If you want to start an argument amongst European patent attorneys, just mention description amendments. As the comments on this and other blogs attest, adaptation of the description is a highly divisive issue and the debate can get decidedly heated (with our US colleagues looking on with bemusement). On one side of the debate are those who believe that aligning the description with the claims is crucial for legal certainty. On the other side, are those with the view that mandatory description amendments are unnecessary, costly, and risky for patentees. The issue is now to be considered by the Enlarged Board of Appeal in G1/25 (IPKat). With the deadline approaching for submitting observations to the referral, it is worth taking a look at the submissions so far.
Description amendments: Background
The requirement for the description to be amended in line with the claims is provided in the EPO Guidelines for Examination, and is followed by examiners and Boards of Appeal with varying degrees of strictness. However, a line of case law has emerged from certain Boards of Appeal, according to which there is no legal basis in the EPC for mandatory adaptation of the description (T 1989/18, IPKat). Indeed, one Board of Appeal found the lack of legal basis for the adaptation of the description so unequivocal as to render a referral to the EBA on the issue unnecessary (T 56/21, IPKat). The divergence between the Boards of Appeal on the issue finally led to a referral last year (G1/25, IPKat).
G1/25: Case background
The referral in G1/25 is derived from T 0697/22, involving a patent owned by Knauf Insulation relating to a hydroponics growing medium. The Board of Appeal referred three questions to the EBA asking for clarification on the necessity (or otherwise) of description adaptation in both opposition and examination proceedings:
"Question 1: 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?
Question 2: If the first question is answered in the affirmative, which requirement(s) of the EPC necessitate(s) such an adaptation?
Question 3: 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?" (emphasis added)
So far, there have been 4 amicus curiae briefs submitted;
- Michael Schmid, Richardt Patentanwälte (6 October 2025)
- Marcel Kortekaas, EP&C Patent Attorneys (9 October 2025)
- International federation of intellectual property attorneys (FICPI) (6 November 2025)
- Olaf S. Roelands, Patentwerk B.V. (13 November 2025)
Analysis of the amicus curiae
 |
| Description amendments |
Article 84 EPC: One direction?
A theme in the arguments against mandatory adaptation (notably by FICPI and Schmid) is the directionality of Article 84 EPC. FICPI argues that Article 84 EPC requires claims to be supported by the description, but there is no reciprocal requirement for the description to be supported or constrained by the claims. FICPI thus argues that as long as the claimed subject matter is present in the description, the presence of additional, unclaimed features (e.g., "A comprising B or C" when only B is claimed) does not logically negate that support.
Schmid's submissions align with this view, arguing that "Article 84 EPC operates in one direction only—from the claims toward the specification. It cannot be applied in reverse to exclude unclaimed subject matter from the description. Consequently, the claims take precedence in determining the extent of protection".
Article 69 EPC as legal basis for description adaptation
The debate over Article 69 EPC in the amicus curiae submissions centres on whether the description's role as an interpretive tool necessitates amendment or its preservation. Schmid argues "description adaptation has limited relevance in national infringement proceedings governed by Article 69 EPC and its Protocol. For example, the doctrine of equivalence, as applied in many EPC member states, considers whether elements not explicitly claimed may nonetheless fall within the scope of protection".
By contrast, Kortekaas views Article 69 EPC as a mandate for consistency between the claims and description, arguing that leaving inconsistencies in the text creates legal uncertainty, risks divergent judicial interpretations across different jurisdictions and would "jeopardize the philosophy of harmonization." Meanwhile, Roelands maintains that Article 69 EPC is primarily a guide for infringement courts and does not provide an explicit legal basis for the EPO to require description amendments during the grant or opposition process.
The relationship between G1/24 and G1/25
The recent G 1/24 decision is used by both sides of the argument. In G 1/24, the EBA ordered that the description should always be consulted for claim interpretation. However, the consequences of this for claim interpretation are not as clear-cut as they might have first appeared. The Boards of Appeal have so far interpreted G1/24 as not permitting patentees to use narrowing information in the description to restrict the meaning of otherwise clear claim language (IPKat). On the other hand, one Board of Appeal found that a broad definition in the description could be used to broaden otherwise clear claim language (IPKat). The approach the UPC will take remains unclear, other than that they agree with the EPO that the description should be "consulted" to interpret the claims (IPKat).
Kortekaas uses G 1/24 as a primary reason for supporting the legal basis for mandatory adaptation of the description. According to Kortekaas, "G 1/24 is a new standard which aims to harmonize the proceedings before the EPO with that of the national courts and UPC. Since the description must be consulted to interpret the claims, it is essential that the description has been subjected to a high quality examination. G 1/25 should not undermine the new standard of G 1/24".
Conversely, Schmid argues that G 1/24 proves adaptation is unnecessary. For Schmid, given that courts are now required to use the description for context, they are perfectly capable of seeing that a narrowed claim excludes broader embodiments mentioned in the text, just as they handle prosecution history estoppel.
Interaction with the UPC
A key question following G1/24 is whether the UPC and EPO are aligned on claim interpretation. The impact of G1/25 on the UPC will depend on where the UPC case law settles on this issue. Kortekaas argues that the EPO must mandate description adaptation to ensure legal certainty and avoid divergent interpretations of the claim in the courts. Kortekaas specifically points to the UPC Court of Appeal decision in NanoString v. 10x Genomics (UPC_CoA_335/2023, IPKat), noting its requirement that the description "must always be used" for claim interpretation as a primary reason for the EPO to ensure internal consistency between the claims and description.
However, Kortekaas also notes in a footnote that '[t]he expressions “are always referred to” in Reason 19 and “shall always be consulted” in the Order of G 1/24 are different from the expression “must always be used” that has been used in Headnote 2 of the order of the UPC Court of Appeal of 26 February 2024 in NanoString v. 10x Genomics (as rectified by the order of 11 March 2024). This difference in wording has led some to question the intended meaning of these expressions. It would help if the EBA reflects on this.' (Emphasis added. On past evidence of the EBA's willingness to provide clarity, PatKat won't be holding her breath).
Conversely, Schmid cites the Hamburg Local Division decision in Agfa v Gucci (UPC_CFI_278/2023) as evidence that mandatory adaptation is unnecessary. Schmid argues that Agfa v Gucci established a clear mechanism for handling discrepancies between the description and claims, whereby specifications in the description that are inconsistent with granted claims cannot be used to broaden the scope of protection (IPKat). Schmid quotes Headnote 3 of Agfa v Gucci, according to which: “specifications in the description that are not consistent with the granted claims cannot serve as a basis for a broad interpretation of a claim”.
Risks of adding matter
FICPI argues that forced adaptation can fundamentally alter the legal and technical interpretation of the patent, potentially leading to Article 123 EPC added matter issues. FICPI argues that if the presence of unclaimed subject matter affects claim interpretation, deleting it would change that interpretation, potentially invalidating the patent under Art. 123(2) or (3) EPC in view of Article 69 EPC and G 1/24. (IPKat).
Sufficiency of Disclosure (Article 83 EPC): The risk of under-enablement
FICPI and Roelands also express significant concerns that deleting "non-relevant" embodiments can leave the remaining claimed invention technically hollow, failing the requirement for a sufficient disclosure under Article 83 EPC. FICPI points out that an unclaimed embodiment often contains specific technical details or examples that provide the necessary support for a feature that remains claimed elsewhere. If these details are excised because they are no longer literally encompassed by narrowed claims, FICPI warns that the remaining description might no longer provide enough information for a person skilled in the art to carry out the invention. FICPI concludes that identifying and removing every passage that could never contribute to the interpretation of an amended claim is not only an undue burden but is practically impossible, but doing so risks creating a disclosure issue that could inadvertently invalidate an otherwise valid patent.
The relevance of "claim-like clauses"
The FICPI submission strongly contests the mandatory removal of these clauses, arguing that such a requirement is technically and legally unjustified (IPKat). According to FICPI, features or embodiments presented in the description cannot be confused with the actual claims provided there is no explicit reference to them as "claims". They warn that forced deletion can be detrimental, as these clauses often serve as a vital source of support for Article 123(2) EPC compliance, particularly in divisional applications where the initial set of claims is often appended to the description for precisely this purpose. FICPI submits that this "cleansing" process may unnecessarily harm applicants by removing the very basis they need for future amendments.
Roelands presents a more nuanced middle-ground on claim-like clauses. Roelands argues that regardless of the presence of clauses or not, the description must contain a verbatim definition of the invention or a reference to the claims to satisfy Article 84 EPC. As such, for Roelands, the adaptation of clauses should only be necessary if they render the verbal support of the amended claims ambiguous, such as when a clause presents a full alternative definition of the invention that contradicts the new claims. Otherwise, Roelands argues that if the clauses do not create such ambiguity, no additional adaptation or removal should be required.
Application-as-filed or granted?
In a recent thought-provoking addition to the debate, a recent article by KatFriend Greg Corcoran highlighted an interesting point of claim interpretation in two recent UPC Court of Appeal decisions. Greg points out that in Meril v Edwards and Amgen v Sanofi the UPC appeared to favour the application as filed over the granted specification as the basis for claim interpretation. Greg notes the UPC guidance in Nanostring to construe the claims and identify the "objective problem" in the application as filed. Greg argues that the UPC thus effectively bypasses description amendments made during grant. By contrast, according to case law of the EPO Boards of Appeal, the description of the patent application is irrelevant for claim interpretation and the claims should be interpreted in view of the granted patent (T 450/20, r. 2.16, IPKat). Since G1/24 found it "a most unattractive proposition that the EPO deliberately adopt a contrary practice to that of the tribunals that are downstream of its patents" (r.16), Greg asks if the Boards of Appeal will recognise and account for the UPC position.
Final thoughts
From this Kat's point of view, mandatory adaptation of the description seems entirely unnecessary. As FICPI points out, no other major patent office requires it. Why then has the EPO chosen to dig its heels in on the issue? The official line appears to be that description adaptation is part of the EPO's "focus on quality". However, quite apart from the legal risks pointed out in the submissions, adaptation of the description also unnecessarily increases the costs for the applicant. Indeed, this Kat heard directly from a partner at a top-tier firm (which shall remain nameless) that the billable hours were one reason he remained in favour of the requirement. Although this Kat feels that AI may very quickly remove this perverse incentive, if it has not already done so.
Regardless of where you stand on the issue, G1/25 will be closely watched.
Further reading
Reviewed by Dr Rose Hughes
on
Monday, January 12, 2026
Rating:
Follow the IPKat on LinkedIn 
Posts
One thing I hope the EBA addresses is inconsistencies between examiners in the type of amendments required to "bring the description in line with the claims". My experience is that some examiners are happy with simply editing the summary to correspond to the claims intended for grant (or replacing it with a note that the invention is set out by the claims) and don't require me to touch the detailed description. Other examiners insist on extensive amendments disclaiming or deleting large parts of the detailed description. If the EBA decide to uphold the amendment requirement, I hope they at least settle on a legal basis and some general guidelines for what exactly is an inconsistency requiring an amendment and what is required to align the description with the claims.
ReplyDeleteI can see from a very strict legal perspective how they could find Q3 inadmissible, but from a practical perspective it would simply be a delay for practitioners and users of the patent system for a few years until the next referral happens in the hope of clarifying what they mean.
It will not come as a surprise that I am of the opinion, that an adaptation of the description to the claims, before grant and before maintenance in amended form, is a necessity. I see three reasons for this.
ReplyDeleteSuch an adaptation is necessary, firstly for the safety of third parties. Any doubt about the scope of protection, by leaving an inconsistency between claims and description has to be removed.
Secondly, such an adaptation is necessary, as it guarantees that the interpretation of claims is the same in procedures before the EPO and before national jurisdictions or the UPC, once the EPO has lost any competence to decide on the fate of a patent.
This does not withhold any post-grant acting judicial body to extend the protection conferred by the claims to equivalents, cf. Art 2 on the Protocol on Interpretation of Art 69 EPC.
Thirdly, such an adaption is the direct result of G 1/24, as interpreted by the boards of appeal in decisions following the issuance of the latter. Any inconsistency between claims and description should be removed by way of amendment in order to avoid the necessity of interpretative somersaults, cf. T 2027/23. G 1/24 can be held to have paved the way for G 1/25.
Any feature mentioned in an independent claim cannot be held purely optional in the description. Any originally disclosed embodiment not falling under the independent claim, does not need to be deleted, but acknowledged as not corresponding to the claimed invention. Those types of inconsistencies have, without any doubt, to be removed.
It is not because post-grant jurisdictions are able to interpret an independent claim and exclude embodiments not covered by it, that it dispenses the EPO not to ensure alignment between the claims and the description.
In the famous Agfa-Gucci case before the CFI LD Hamburg, UPC_CFI_278/2023, the mention of “chromatic” in the description should have been deleted before grant, as the claim was limited to “achromatic”. As the description was not adapted, Agfa was prompt in devising an Angora cat including “chromatic” in its claim for infringement.
A few observations.
ReplyDeleteTo my knowledge, the only provision of the EPC which concerns "doubt about the scope of protection" is the clarity provision of Article 84 EPC. That is a provision which concerns the claims, and not the description.
I am also blissfully unaware of any provision of the EPC, or indeed any hint of such a provision in the TP to that Convention, which affords the EPO both the role and the responsibility for guaranteeing that "the interpretation of claims is the same in procedures before the EPO and before national jurisdictions or the UPC". That sounds a lot like judicial imperialism, and makes no sense if - as is certain to be the case - the EPO cannot be relied upon to get claim interpretation 100% correct in 100% of cases.
Finally, any interpretation of the claims should follow the principles of Article 69 EPC. Adaptation of the description is simply not relevant to interpretation of the claims. If, due to disclosures of the description, the subject matter of the claims is unclear, then this points to a lack of clarity of the claims ... for which the remedy is amendment of the claims. On the other hand, if the subject matter of the claims is clear, then there is no doubt as to the subject matter for which protection is sought, and hence no need to amend either the description or the claims.
@ Proof of the pudding,
DeleteArticle 84 EPC reads:
The claims shall define the matter for which protection is sought. They shall be clear and concise and be supported by the description.
There is thus a direct link between the claims and the description.
In T 2766/17, Catchword, it was held that the description cannot contain statements contradicting the plain claim wording as it may cast doubts as to the intended meaning of this wording.
In T 758/13, Reasons 2.1, it was held that purely formal support by the description through repetition of the mention of a claimed feature does not meet the requirements of Art 84 EPC.
Even if you appear to be “blissfully unaware of any provision of the EPC, or indeed any hint of such a provision in the TP to that Convention, which affords the EPO both the role and the responsibility for guaranteeing that "the interpretation of claims is the same in procedures before the EPO and before national jurisdictions or the UPC"”, this is an aim which is pursued by the boards of appeal.
See for instance. T 2027/23, Reasons 3.5.4 (c), which states that “the second takeaway from decision G 1/24, namely a harmonised claim interpretation between national courts or the UPC and the Boards of Appeal”.
Indeed, in G 1/24, Reasons 16, one reads: “the Enlarged Board agrees with the harmonisation philosophy behind the EPC (see G 5/83, Reasons 6, and G 3/08, Reasons 7.2.2)”
There is thus nothing like judicial imperialism at stake, and if the EPO might not be 100% right on the interpretation of the claim, the same can be said of any court deciding in post-grant procedures.
What is to be avoided, cf. T 2027/23, op cit, is that “patentees could get away with different claim interpretations for validity, on the one hand, and for infringement on the other”, cf. the famous Angora cat. This is what Agfa tried to do.
In G 1/24, Reasons 7, the EBA held that “Article 69 EPC and the Protocol are arguably only concerned with infringement actions before national courts and the UPC”. As far as the EPO is concerned, it is thus incorrect to claim that “any interpretation of the claims should follow the principles of Article 69 EPC”.
However, Art 69 EPC and its Protocol of interpretation, do not hinder a national court or the UPC to consider other forms of infringement, e.g. by equivalents.
The subject matter of the claims might be clear as such, but there can be doubts as to the subject matter for which protection is sought, when the description contains statements allowing either a limited or a broadened interpretation of the claimed features. This is what the boards applying G 1/24 consider a discrepancy or inconsistency between claims and description.
In all decisions of the boards, following G 1/24, it has been made clear that if the applicant/proprietor wants to see the interpretation of claim found in the description to be taken into account, the claim has to be amended.
When the claim has been limited, any statement allowing a broader interpretation of a claim has to be removed from the description. The description has to be amended in order to remove the discrepancy/inconsistency between the original description and the claims.
It is thus an absolute necessity to amend either the description or the claims.
G 1/24 has not brought for lots applicants/proprietors what they had hoped, i.e. that the interpretation found in the description of claimed features has to take precedence over the interpretation the skilled person gains from the plain wording of the claim.
If G 1/24 is correctly applied, the times when the patent was its own dictionary have gone. It is to be hoped that G 1/25 will simply continue and complete the move started with G 1/24.
Article 84 EPC does indeed specify that "The claims ... shall ... be supported by the description". But what on earth is the connection between that requirement and "doubt about the scope of protection"?
DeleteAre you saying that doubts over claim scope arise every time that the subject matter defined by a claim is unsupported in some way? Even if the claims meet the clarity requirement of Article 84 EPC??
I fully acknowledge that you, and many of your ex-colleagues at the EPO have concerns about patent descriptions which contain unclaimed subject matter. But just having a "bad feeling" about that does not equate to grounds for invalidity under the EPC.
Determining whether there is a ground of invalidity is a step-wise process that involves interpreting the wording of a provision of the EPC and (only) then determining whether that interpretation gives rise to an objection when considering the facts of a specific case.
One simply does not work backwards by assuming that certain facts give rise to an objection and then, with that assumption in mind, working out which interpretation of a provision of the EPC might justify such an objection. However, it is plain as a pikestaff that this is precisely the kind of “cart-before-horse” approach that Boards of Appeal have adopted when it comes to objections to “unadapted” descriptions. This point is particularly evident from the Smorgasbord of different legal provisions that the Boards have attempted to use to justify addressing various “inconsistencies”. If the legal basis for adapting the description were so clear-cut, then there simply would not be such a divergence of opinion on precisely where that legal basis can be found.
At the end of the day, if the claims of a patent are clear, how can there be any "doubt about the scope of protection"?
If there is such doubt, why does it only raise a support objection under Article 84 EPC, as opposed to a clarity objection?
If the EPO can work out which disclosures support the claims and which do not, why would they worry about the national courts (or the UPC) not being able to do the same?
And should the EPO be in the business of determining what does and does not fall within "the scope of protection"? Does that not raise issues concerning Article 69 EPC and its Protocol that the EPO simply should not be getting involved with?
@ Proof of pudding,
DeleteIf you had read carefully what I said, you would have realised that a claim can be clear as such, but statements in the description can cast doubts about the exact scope of protection. I maintain that such a claim is not supported by the description.
For example, if the claim is, due to prior art, limited to an “achromatic” base colour, but the description still includes a “chromatic” base colour, then the claim is not supported by the description, as the description allows a broader scope of protection than that which can be derived directly and unambiguously from the claim wording.
This is exactly what all decisions following G 1/24 have explained. If an applicant/proprietor wants the scope of protection to be that derivable from the description, then the claim has to be amended accordingly.
I have never said, or never heard any of my colleagues saying that they have a "bad feeling" about leaving statements in the description which are not in line with what is actually claimed. It is by no means a matter of feeling, it is a matter of fact: the claims are not supported by the description and that’s it.
You seem to carefully ignore the fact that Art 84, clarity and support, is not a ground for invalidity. It is thus the duty of the EPO to grant or maintain patents which do not allow the proprietor to change its interpretation of the claimed subject matter as he thinks fit.
It might have been better if the boards had been more precise, but a general reference on Art 84 should be sufficient to people with a mind willing to understand.
National courts (or the UPC) are certainly able to work out which disclosures support the claims or not, but this does not dispense the EPO to abide by Art 84 with its two prongs, clarity as such and support.
The EPO should certainly not get involved with Art 69 and the Protocol, but it has to insure that the proprietor cannot change its interpretation as it suits him: narrow when it comes to validity but broad when it comes to infringement.
When in the past, granted European patents were only dealt with by national courts, the problem was not so manifest, and each court could decide on the interpretation it wanted.
With the upcoming of the UPC and its wider geographic range, it is necessary to put a hold to this way of behaving by proprietors and request alignment between the claims and the description, i.e. to apply Art 84 in full.
It cannot be given an absolute guarantee that a national court or the UPC cannot end up with a different interpretation of claimed subject matter than that of the EPO, but by aligning claims and description, the likelihood of such a discrepancy in interpretation is minimised.
The duty of the EPO is not to please applicants/proprietors, but to grant or maintain patents which are not open to multiple possible interpretations.
You have stated that “If an applicant/proprietor wants the scope of protection to be that derivable from the description, then the claim has to be amended accordingly”.
DeleteI could perhaps agree with that statement. The problem that I have is that your other comments seem to assume that a claim can meet the clarity requirements of Article 84 EPC even where “the description allows a broader scope of protection than that which can be derived directly and unambiguously from the claim wording”. I simply do not see how that could possibly be the case.
This all seems to link back to issues of claim interpretation, namely “EPO-style” primacy of the claims vs. interpretation in accordance with Article 69 EPC and Articles 1 and 2 of the Protocol. Indeed, your comments highlight the fundamental disconnect between the EPO’s claim interpretation and description adaptation practices, where:
(a) when interpreting the claims, and by applying “EPO-style” primacy of the claims (or, as you like to say, rejecting the concept of a patent being its own dictionary), the EPO studiously ignores potentially contradictory disclosures of the description; but
(b) when raising objections to the description, the EPO suddenly starts worrying about what might happen if, when interpreting the claims, a national court does not apply “EPO-style” primacy of the claims (and therefore accepts the concept of a patent being its own dictionary).
This is a problem that is entirely of the EPO’s own making. It is also a problem to which there is an obvious solution: the EPO should interpret claims in a manner that is compatible with (or at least not so clearly incompatible with) claim interpretations as conducted by national courts and the UPC.
As to the "bad feeling" that I mentioned, please interpret that as referring to your view (and the view of certain of your ex-colleagues) that the EPC does not “allow the proprietor to change its interpretation of the claimed subject matter as he thinks fit”. The fundamental question here is: where does it say that in the EPC?
The proprietor does not get to change the interpretation of the claims. That is fixed by the wording of the claims in combination with information / evidence which aids its interpretation, whether that be intrinsic evidence (ie disclosures of the patent application as originally filed) or extrinsic evidence (such as common general knowledge or a declaration of a person of ordinary skill in the art). Thus, the only avenues open to the proprietor are the provision of additional (extrinsic) evidence and/or arguments relating to how the evidence should be used / understood.
In an ideal world, a patent proprietor’s pre- and post-grant arguments would be entirely consistent with one another. However, we do not live in an ideal world, and nothing (not even “strict” adaptations of the description) can prevent different arguments being used at different times. There can even be situations where such differences are entirely justified, eg due to the emergence of new evidence, or a new understanding of the original evidence.
Let us be honest, what the EPO’s description adaptation practice really seeks to achieve is a poor man’s file wrapper estoppel. Whist establishing some form of file wrapper estoppel might be an interesting objective, it certainly has no basis in the EPC. And, in any event, “strict” adaptations of the description are an exceptionally bad (and likely ineffective) way to set out achieving that objective.
@ Proof of the pudding,
DeleteThe sentence “If an applicant/proprietor wants the scope of protection to be that derivable from the description, then the claim has to be amended accordingly” is not my creation. It is the conclusion one has necessarily to draw from the decisions following G 1/24, e.g. T 2027/23.
I would allow myself to add that you manifestly disagree with the conclusions drawn by the boards after G 1/24. It however happens that the last say, when it comes to validity in opposition, is with the boards and not with the applicants/proprietors and their representatives.
You do apparently not see how it could possibly be the case that “the description allows a broader scope of protection than that which can be derived directly and unambiguously from the claim wording”. I will therefore repeat an old example of mine: if the claim is about a spade, and the description explains that under a spade one can understand any kind of garden implement, it is manifest that on the one hand, the claim is limited to a spade, but with the description the applicant/proprietor wants also to seek protection for any kind of garden implement. One statement is not compatible with the other one.
When it comes to validity, the applicant/proprietor will insist that its claim is limited to a spade, but when it comes to infringement, the applicant/proprietor will insist that the scope of protection also encompasses any kind of garden implement. This is not what a patent is for.
This behaviour is exactly what Agfa did when it attacked Gucci before the CFI-LD Hamburg. As the description contained a statement covering “chromatic”, Agfa ignored the limiting wording of the claim to “achromatic” and sued for infringement on the basis of the description mentioning “chromatic”. If this does not boil down to changing the interpretation, then I wonder what can be understood under changing the interpretation. This is exactly what the board queried in T 2027/23, referring to the well known Angora cat.
The EPO’s description adaptation practice really does not seek to achieve a poor man’s file wrapper estoppel. There is no file wrapper estoppel in the European patent system. The European patent system simply wants to avoid inconsistencies between claims and description.
By the way, the CFI-LD Hamburg noted that there were several passages mentioning “chromatic” in the specification of the patent which “should have been deleted”. This is exactly the purpose of G 1/25. Following the clarification by G 1/24, the work has to be finished with G 1/25.
Doing what is suggested by the UPC-CFI LD Hamburg cannot be seen as requiring a “strict” adaptation[s] of the description which would likely be ineffective. On the contrary, it is highly effective against applicants/proprietors trying to foster Angora cats.
If during prosecution a claim is limited, any statement in the description contrary to what the skilled person would derive directly and unambiguously from the claim wording should be either deleted or marked as such.
It is not just in an ideal world, that a patent proprietor’s pre- and post-grant arguments should be entirely consistent with one another. In the not-ideal world we live in, the EPO is there to prevent different interpretations of one and the same claim being used at different times. This is an important aim which deserve support, irrespective of whether national courts and the UPC can draw their own conclusion from an ill adapted description.
With reference to your example, you have somehow overlooked the obvious (alternative) solution: the claim should be interpreted by reference to the description ... meaning that "spade" is afforded the definition provided in the description.
DeleteIs that not how the national courts and the UPC would interpret the claim? So why would the EPO deliberately adopt a divergent (and completely incompatible) approach to claim interpretation? Does that divergence of approach not generate Angora cats? And, if so, why on earth would the EPO deliberately decide to do that?
@ Proof of the pudding,
DeleteYou seem to have overlooked, willingly or not, how the boards are applying G 1/24. From all decisions following G 1/24 a clear line of case law has emerged: if the applicant/proprietor wants the interpretation given in the description to be taken into account, be it a broadening or a limiting one, the claim has to be amended accordingly.
In other words, there is no alternative solution whatsoever, although this is what you manifestly would like to be taken into account. Forget what is in the claim, but look at the description. At this rate, claims are superfluous.
If the description allows a broadened interpretation, the prior art against the claim will be broader than that warranted by the mere, limiting, claim wording, and will be commensurate with the broad interpretation allowed by the description.
If the description allows a more limiting interpretation, the prior art against the claim will be commensurate with the broad wording of the claim.
In presence of an inconsistency between claims and description, created by the different possible interpretations, national courts have no choice but to look at the description to determine the actual scope of protection.
The EPO does by no means deliberately adopt a divergent and completely incompatible approach to claim interpretation.
With G 1/24 primacy of the claim has been re-established, and it is good so. I am at a loss to see how that divergence of approach can at all generate Angora cats. If the broad or limited interpretation in the description has to be taken into account into account it has to be brought into the claim.
This will avoid the proprietor to allege that with a limited claim, it can escape broad prior art, but when it comes to infringement, in view of the broadening in the description, he can sue for infringement. We have here a first form of Angora cat.
This is exactly what Agfa did in the famous Agfa-Gucci case before the CFI-LD Hamburg. Agfa sued for infringement on the basis of “chromatic” and “achromatic”, although the claim was clearly limited to “achromatic”. In its decision, the CFI LD Hamburg, UPC_CFI_278/2023, noted that the mention of “chromatic” in the description should have been deleted before grant, as the claim was limited to “achromatic”. This exactly what has to be done in matters of adaptation of the description: delete any inconsistency between claim and description, allowing different interpretations.
Conversely, this will avoid a proprietor to claim infringement on the broad wording of the claim, but at the same time fence off prior art in view of the limited interpretation in the description. We have here a second form of Angora cat. I allow myself to invite you to look a T 2027/23 which was abundantly clear on the topic.
Actually, with G 1/24 the EPO has deliberately decided to go back to basics. This approach ought to be completed with G 1/25. No hiding in the description statements which are at odds with the interpretation which derives directly and unambiguously from the plain wording of the claim.
I do not overlook the interpretation of G 1/24 by certain Boards of Appeal. I simply doubt whether "EPO-style" primacy of the claims is truly consistent with that decision. In essence, if that is true, then the word "consult" in the EBA's ruling has no substance or practical effect. That hardly seems plausible.
DeleteOn the other hand, and given your insistence that the EPO is absolutely correct to apply "EPO-style" primacy of the claims (ie essentially interpret the claims in isolation from the description), I find it difficult to find any basis in fact for your assertion that "The EPO does by no means deliberately adopt a divergent and completely incompatible approach to claim interpretation".
Let's be honest, the entire motivation for the EPO's description adaptation practice is the fact that, compared to the EPO, the national courts and the UPC adopt a different approach to claim interpretation. Reasons 19 of G 1/24 was quite clear in stating that this was undesirable. You will therefore forgive me for struggling to find any persuasive logic in a position that supports EPO practice which sustains this divergence of claim interpretation approaches, whilst simultaneously insisting that the EPO does not deliberately adopt such a divergent approach.
@ Proof of the pudding,
DeleteWhat you call the “EPO-style" primacy of the claims, is, contrary to what you think, truly consistent with G 1/24. The word "consult" in the EBA's ruling has a lot of substance and practical effect. How the bords apply G 1/24, is more than just plausible.
It also happens that you barely ever reply to practical examples submitted to you. You prefer to stay much more with general statements, as most probably you do not have a plausible answer.
I will try once more to explain the situation. From a clear claim the skilled person cand directly and unambiguously derive interpretation A. When consulting the description, a board finds either interpretation BA.
What all the boards, applying G 1/24, have decided, is that, when the applicant/proprietor wishes interpretation B or C to be taken into account, the claim has to be amended accordingly. Without “consulting” the description this conclusion could not be reached.
The applicant/proprietor cannot fence off broad prior art relevant to the true interpretation A of the claim by alleging, that interpretation BA.
Claim interpretation A is what is valid, and it cannot be replaced by interpretation B or C as the applicant/proprietor thinks fit. “Consulting” the description, does not mean that interpretation B or C has to prevail over interpretation A which derives directly and unambiguously from what the skilled person gets from the claim wording.
National courts and the UPC may adopt a different approach to claim interpretation according to your wishes, but the EPO is not bound by the rules of claim interpretation devised by another judicial instance.
If the description is adapted to the claim, i.e. when interpretation A in the claim is replaced by interpretation B or C, then there will be no doubt about what the claim actually protects as far as validity is concerned. It will the be B or C, but not A.
It is only because the EPO has tolerated for too long that the description was not adapted to the claims, that national courts and the UPC had to carry out interpretative summersaults in order to find out exactly what was covered by the claims.
I fully support Reasons 19 of G 1/24, but with G 1/24 and possibly G 1/25, those decisions will ensure full harmonisation of the interpretation as far as validity is concerned. This will not withhold or hinder national courts and the UPC to decide, on the basis of Art 69 and the Protocol, on direct/indirect infringement, or infringement by equivalents.
I allow myself here to refer to Reasons 7 of G 1/24: “Article 69 EPC and the Protocol are arguably only concerned with infringement actions before national courts and the UPC. Such a conclusion can be drawn from the wording of Article 69 EPC and the Protocol, the drafting history of these provisions, and from their position in the EPC”.
In Reasons 9 of G 1/24, the EBA went as far as to say: “Given the above, the strictly formal answer to Question 1 would be "No".” The “above” means that neither Art 69, Reasons 7, nor Art 84, Reasons 8, can be used as clear legal basis, in terms of an article of the EPC, for claim interpretation when assessing patentability.
If patentability is be assessed as suggested by the EBA and applied by the boards, there will not be different interpretations between the EPO and national courts or the UPC, as far as validity is concerned.
The example that you provide points to a lack of clarity of the claims. That should be dealt with by amendment of the claims.
DeleteYou must have missed the part of my previous comment where I made it clear that I was discussing the scenario where the claims have already been found to meet the requirements of Article 84 EPC. You also appear to believe that the national courts and the UPC should follow "EPO-style" primacy of the claims ... when it is abundantly clear that they do not, and where it is also clear that this very fact is the entire motivation for the EPO's description adaptation practice.
I refer you to Roche's excellent amicus brief, which addresses many of the points that you have raised.
@Mr Thomas -
DeleteYou refer to the Agfa/Gucci case as supporting your position. But you seem to have overlooked that -
- the court did not explain WHY the allegedly conflicting disclosures in the description "should have been deleted" with reference to any applicable law. It seems likely that the subtext here is simply that this is their understanding of current EPO practice - they were not being asked to rule on whether that practice is correct or not; and
- despite the presence of these conflicting disclosures, the court was able to come to a claim interpretation that got rid of the Angora cat anyway: see headnote 3 ("Specifications in the description that are not consistent with the granted claims cannot serve as a basis of a broad interpretation of a claim") and the paragraph spanning pages 23-24 of the decision (" Chromatic and achromatic colours are mutually exclusive, and a colour only similar to achromatic is not achromatic. However, regardless of para [0029] being outside the scope of the patent, even when considered otherwise para. [0029] could not (and does not) override the clear definition given in para. [0021] due to the
necessity of legal certainty")
Thus the court treated this analogously to a "disclosed but not claimed is disclaimed" scenario. An elegant way of putting Angora cats back in their box, don't you agree?
The practice which you advocate, requiring strict conformity of claims and description, seems to be a solution in search of a problem.
@ Proof of the pudding,
DeleteThe example I have provided does not point to a lack of clarity. Where would you see a lack of clarity? For the skilled person, the wording of the claim allows directly and unambiguously interpretation A. When the description allows interpretation B=less than A or C=more than A, then it follows that there is an inconsistency between claims and description, but not a lack of clarity.
Most of the decisions following and referring to G 1/24, do not mention Art 84, but Art 54 and 56, as it is primarily a question of patentability and which prior art is applicable. I refer here e.g. to T 2027/723, T 0356/23 (limited interpretation possible according to the description) and T 1849/23 (broader interpretation possible according to the description).
Some decisions following G 1/24 deal with clarity, e.g. T 0981/23, T 0389/24, or added matter, e.g. T 0873/23.
In T 1588/23, the board held that unclear features cannot be ignored under G 1/24.
In T 0405/24, the board noted that, even if the order of G 1/24 could indeed be extrapolated to the assessment of compliance with Art 123(2), there is no indication in G 1/24 that “consulting” or “referring to” the description and drawings could translate to adopting a claim interpretation which ensures that the disputed feature is originally disclosed and thus necessarily complies with Art 123(2).
Such an approach which inherently assumes that there may be only one “correct” interpretation of a claim feature, namely the one derivable from the original description as its intended meaning.
Furthermore, I never said that national courts and UPC should follow the "EPO-style" primacy of the claims. If you had read carefully what I have written, I merely mentioned that as far as validity is concerned, when claims and description are on a par, there is no risk of diverging interpretation between national courts and the UPC vs. the EPO.
This leaves national courts and the UPC to decide on direct/indirect infringement and on infringement by equivalents, once the claim is held valid. What is to be avoided, is that the proprietor comes up with a different interpretation when it comes to infringement and/or validity.
You referred to “Roche's excellent amicus brief”, but I fail to see any amicus curiae from Roche. I do however see here a manifest problem of clarity on your side.
I allow myself to refer you once more to the now famous Agfa vs. Gucci, case, cf. UPC_CFI_278/2023. In this decision, the CFI LD Hamburg, noted that the mention of “chromatic” in the description should have been deleted before grant, as the claim was limited to “achromatic”.
Contrary to what you and a lot of your colleagues are thinking, this UPC decision cannot be taken as evidence that mandatory adaptation is unnecessary. It is however true that the UPC was not fooled by maintaining “chromatic” in the description.
The CFI LD Hamburg correctly concluded that “specifications in the description that are not consistent with the granted claims cannot serve as a basis for a broad interpretation of a claim”, but in the same decision, it queried the lack of adaptation of the description with the claims as granted.
The Roche brief is on the Register at: https://register.epo.org/application?documentId=ML12AT8R10T2DWK&number=EP07704142&lng=en&npl=false
DeleteWith reference to Agfa, I refer you to my comment immediately above. Is there not an absolutely massive, glaring, obvious contradiction in your position that, on the one hand, the adaptation on the description to remove "chromatic" is absolutely necessary to avoid different interpretations for infringement and validity; and on the other hand, that the court "was not fooled by maintaining "chromatic" in the description"?
The referring board in T 697/22 noticed as much at point 16.8 of its reasons.
You argue that:
Delete"For the skilled person, the wording of the claim allows directly and unambiguously interpretation A. When the description allows interpretation B=less than A or C=more than A, then it follows that there is an inconsistency between claims and description, but not a lack of clarity".
In a nutshell, this highlights the differences between our claim interpretation philosophies. For me, "consulting" the description means taking into account clear, explicit definitions provided in the description - even if those definitions do not align with "standard" definitions. For you, such definitions are to be disregarded for interpreting claims that, on their own, impart a clear meaning to those skilled in the art.
In other words, you adopt what I call an "EPO-style" primacy of the claims approach. On the other hand, my philosophy is far more aligned with the approach of the national courts and the UPC, which is to interpret the claims in the light of the whole patent specification.
You will therefore understand why I conclude that the claims lack clarity in your hypothetical example. It is simply the result of adopting a more "court-like" approach to the interpretation of the claims. That is, if the definitions in the description are taken into account, the interpretation of the claims becomes unclear - with the result that amendment of the claims is necessary, eg by incorporation of the relevant definition(s) from the description.
Strictly speaking, there is a divergence of approach even when the description and claims are fully aligned. That is, the two different approaches are still as outlined above ... it is just that there may be circumstances under which those different approaches can arrive at the same conclusion.
If you cannot find Roche's amicus, you will need to look at the online register entry for EP2124521. It is amongst the briefs filed on 29 January.
You must not be very busy as you have time reply by return of post.
DeleteWhether willingly or not, you are mixing two issues: alignment of the independent claim and the description and the position of the CFI-UPC with respect to the Angora cat. Try to read again what I have said. It is glaring, but apparently not for you.
I repeat once more, what the EPO is doing, is to avoid proprietors to change their view as they think fit. Small interpretation when it comes to the claim and broad interpretation based on the description when it comes to infringement.
It is therefore irrelevant what a national court or the UPC does in order to squeeze out any Angora cat. If the description would have been aligned with the claim limited to “achromatic” base colour, Agfa would never have been tempted to claim infringement on the basis of a “chromatic” base colour.
I invite you to have a good look at T 2027/23. It speaks about Angora cats and confirms what I have been saying all along.
@ Anon Y. Mouse,
DeleteI maintain that the UPC-CFI LD Hamburg has queried the fact that the description had not been cleared of the word chromatic before grant. This is a matter of fact and claiming that the court did not explain why is playing with words. Nothing in a decision is innocent, and I look at this comment as a kind of obiter dictum.
I further maintain that the UPC-CFI LD Hamburg was not fooled by the fact that the description had not been adapted to the claims. The UPC-CFI LD Hamburg could have stayed with the statement that "Specifications in the description that are not consistent with the granted claims cannot serve as a basis of a broad interpretation of a claim", but it added the comment on the lack of adaptation of the description. This was not innocent.
In any case , this is not a reason good enough for not admitting that there was an inconsistency/discrepancy between the independent claim as granted and the description.
How to deal with such inconsistencies/discrepancies is exactly the question the referring board is asking the EBA for a solution.
It is not be cause a national court has found a way of putting Angora cats back in their box, that the EPO should continue to grant patents which allow nurturing Angora cats. Its duty is to provide patents which are not full of traps, i.e. in which the claims says A and the description says actually B less than A or C more than A, so that the proprietor can change interpretation as he thinks fit.
Requiring strict conformity of claims and description, is the solution of the problem of inconsistencies/discrepancies between independent claims and description. You manifestly do not want to see a problem where there is a huge one.
@ Proof of the pudding, 02.02.2026 at 16:46:00
DeleteYou still not seem to have understood the message the boards convey. For example, in T 2027/23, the claim is absolutely clear as such, but the board refused any interpretative summersault when comparing the independent claim with statements in the description.
Your way of "consulting" the description which means taking into account clear, explicit definitions provided in the description - even if those definitions do not align with "standard" definitions” found in the claims, is not the way the boards see it. At the end of the day, the boards are deciding when validity is at stake, and it is their good right to decide as they do.
If the applicant/proprietor wants the definition in the description to be taken into account, the claim has to be amended accordingly, i.e. by bringing in the claim B=less than A or C=more than A instead of merely A.
Even if for the sake of arguments, it could be held that the claim with A is unclear over the statements in the description, B=less than A or C=more than A, then this unclarity has to be resolved before grant, if the claim cannot be reduced to A in view of prior art against A or C=more than A.
Your philosophy might look far more aligned with the approach of the national courts and the UPC, which is to interpret the claims in the light of the whole patent specification, but at this rate, as I said it before, we can get read of claims.
Third parties should not have to delve deep in the description to have an idea what the actual scope of protection could well be. This does in no way hinder national courts or the UPC to decide on direct/indirect infringement or infringement by equivalents.
When a national court or the UPC has to delve in the description in order to determine the actual subject-matter of the claim or the actual scope of protection, it is because the granting authority did not do its job properly before grant.
I maintain that in case of an inconsistency/discrepancy between claims and description, the description has to be adapted to any independent claim.
To be clear, your whole comments as that of Anon. Y Mouse actually comfort my views.
You say it yourself: “if the definitions in the description are taken into account, the interpretation of the claims becomes unclear - with the result that amendment of the claims is necessary, e.g. by incorporation of the relevant definition(s) from the description”. This is the view adopted by the boards following G 1/24. And this job need to be finished with G 1/25.
I fail to see any circumstance under which the different approaches can arrive at the same conclusion. Could you please provide an example? I am for practical examples not theoretical views.
OK, now seems like an opportune moment to bring an end to the back-and-forth between us on this thread.
DeleteYou asked me to address a hypothetical scenario. I answered and explained the reasoning behind my answer. You disagree with my reasoning. Fair enough. I disagree with your reasoning, especially that which appears to presume a role for the EPO in policing the arguments that patentees might attempt to make before national courts or the UPC.
Perhaps reasonable minds can differ on this point. So let's just agree to differ, rather than generate another thread with the kind of "increasingly intemperate language" noted in the amicus filed under the pseudonym Jackson Lamb.
@ Proof of the pudding,
DeleteLet’s agree to stop the discussion. We will eventually see which opinion will prevail. As far as G 1/24 is concerned, I must come to the conclusion that it was certainly not yours.
Thanks for telling us that Jackson Lamb is a pseudo. Boards of appeal never consider anonymous third parties observations. May be the EBA should do the same as the boards, when it comes to amicus curiae filed under a pseudo, an other form of being anonymous.
I noted Jackson Lamb’s comment about the "increasingly intemperate language" and its reference to T 438/22 commented in my blog. I fail to see any “intemperate language” in the comments of this decision in my blog. The same applies to the comments in my blog of T 2027/23.
I invite you to read Points 2 and 3 of the catchword in T 438/22. Points 2 and 3 of the catchword are certainly not going in the direction you would like it to go. Adding that claim-like clauses should be deleted is a point view as valid as yours.
I repeat that for the sake of legal certainty, patents when granted or maintained should not allow the proprietor to change its point of view as he thinks fit, and it should not be necessary to delve deeply in the description to know what actually falls under the scope of protection of the claims.
As the EPO is not competent to decide infringement and validity (after the opposition period) it is important to amend the description “if the latter contains subject-matter which manifestly impedes a common understanding, it is legitimate to insist on its removal under Articles 84 and 94(3) EPC and Rules 42, 48 and 71(1) EPC”.
When a national court decides on validity, it has the power to iron out inconsistencies between claims and description left by the national granting authorities. In the EPC system, the BA do not have the possibility to iron out inconsistencies between claims and description left by first instance divisions of the EPO in the absence of an appeal.
Two anonymous amicus briefs were submitted in G 4/19 and acknowledged by the EBA (see point X of the summary of facts and submissions). A number of anonymous amici also submitted briefs in G 1/21. There is no indication that these were disregarded in either case, albeit that the EBA rarely comments explicitly on *any* amicus brief specifically in any event.
DeleteI think that your understanding of "let's agree to differ" must be different to mine! Anyway, thanks for pointing out further reasons why you believe that my opinion is incorrect (and why, in your opinion, Jackson Lamb's comment is either misplaced or directed at comments other than your own).
DeleteAs you have raised a further topic (G 1/24), I will permit myself one further observation. That is:
- your amicus for G 1/24 argued that the description and figures should not be generally consulted when interpreting the claims to assess patentability; but
- the EBA held that "The description and drawings shall always be consulted to interpret the claims when assessing the patentability of an invention".
I will concede that the EBA's ruling in G 1/24 did not precisely align with my views (that the claims should always be interpreted in context, including in the light of the description). But, bearing in mind the evidence of the written record, are you seriously trying to imply that your views in some way "prevailed" in the EBA's ruling?
@ Proof of the pudding,
DeleteI can admit without hesitation that I was as wrong as the president was in G 1/24, so there is no need to make a big deal of it. Why do you have to come with such a derogatory comment?
I never, whether seriously or not seriously, tried to imply that my views prevailed over those of the EBA. Have you used an AI having hallucinations to come up with this comment?
When lecturing, I always make clear that decisions of the EBA impose themselves, not only to all the parties, but also to first instance divisions and boards of appeal. Anything else would be a betrayal of the audience. So stop spreading nonsense about alleged statements of mine.
In view of the tone of the statements uttered by “Jackson Lamb”, I am getting an idea of who could hide beside this pseudo. The statements, and the tone in which they are uttered, are quite revealing.
The EBA's ruling in G 1/24 did not align with your views, precisely or not, as you did not simply want the description to be always taken into account, but you considered that interpretations in the description should supersede interpretations the skilled person can derive directly and unambiguously from the claim wording.
Furthermore, you still defend this point of view in spite of the very clear line of case law of the boards of appeal in the decisions following G 1/24.
By the way, I will make a summary of all the amicus curiae for G 1/25. Should you be displeased with my comment, and complain to the webmaster, you know what you can expect.
@ Anon Y. Mouse,
DeleteAs time changes constantly, the position of the EBA with respect to anonymous submissions might change as well, and become aligned to that of the boards.
What are you actually pursuing with your comment? Let the hamster wheel keep running? Unless you force me too, I will stop it.
As far as the Roche comment is concerned, it is by no way surprising as the company benefited from T 56/21 and the similar decisions.
I am not "pursuing" anything; merely pointing out that anonymous or pseudonymous submissions are not automatically to be disregarded just because of this, and that there is historical precedent for such submissions in proceedings before the EBA.
DeleteIncidentally, while the Technical Boards might generally disregard anonymous third party observations, this is not always the case. In T 56/21 the Boards invited the Appellant to comment on a set of anonymous third party observations. While this may be very unusual, perhaps even unique so far (though it is not possible to search the database exhaustively to confirm this), T 56/21 disproves that the Boards "never consider" such submissions.
I am happy to see that my amicus brief has been picked up even if it was a day late thanks to technical problems here at Slough House.
DeleteI am not sure what DXThomas means by "the tone of the statements uttered" if he is referring to the contents of my amicus brief. All I said was that discussions on this topic had previously been characterised by "increasingly intemperate language". This was a footnote to a wider point about the topic giving rise to acrimonious debates. I think this thread helps to illustrate my point.
I will not be commenting further.
PS. I am not Proof of the Pudding.
@ Anon Y. Mouse-04.02.2026 at 14:33:00 GMT
DeleteIt is due to T 56/21 and a handful of decisions alike that we have now G 1/25. On the other hand, it is good that the EBA has been invited to take a decision on the topic of adaptation of the description.
Yourself and Proof of the pudding are full of praise for T 56/21, and of Roche’s amicus curiae, but this decision is for me anything but a model of coherence. In view of the argumentation brought forward, the board in T 56/21 wanted absolutely to issue its decision before G 1/24, which was not very elegant. The board has also chickened out as it had announced that it would refer question to the EBA. Not a very exemplary attitude.
Every rule might suffer an exception, but I invite you to indicate to the audience in which other recent decision than T 56/21, a board has taken into account anonymous submissions. The board in T 56/21 wanted to get some more reasons for its views. I am inclined to think that yourself and the board in T 56/21 are/were really scraping the barrel to find arguments. .
@ Jackson Lamb,
DeleteYou know very well what I meant by "the tone of the statements uttered". I would allow myself to say that if you want to catch the attention of the EBA, you have to come with a proper legal argumentation, not merely provocative sentences and a collection of documents in favour of your opinion.
As you explain that you are “a European Patent Attorney in a large private practice firm with offices in several European countries”, you can a priori not be Proof of the Pudding. As you seek anonymity, nothing is however certain. I actually know who is hiding behind the pseudo Proof of the pudding, and he does not correspond to your alleged situation, even if in the past he was working for a firm of patent attorneys.
I allow myself to add that by wanting to hide deep in the description interpretations which are either more limited or broadened, applicants/proprietors are, in my opinion, defeating the whole patent system and the raison d’être of the claims.
It is good that G 1/24 has brought an end to this and that divisions of first instance will be forced to read the description before searching and before granting. What has been started under G 1/24 should be achieved with G 1/25.
I fully stand behind what I have said on IPKat in various threads, especially the following: “It should not be allowed to leave in the description statements which allow to raise doubts about the interpretation of the claims. This has nothing to do with patronising national courts or not leaving them without any possibility to interpret the claims.”
This is what I considered to be at the core of G 1/25, following G 1/24 The best example is what the CFI-LD Hamburg said: “chromatic” should have been deleted before grant. Was the CFI-LD Hamburg so wrong with this statement?
Having been in patents since the early 1970's, what I hope for here is for the EBA to put its finger on the judicial scales in a way that will start a trend towards greater legal certainty, in post-issue litigation over infringement and validity. I am one of those who hopes for the EPO to honour the provisions of Art 84 EPC to grant only claims that are clear, by decreeing that it is the originally filed description that must deliver "support" for those claims. Prosecution amendments to the claims are admissible if they clarify a claim that, as filed, was not clear enough as to the scope of protection sought. But forcing perfect consistency between the description and those clarified claims is, I say, a misconceived crusade for more legal certainty. Why, because in the real world, conforming the description doesn't enhance legal certainty. What it does instead is to open various mischievous, misconceived and even egregious validity attacks on the issued claims which otherwise would not be available.
ReplyDeleteDXT has suggested that the price (file wrapper estoppal) of leaving the description alone is too high, too burdensome for the tyro UPC judges. But litigants will pore over every detail of the prosecution history, regardless whether or not the description is conformed and judges will have to look intensively at the description originally filed, just as much in the case of a conformed description as in an unconformed description. I think therefore that the price DXT imagines is not as high as he supposes.
@ Anonymous,
DeleteIt happenes that have also been in the patent business since the early 70’s, and at the EPO since the day 1.
I can agree with you that Art 84 should be given more credit than has sometimes been the case at the EPO. As Art 84 is not a ground of opposition, many EDs allow themselves to cut this corner and not bother about clarity as such and support of the claim. This is time gained and leads to a higher production….
Faced with an unclear claim and a not adapted description can end in a very arduous opposition procedure. It is not different in post-grant procedures. G 1/24 is a good move in this direction as it should eventually bring all EDs to read the description before grant.
As a former search examiner, I have difficulties in understanding that a search can be carried out without reading the description beforehand. The same goes when examining.
Art 84, support, cannot be interpreted as meaning that the subject-matter of the claims is to be found somewhere in the description, i.e. that it goes only from the claims to the description. The reverse way is also to be looked at.
Claims and description form a whole in which one part cannot be played against the other be it during grant and during post-grant procedures. This is how I have been trained and later advised the examiners under my responsibility.
When an independent claim is limited by optional features present in the original description, but at the same time the description is not adapted, such a claim is, in my honest opinion, not supported by the description, as it allows later the proprietor to claim a different interpretation in infringement.
The day the notion of file wrapper estoppel will be introduced in the EPC, then I can agree that the description will not have to be adapted. This does however require a diplomatic conference. Doubts are permitted about the existence of the corresponding political will.
What the boards have done in applying G 1/24, is to bring, at the end of the day, applicants/proprietors to expurge from the description any discrepancy/inconsistency between claims and description.
In any case, when assessing added-matter or the same invention with respect to the priority one has to look at the file history, either in the application itself and/or in the priority document, when there have been amendments during examination.
Courts acting in post grant procedures are free to go through the file history, but this is a notion I have not seen used heavily in national proceedings, but in case of added matter, wher it is necessary
The CFI LD Munich has tried in the Hanshow case (UPC_CFI_292/2023) to use the file history for interpreting the claim. The UPC CoA (APL_8/2024-UPC_CoA_1/2024) has quickly put an end to it and made clear that claim interpretation has to be according its decision in the 10X and Harvard/Nanostring case (UPC_CoA_335/2023 App_576355/2023).
It is manifest that litigants will pore over every detail of the prosecution history, regardless whether or not the description is conformed and judges will have to look intensively at the description originally filed, just as much in the case of a conformed description as in an unconformed description.
However, this does not withhold the EPO to apply G 1/24 and Art 84, and thereby request applicants/proprietors to remove any inconsistency between claims and description. If this is not conforming claims and description, what else can it be? This is why I am also of the opinion that G 1/25 should be the continuation of G 1/24.
Daniel, sorry but I'm not convinced that the courts in Europe have easy work and the courts elsewhere in the world have hard work because the EPO insists on removal of all inconsistencies between description and claims. What is my explanation then, that alone of all the world's Patent Offices, the EPO requires such removal? I suggest to you that it is because the presence of inconsistencies offends the praiseworthy sense of "good order" that all EPO Examiners share, whether serving in an ED an OD or as member of a TBA. As you say, the description and the claims are different parts of a unitary document which should be read as a whole. Any internal inconsistency is therefore offensive.
ReplyDeleteTherefore, they think, to let through to grant any specification with any such inconsistency within it is to fall down on the job of examining every such application for perfect compliance with every provision of the EPC.
Do they think that? I suspect that they do and, until now, and uniquely in all the world, there has been no superior legal instance to set limits to their zeal to do what they see as a competent, correct and thorough job.
Dear Max Drei,
DeleteI read your comment with interest, but you will not be surprised when I tell you that you did not convince me.
To me, it is irrelevant whether the presence of inconsistencies “offends the praiseworthy sense of "good order" that all EPO Examiners share, whether serving in an ED an OD or as member of a TBA”. The EPO and the TBA are there to apply Art 84 and the decisions of the EBA, which includes G 1/24.
When you are of the opinion that inconsistencies between description and claims should be ignored by the EPO, you thus ignore the provisions of Art 84, and you consider that the EBA should not have issued G 1/24.
I would not say that any internal inconsistency is therefore offensive, it simply does not correspond to the requirements of Art 84 and what has been decided in G 1/24.
Whilst I can agree with you that claims and description ought to be considered as whole, I have hastened to specify that one part, the description or the claims, cannot be played against the other, the claims or the description, be it during grant and during post-grant procedures.
There has to be, as much as possible, a one-to-one relationship between claims and description. This hinders applicant/proprietors to change the interpretation as they see fit.
The way the boards apply G 1/24 is exactly to avoid inconsistencies between the two. I therefore repeat that G 1/25 should be the continuation of G 1/24, as both deal with inconsistencies.
Whether other countries, especially outside Europe, do not amend the description is actually irrelevant. The EPO and the TBA are here to apply the EPC as interpreted by the EBA.
When it comes to the interpretation of the EPC there is actually a superior instance, the EBA. Talking about “zeal” in relation to removing inconsistencies, is, to say the least, inappropriate and greatly surprises me coming from a practitioner whom I hold in high regard.
At grant, strict conformity with all requirements of the EPC will never be achieved. One example is that in opposition at the EPO or in national/regional procedures, a patent can be revoked for added matter.
You might also say here that the EPO and the TBA are very “zealous” in chasing added matter in a way which is not practiced elsewhere, but here again, the EPO and the TBAs are applying the decisions of the EBA.
That the way the EPO and TBAs deal with added matter, is not to the liking of lots of people, especially on the other side of the Atlantic, is again irrelevant. But at least all parties know where they stand. The same should be valid for inconsistencies between description and claims.
It is to be welcomed that the strict stance adopted by the EPO when it comes to added matter is to be extended to inconsistencies between claims and description
It needs more to be convinced that my position, and that expressed up to now by the EPO and the TBAs, under the guidance of the EBA, is flawed.
Thanks, Daniel. Of course it "needs more" to convince you to change your personal opinion. But here I am using the thread to try to reach a wider readership, and perhaps their opinions are less firmly held, and more open to persuasion, than is yours. In any event, then the EBA issues its opinion, it will make interesting reading, for both of us.
DeleteOne thing though: it came as a surprise to me that you find "zeal" to be a slur. I had thought it to be a compliment. Sorry for any unintended insult.
Dear Max Drei,
DeleteIt is clear that what you call the “wider readership”, are your colleagues. Their point of view is for most of them well known: no adaptation of the description.
In their vast majority, their position and yours can be summarised as follows: it is not mandatory to adapt the description to the claims to ensure legal certainty for third parties. While such adaptation may facilitate interpretation of the claims, it remains an excessive requirement not grounded in the provisions of the EPC.
My personal opinion is to the contrry, and is based on my experience as former member of the EPO and in the analysis of case law of the boards and of the EBA.
You will see my detailed point of view on G 1/25 in the Amicus Curiae I have send to the Registrar of the EBA and which should be published soon.
As far as the use of “zeal” is concerned, the difference in perception might lie in our different mother tongues. In the present context, using the term “zeal” had a clear negative connotation for a Frenchman, hence my comment. As far as I am concerned, the matter is settled.
Daniel I am very much looking forward to your Amicus Brief, as soon as it becomes available.
DeleteAs to "zeal" I asked MS Copilot whether it has, in French or German, any negative connotations. When I read the answer from the AI I realised at once why my use of that English word upset you. Perhaps we can add this unfortunate example to the list of "booby traps" when shifting from one EPO language to another. Rather like, "Die sensible Dame mit dem braven Hund".
As to readership, both yours and mine, actually my hope is that at least one member of the EBA follows the IPKat blog and will read this thread. As I see it, external members of the EBA have an important role, reporting to it from the courts called upon to enforce (or not) patents duly issued by the EPO and calling into question things that, within the EPO, are seen as set in stone.
Dear Max Drei,
DeleteAs far as readership of blogs is concerned, I am pretty sure that not only IPKat, but also other blogs are regularly read by members of the BA and hence of the EBA.
As far as non-EPO members of the EBA are concerned, I do not think that they have a determining influence in the decision making of the EBA. I have witnessed many OP before the EBA when discussing referrals. I remembered one occurrence in which the non-EPO member dared asking relevant questions. In general they are much more silent than vocal. I would therefore not expect too much from non-EPO members of the EBA, as on top of this they are in the minority.
In order to insure that really independent views are considered when it comes to referrals, the members of the EBA should not be members of the BA. This is a problem which has been built in the EPC, and which in the long run, appears problematic. The EPO had proposed in 2004/2005 a reform of the BA/EBA system aiming at true independence. Alas, after the Diplomatic Conference of 2000, the contracting states were not willing to embark on this topic. The reform of 2016, might have increased the perception of the independence of the boards, but not their true independence.
At the UPC the situation is different, but I am not sure it is better. According to RoP 234 UPC, a court of first instance can take a “Decision to refer” to the full Court of Appeal, when the case is of exceptional importance and, in particular, where the decision in the action may affect the consistency and unity of the case law of the Court.
This action is comparable to a referral under Art 112 EPC. At least at the UPC, the members deciding on the referral are not at the same time members of the Court of First Instance.
On the other hand, such an action is only provided for in the RoP and not in the UPCA itself. This means that this RoP can be amended, or even deleted, by the Administrative Committee of the UPCA under Art 41(2) UPCA. An action for rehearing, comparable to a petition for review under Art 112a EPC is enshrined in the UPCA itself, cf. Art 81 UPCA + RoP 245 UPC.
There is one common problem at the UPC and at the BA, a decision to refer is left to the discretion of a first instance panel. At the EPO, boards are very reluctant to refer questions to the EBA, even the corresponding request of parties appears justified. I guess that it will be the same at the UPC.
As far as “zeal” is concerned, I appreciate that you looked into it. I was not pleased when I read it and reacted. Quite normal, I think. For me its all OK, and it is an example of word to be used with care outside one’s mother tongue. Ende gut, Alles gut.
There are other words which can be called false friends (faux amis). One of them is “eventually”. In English it expresses a certainty, in French or German a possibility. It could not be more different!
Regarding “eventual”, this is one that I come across often and I am surprised that it is not better known to practitioners either side of the language divide as it can make a significant - even critical - distinction in understanding of patent language.
DeleteThe best illustration I am aware of so far, however, comes from football. To quote the Wikipedia article on “Dunglish”:
The word eventueel in Dutch means potentially and not eventually, which is uiteindelijk in Dutch. This mistake caused a row between the Scottish and Belgian football associations when the Belgian football association invited delegates from various associations over for the "eventual qualification of the Belgian national football team" before the play-offs against Scotland started. While the Scottish federation accused the Belgians of sheer arrogance, the Belgian association had actually meant to hold the event after a "possible" qualification.
Wasn't aware of the Dunglish page. Thanks for that. I was aware of "Globish" and will see now if there is a Wikipedia page for that.
Delete@MaxDrei & @DXThomas
ReplyDeleteI am pleased to follow up on MaxDrei’s outreach to the « wider readership ». If I am forgiven for a kind of inclusive neologism, I would dare to refer to a « writership », since all of us are fairly prolific commenters on IP blogs.
I am also pleased to note that the EPO appears to have put an end to the practice of substantial amendments entered by EDs at the Rule 71(3) stage without prior consultation with the applicant. The EPO has come to realise in the Spring of 2024 that this practice was a massive violation of Art 113(2) and a significant quality issue. Another significant development was the 2024 revision of the Guideline on clarity.
The feedback from SQAP panels each made up of 2 EPO experts and 2 patent attorneys to audit 6 files in respect of the grant procedure may have played a role in the EPO’s change of tack. I know from my participation in these panels that there was consensus of the panels for strongly suggesting to ban this practice. This was mentioned in the 2023 and 2024 SQAP reports attached to the EPO quality reports. This is why I disagree with Mr Thomas when he pits representatives against the EPO. The discussions within my SQAP panel have been extremely constructive and the energy of the directorate quality to promote the SQAP audits and provide ongoing support were much appreciated. The EPO deserves great credit for that. I am not aware of anything comparable in the other patent offices.
Mr Hagel,
DeleteAmendments to the description. and even sometimes to the claims, at the R 71(3) stage should, according to the valid instructions, always have been limited to minor corrections, mainly typos and the like.
During my active time, if the examiner in charge wanted to make amendments which had a direct influence on the scope of protection, he had to get at least approval by telephone from the applicant, and the approval had to be entered in Form 2035. This approval was not binding on the applicant, who could still disagree upon receipt of the communication under R 71(3).
It is the undue pressure for more productivity/production exerted by the upper management of the EPO which lead examiners to cut corners and to try to twist the applicant’s arm by proposing a grant, but at the same time making large amendments to the description and/or the claims. This move also saved a further communication, which count for nothing in the production appraisal system.
For the same token, it was a matter of fact that in some technical areas, the description was never read before the search or before grant. If this had been done the UPC CFI would not have had to complain that the description was not in conformity with the claims. It is to be hoped that G 1/24, and later G 1/25, will put an end to this practice.
It was thus high time that the EPO reacted and reintroduced a strict limitation of amendments at the R 71(3) stage, or at least requires a proper justification for a request for amendment. The EPO does not have to be praised for re-establishing a procedure which should always have been followed, and the deviation from it has only been tolerated as it increased production.
On this topic, and contrary to what you allege, I never pitted representatives against the EPO. That I have always been, on this topic, in full agreement with representatives and considered the attitude of some first instance divisions as not acceptable is, in my opinion normal. Requesting from first instance divisions to abide by the law, can in no way be considered an action against the EPO. Here you went way too far in accusing me of such an attitude.
Indeed, when lecturing on EPO procedures, and hearing the complaints, which often had nothing to do with the topic at stake, I have always advised the audience to resist such moves at the R 71(3) stage. I am also of the firm opinion that any amendment requested by the ED before grant has to be properly justified and not merely summarised, as I have seen, with a mere “cf. Art 84”.
This is why I am of the opinion that before grant, or before maintenance in amended form, the description has to be read and aligned with the claims. On this topic, I certainly have a different opinion than that of the vast majority of representatives, but this is my sincere belief.
Mr Thomas,
ReplyDeleteThank you for your answer. I am pleased that we agree to a large extent as to the 71(3) practice.
I also agree with you as to the need for examiners to read the description. And this should be done before the search, so that the scope of the search can take into account significant information such as the problem as exposed by the applicant, salient technical effects and possible fallback positions. In addition, when the description is "consulted" per G 1/24, this makes for more efficient examination. Logically, after G 1/24, the management can be expected to instruct and train examiners accordingly.
I
Mr Hagel,
DeleteIt is nice to hear from you that you were pleased that “we agree to a large extent as to the 71(3) practice”.
It remains that you alleged that I was, in this matter, pitting representatives against the EPO. I wish to retain that this allegation is totally incorrect, and that you have not withdrawn it as it should have been the case.
Your considerations relating to G 1/24 cannot compensate for this missing statement.
If a post considers that the colleagues of a patent practitioner are of the view that there should be no adaptation of the description, and the person writing that post has indicated that this is the current practice of the EPO, is this not pitting representatives against the EPO?
DeleteIs it? Perhaps "pitting" means one thing to a person whose first language is English and something different to a person whose first language is French. Or is it that context is everything and the meaning of pitting is dependent on context. Or both.
DeleteIn view of the last comments, I draw one conclusion: people no having as mother tongue, the language of the blog should avoid using terms which could lead to different interpretations or confusion.
DeleteIn Reasons 2, 1st §, of T 1756/23, the board noted that “the amendments underlying feature (g), which were apparently proposed by the examining division in its communication under Rule 71(3) EPC, introduce subject-matter which extends beyond the content of the application as filed Claim 1 AR3 infringed Art 123(2)”.
When looking at page 2 of Form 2025 one reads; “CLAIMS 1, 5: As agreed over the phone with the Applicant on 25-11-2020 (Guidelines C-VII, 2.6).
In this case the ED abided by the Guidelines, the then applicant tacitly agreed to the amendments by filing the translations and paying the granting fee, but the amendments introduced added-matter.
It till consider that reminding of the valid procedure, cannot be considered as taking position in favour of representatives and against the EPO.
Mr Thomas,
DeleteI was not correct when I wrote that you were pitting representatives against the EPO and I withdraw this comment. In your reply to MaxDrei of 17 January, you actually emphasised a division of opinion (I agree that « pitting » was overstated) between representatives (« your colleagues ») and yourself (not between representatives and the EPO.)
As far as I am concerned, the matter is settled and I acknowledge that you have withdrawn your comment.
DeleteI see that Daniel Thomas and Francis Hagel have now filed their respectiobservations in G1/25 on the EPO file. I offer the following comment here, to stir the pot a little more.
ReplyDeleteAlthough the EPO provides a service to inventors and applicants, its duty is to the public, society, the common good. The exclusive right given by the patents it grants are awesome in their power. It is reasonable then, that the EPO requires of applicants a set of documents to grant, which declare what the invention is, and then define in the claims the subject matter which the patent is to protect. That's not too much to ask, is it? Next point: just because the EPO is the only one of the world's large Patent Offices which requires the descrcription to be purged of inconsistencies does not mean that the EPO is wrong to insist on it. Increasingly, the world will come to look to multi-national Europe to set reasonable international standards for doing business. Finally, when patent dratters know that conformity is going to be required, they will have a stronger incentive to do a thorough patentability analysis of the prior art, before they start drafting. This will raise drafting standards, for the benefit of the public.
That said, it is the A (or WO) publication that is of interest to engineers and scientists. The B publication is not for them; rather it is for the lawyers, and they can understand a document that is a composite of the as-filed disclosure in support of the as-granted claims.
I await with great interest the reasoning in G1/25.
Dear Max Drei,
DeleteI read with great interest the following comment of yours: “just because the EPO is the only one of the world's large Patent Offices which requires the description to be purged of inconsistencies does not mean that the EPO is wrong to insist on it”. This is the first time I read something like this from you.
I do agree with you that often the silent player in the game is forgotten: the public at large, or in other words, the society. I do also agree with you that the “exclusive right given by the patents it grants are awesome in their power”. It is certainly reasonable “that the EPO requires of applicants a set of documents to grant, which declare what the invention is, and then define in the claims the subject matter which the patent is to protect” It is certainly not too much to ask.
I am also awaiting with great interest the reasoning in G1/25.
Me again, Daniel. I am glad we agree on many things. But I have to say that I also agree with much of what Pudding has written in this thread today.
DeleteI think that a B publication can declare what the invention is, and define what the scope of protection is, even when the EPO refrains from enforcing strict conformity between the description and the claims after their amendment during prosecution.
Applicants have to police their own amendments very carefully, lest their granted patent falls into the 123(2)/(3) fatal trap. I envisage similar self-policing in relation to Art 69 EPC. Courts are free to punish applicants who deliberately seed their descriptions with Angora cats. Courts are capable of construing claims on the basis of the originally filed description, taking it for what it is, a declaration of what applicant thought at the filing date was the invention.
You surmise that G1/24 sets up the EBA for a declaration in G1/25 that the description must be adapted. Conversely, I suspect that the UPC court of Appeal, in its new definition of the obviousness enquiry, has set the scene for a G1/25 declaration that the description does not need to be conformed. Hey Ho, isn't life interesting!
Dear Max Drei,
DeleteI am at a loss when on the one hand you agree with me, stating that “just because the EPO is the only one of the world's large Patent Offices which requires the description to be purged of inconsistencies does not mean that the EPO is wrong to insist on it”, but on the other hand you “agree with much of what Pudding has written in this thread today”. One or the other, but not both at once.
I can agree that a “B publication can declare what the invention is, and define what the scope of protection is”, but this is, in my opinion, only possible when the EPO enforces strict conformity between the description and the claims after their amendment during prosecution.
I can also agree that “courts are capable of construing claims on the basis of the originally filed description, taking it for what it is, a declaration of what applicant thought at the filing date was the invention”. The problem is that there is no file wrapper estoppel in the European patent system. Instead, Art 84 requires that claims are supported by the description, and Art 84 only applies before grant and before maintenance in amended form, even taking into account G 3/14.
I said it many times, it is not because national courts or the EPC are able to make head or tail of an ill adapted description, that the EPO has to ignore Art 84 to please applicants/proprietors.
Art 84 is not to be used when it comes to post-grant or post-opposition procedures, so that your conclusion that the “UPC court of Appeal, in its new definition of the obviousness enquiry, has set the scene for a G1/25 declaration that the description does not need to be conformed”, does not seem correct.
Courts are certainly “free to punish applicants who deliberately seed their descriptions with Angora cats”, but this is, as stated in T 2027/23, exactly what the EPO is there to avoid.
I regularly hear about fears of falling in the Art 123(2)/(3) fatal trap. I am still waiting to see one example that this has happened. It is clear that any amendment is potentially opening the Art 123(2)/(3) fatal trap, but avoiding Angora cats has nothing to do with this fear. I only know of one case in which this happened, it was in the UK, but there it was the applicant/proprietor who fell in the trap he created himself.
I agree with you with the need for a search and analysis of the prior art for efficiently drafting an application. This is the applicant’s homework. This is also valuable for a FTO analysis of the invention.
ReplyDeleteOn the EPO side, it has always seemed necessary for examiners to thoroughly read the description before they start the search and the assessment. But in too many cases, the review of the description was cursory. This should improve after G 1/24.
Today on the EPO Register there appears an Amicus Curiae brief from the UK Bioindustry Association. Does anybody agree with me that it includes a flaw in its reasoning, as follows.
ReplyDeleteThe BIA argues that, following G1/24, the way to address the B publication is to see the description as a record of what applicant saw as the invention, as sought to be protected at the filing date. But then the BIA goes on to argue in favour of voluntary amendment, to rid the description of inconsistencies which the applicant sees as problematic to claim interpretation.
The EPO is guardian of the public interest. If applicant shall have the right to polish the description during prosecution at the EPO, to purge it of all text that might cut against the patent owner in infringement litigation, should not the EPO, in pursuance of its duty to the public, have a right to impose on applicant amendments calculated to remove unfair benefits to a patent owner during enforcement of their patent rights?
In addition, the Amicus brief from the AIPLA (6500 patent practitioner members) observes that the description, if left alone, provides a common source of support for the the granted claims, the same all oder the world, which is good for legal certainty, all over the world. All fine and good, but hang on a minute. Does that not imply the imposition of a ban on amendment of the description during pre-grant prosecution at the EPO?
DeleteOr shall the applicant be allowed to fine tune and tweak the description during prosecution but not the EPO in pursuance of the public interest? Whatever happened to "equality of arms"?
Dear Max Drei,
DeleteI read with great interest the Amicus Curiae brief from the UK Bioindustry Association.
Whilst as a matter of principle, the UK Bioindustry Association does not want the description to be adapted, as for the association, the EPC does not require mandatory adaptation of the description to be consistent with amended claims, on the other hand the UK Bioindustry Association is in favour of voluntary description amendments, which “may be appropriate in specific cases”. Those cases are not even specified.
This position is quite odd as it leave the decision to amend to the applicant/proprietor. It is somehow surprising to see a group of applicants ignoring the EPC, especially Art 84, but are willing to adapt the description only when it suits its members.
To my knowledge, the EPC is valid for all domains of technique, and there are no specific legal rules in the EPC to take into account the specificity of biotech invention. I remembered that in the past, biotech candidates even requested a special biotech EQE. Needless to say that this wish stayed a wish.
It is not because “life sciences applications are drafted globally and at an early stage”, that they can be dispensed with later mandatory description amendments. There is nothing unfair for the EPO requiring such amendments. The burden is the same and nothing is unfair and disproportionate if the biotech sector is submitted to the same rules as any other technical area.
That the sector is such that applicants often maintain very large global portfolios and must navigate the many competing legal requirements imposed on it by different patent offices, does not imply a special treatment.
I see here between the lines the fear of the sector by acknowledging certain limitations of the scope that their patents could also be limited in the US. This is not an acceptable reason for requesting a derogatory treatment. The law is different on both sides of the Atlantic, but this should not come as a surprise.
Two points come to my mind, Daniel, when reading your comment of Thursday.
DeleteFirst point. Over in the USA, there is a thing called "patent profanity" which is concerned with the negative effects of court decisions on those who draft patent specifications. One example is the prohibition on ever including in the description the word "invention" in relation to the subject matter sought to be protected. Do we want that over here, for those engineers and scientists who read A publications for their technical content. But we are likely to get that here, if the EPO President has his way, with his Amicus brief in which he identifies the core G1/25 issue as a prosecution amendment which introduces "an inconsistency". One way to reduce the emergence of an "inconsistency" is to avoid saying at the outset what the invention is. Follow Robin Jacob's advice, EBA. Study what the Americans have done but then don't make the same mistake over here. I say again, also to the judges of the UPC, there is a difference between "the invention" as declared in the patent application and "the subject matter for which protection is sought" as defined in the claim.
Second point. Comparing the President's analysis with your "spade" hypothetical, if the A-publication defines a spade as an earth-moving tool, and the prosecution amendment in claim 1 is from "earth-moving tool" to "spade", well then that amendment has not introduced an inconsistency, has it? I am reminded of that old logic example "All dogs have a tail. This animal has a tail. Ergo, this animal is a dog". Just because the description states that a spade is an earth-moving tool does not mean that all earth-moving tools are spades. In the same way, narrowing the claim from earth-moving tool to spade does not necessarily introduce an inconsistency with a description that states that a spade is an earth-moving tool.
It is going to be quite a challenge, for the EBA to tell us what test to apply, to decide whether all inconsistencies, no matter how trivial, must be removed, or whether there are some that can remain on the basis that they are not going to be what crashes the process when the patent is litigated. De minimis non curat lex.
Another point for you, Daniel. In the UK, there is a thing called "Cake-ism", which means, having your cake and eating it too. Which means enjoying not only eating your cake but also "having" ie keeping it in the cupboard to eat at some future date. Impossible, but that doesn't stop them advocating it. The proponents of BREXIT, for example, were ridiculed as proponents of cake-ism.
DeleteMethinks all those Amici who want the right to amend their desciptions but at the same time want to forbid the EPO (acting in the public interest) from requiring the description to be conformed are just as much cake-ists as the BREXIT-ers.
As to the case in front of the EBA, I am thinking that "gathered" is a very suitable foundation for a debate about how much "conformity" is needed, in the public interest. I hope therefore that the EBA comes up with a line of reasoning that is good for the next 50 years and provides a roadmap for not only Europe but also the rest of the world.
It's not really "cake-ism" if you simply accept that there may be times when it is genuinely necessary to amend the description (for example, when there is an inconsistent or contradictory definition to a claim term, or simply an error in the description that should be corrected or removed), but that expected, procedural "inconsistencies" resulting from the very standard practice of inserting a formerly, i.e. at the filing date, "optional" feature into claim 1, should not meet such a requirement.
DeleteDear Max Drei,
DeleteFirst point, I am not concerned with what is happening in the USA under US law. You might speak about “the prohibition on ever including in the description the word "invention" in relation to the subject matter sought to be protected.” In this respect I would like to draw your attention to R 42 in which the word “invention” is mentioned not less than 8 times and to R 43 in which the word “invention” is mentioned not less than 7 times. Therefore, I cannot see a problem if an applicant/proprietor speaks about his “invention” in its original disclosure.
Whether the word “invention” is used in the original disclosure or not, I have difficulties to see how it could reduce the emergence of an "inconsistency". Not saying what the invention actually is may lead to problems of sufficiency, or if the latter is given, to problems of clarity and/or patentability.
If the A-publication defines a spade as an earth-moving tool, and the prosecution amendment in claim 1 is from "earth-moving tool" to "spade", such an amendment manifestly introduces an inconsistency. An earth-moving tool can be an excavator which is way different from a spade. There is no inconsistency only if the claim and the description mention either an "earth-moving tool" or a “spade” as both words define quite different items.
If the claim mentions a spade, but the description mentions an earth moving tool, prior art could be represented by a bulldozer, which is also an earth-moving tool as well as an excavator. If the claim mentions an “earth-moving tool” but the description mentions a spade, prior art can be represented by an excavator or a bulldozer. The difference in subject-matter claimed is by no way trivial, and hence your Latin citation appears not relevant.
To come back to the famous Agfa/Gucci case, it is by no way trivial if the description at grant mentions a “chromatic” base colour, but the claim has been limited to an “achromatic” base colour during prosecution.
The core G1/25 issue is indeed a prosecution amendment which introduces "an inconsistency". “Chromatic” and “achromatic” could not be more different. Why do you think that the UPC-CFI Hamburg complained that the word “chromatic” had not been deleted before grant? Certainly not to please the EPO’s president.
Any amendment to an independent claim has to find a corresponding amendment in the description. This is to insure that what is said in the description does not allow a different interpretation of what is actually claimed. See Agfa/Gucci.
Another old example of mine: if the independent claim is limited by the subject-matter of a dependent claim, then the description cannot state that the subject-matter of the dependent claim is still merely optional, as it is the case for every dependent claim.
Dear Max Drei,
DeleteFirst point, I am not concerned with what is happening in the USA under US law. You might speak about “the prohibition on ever including in the description the word "invention" in relation to the subject matter sought to be protected.” In this respect I would like to draw your attention to R 42 in which the word “invention” is mentioned not less than 8 times and to R 43 in which the word “invention” is mentioned not less than 7 times. Therefore, I cannot see a problem if an applicant/proprietor speaks about his “invention” in its original disclosure.
Whether the word “invention” is used in the original disclosure or not, I have difficulties to see how it could reduce the emergence of an "inconsistency". Not saying what the invention actually is may lead to problems of sufficiency, or if the latter is given, to problems of clarity and/or patentability.
If the A-publication defines a spade as an earth-moving tool, and the prosecution amendment in claim 1 is from "earth-moving tool" to "spade", such an amendment manifestly introduces an inconsistency. An earth-moving tool can be an excavator which is way different from a spade. There is no inconsistency only if the claim and the description mention either an "earth-moving tool" or a “spade” as both words define quite different items.
If the claim mentions a spade, but the description mentions an earth moving tool, prior art could be represented by a bulldozer, which is also an earth-moving tool as well as an excavator. If the claim mentions an “earth-moving tool” but the description mentions a spade, prior art can be represented by an excavator or a bulldozer. The difference in subject-matter claimed is by no way trivial, and hence your Latin citation appears not relevant.
To come back to the famous Agfa/Gucci case, it is by no way trivial if the description at grant mentions a “chromatic” base colour, but the claim has been limited to an “achromatic” base colour during prosecution.
The core G1/25 issue is indeed a prosecution amendment which introduces "an inconsistency". “Chromatic” and “achromatic” could not be more different. Why do you think that the UPC-CFI Hamburg complained that the word “chromatic” had not been deleted before grant? Certainly not to please the EPO’s president.
Two points, Daniel. First, the content of the description which "supports" the claims is disclosure. In keeping with that duty to disclose, US-style drafting uses the word "disclosure" instead of "invention". Readers of the A publication are interested in disclosure, less so in what the drafter of the description supposed to be the "invention" for which protection is being sought. Do they even bother to read the claims? I doubt it. So, I'm not sure how eschewing the term "invention" in the description can result in failure of the patent application under any of the Articles of the EPC.
DeleteSecond, on the question whether an inconsistency is or is not fatal, please see the reply I just typed to Pudding and Mouse. As to the Latin, whether any given inconsistency is no more than a trifle is the question I want the EBA to answer for me.
@ Max Drei-02.02.2026 at 20:42:00 GMT,
DeleteHow many times do I have to repeat that what matters here, is not what happens in the US and the way US applicants draft their specifications. What matters here is the EPC which is also valid for US applicants. In view of the way US applicants draft their specifications, it is no wonder that they come into problems when their applications end up at the EPO, just think of added matter. By the way, AIPLA want that the description remains unaltered, i.e. that the US system applies at the EPO.
You are as well mixing two issues. First, published patent applications are a source of technical information, cf. the theory of contract underlying the patent system. Second, the granted patent offers a temporary monopoly which, for the sake of the security of third parties has to be defined as clearly as possible, based on a description and claims as clear as possible and which cannot be based on varying interpretations of the proprietor, i.e. narrow when it comes to validity, but large when it comes to infringement.
Any inconsistency between independent claims and the description is by no means trivial. As just said above, it is by no means trivial when the proprietor can change its interpretation as he sees fit.
In answer, Daniel, I would say that eradicating all inconsistencies is not enough to prevent different parties from arguing for different interpretations of the claim. They are likely to be doing that anyway, with or without any inconsistencies.
DeleteBut with your final sentence, you offer a test of whether a seemingly harmless inconsistency is or is not objectionable: if it serves as a basis of support for a claim construction other than that given by the clear meaning of the words of the otherwise allowable claim, then perhaps it has to be purged because its presence has a knock-on effect of rendering the claim offensive under Art 84 EPC. Isn't this what CIPA argues in its Amicus brief?
The EPO Register is now full to bursting with Amicus briefs filed in the last two days.
ReplyDeleteNotably, Apple's lawyers plead for adaptation. I can well imagine that the combination of Apple + EPO President + considerations of the EPO's duty to the common good and the public interest will weigh heavier on the EBA than all the multitude of Amici who urge the EBA to come down on the side of not insisting on adaptation.
CIPA argues for adaptation on a case-by-case basis, with the party urging adaptation having (also in pre-grant proceedings) the burden of proving that the unadapted text would offend the provisions of the EPO. Thinking as a representative of an applicant, I'm not sure I want on me and my client the burden of proving to the ED that my requested adaptations of the text of the A publication are necessary, in order that the patent application shall meet all the provisions of the EPC. What do other readers think?
In the discussion about the requirement to adapt the description, the term « requirement » is misleading. Let’s assume an inconsistency between the amended claim and the description casts such doubt on the meaning of a term of the claim as to violate the clarity requirement of Art 84. An ED is not entitled to amend the description at the R 71(3) stage without the applicant’s agreement (Art 113(2)), as clarified in recent EPO documents. The ED may indeed issue a 94(3) communication objecting to a lack of clarity. But it is then for the applicant to decide on how to vercome the objection. The applicant may opt for amending the claim rather than the description.
DeleteNote that the applicant has to pay a price for its sole responsibility for the patent as granted. It implies that the EPO incurs no liability even if an amendment entered by the ED creates a new matter issue, as was the case (T 1473/19 ?) in which the removal of a comma in the claim resulted in the revocation of the patent by the Board.
I have looked at the post-G 1/24 proceeding of the « gathered sheet » patent. The patent proprietor submitted an auxiliary request to amend the description by deleting para [0035] which offers three definitions for the term « gathered sheet ». During oral proceedings in December, the Board took the position that this request is inadmissible for violation of Art 123(3). If this is confirmed, it will be interesting to look at the Board’s reasoning.
As to the submission of Apple's lawyers, I note that it is a creative use of an amicus curiae. This submission is actually used to provide arguments in a specific Apple case.
Max, would it be too much to hope that, when it comes to weighing the arguments in the various amici, the EBA would pay more attention to the substance of the arguments than they would to the identity of the author?
DeleteThe substance, Pudding? Even in the CIPA Amicus Brief, I couldn't find a test, how to determine whether any given "inconsistency" renders the claim unclear and thereby offends the provisions of Art 84, EPC.
DeleteI think CIPA, even in its role as a friend to the court, has dodged the question, the substance, the core issue. What say you?
@Max: but that's just the problem, isn't it? Several of the amici point out that "inconsistency" is ill-defined and (ironically) can mean different things to different people or deciding bodies due to its inherent subjectivity.
DeleteDoesn't that demonstrate neatly that the whole premise of requiring amendment of the description to remove any and all "inconsistencies" is ill-founded, as a case of the perfect being the enemy of the good? We all know that any claim, any description, is *always* going to be susceptible to multiple interpretations. That is the very nature of language.
Max, it might be more instructive to look at your question from a different perspective: is it even possible to formulate a definitive test for the presence of an "inconsistency"?
DeleteThere are numerous barriers to formulating such a test.
For example, what level of connection to the subject matter of the claims is required for a disclosure of the description to even be assessed for consistency? That is, can we discount disclosures that merely present background (or purely factual) information? And what about passages indicated as relating to a "disclosure", instead of "the invention" (or an embodiment or aspect thereof)?
Once we have overcome that hurdle, we then need to address the topics of how to determine the reference point (ie the precise subject matter with which the disclosures of the description should be consistent) and how to determine when a difference in subject matter rises to the level of an "inconsistency" that a provision of the EPC dictates must be addressed.
Given that neither the EPC not the travaux discuss the concept of (in)consistency between the description and the claims, I would say that there are no clear pointers to the correct tests for any of the above issues. This, in turn provides a clear pointer to the answer to Q1 (and to Q3) being "no".
The way I see it, Pudding and Mouse, the pursuit of meaning in the term "inconsistency" is not the key issue. What is key is the extent to which content in the description can render "unclear" a claim that is, otherwise, "clear". One is required to "consult" the description when construing a claim. Of course. Fair enough. Uncontroversial, I would think. But what if such consultation reveals an inconsistency? Is that enough to render an otherwise clear claim unclear? How shall I know, which inconsistencies render the otherwise claim unclear, and which leave the clarity of the claim untouched? Will the EBA tell us? I do hope so.
DeleteI doubt the Enlarged Board will give guidance as to the level of clarity required as this will be dependent on the facts of the individual case. My feeling is that the absolute requirement of Article 84 for the claims to be 'clear' will be interpreted by the EBoA such that if there is an inconsistency between the description and the claims with the result that the claims are not clear, then the description or the claims should be amended.
DeleteYes, Kant, but that is indeed the conundrum, isn't it, the issue that the CIPA Amicus brief zooms in on. What level of inconsistency does it take, to render a claim that is, in itself, clear enough to meet Art 84 EPC and turn it into a claim that offends Art 84 EPC.
DeleteWill any inconsistency do the trick, no matter how trivial, or are there some inconsistencies so unimportant that they fail to carry the claim over the border, from the patentable zone of compliance with Art 84 clarity standards into the kill zone of non-compliance with Art 84 EPC. When all one has is words, it is pretty much impossible to define with 100% clarity the subject matter for which protection is sought. Likewise, I suspect it is almost as hard to purge the description of all text that can be used in support of an argument that there is "an inconsistency" between words of the description and those of the claim. Where to draw the line? I don't know but CIPA is suggesting something.
One way to avoid inconsistencies arising, upon making a post-filing narrowing amendment of the claim, is to adopt US-style drafting and refrain from use in the description of the word "invention" instead using the word "disclosure". Do Europeans want US drafting standards over here? I suspect not.
I note that the last two amicus curiae submissions, dated 31January and 2 February, bring critical observations regarding the effect of removing inconsistencies by amending the description. The authors are Martin Wilming and a patent attorney writing under the pseudo « Jackson Lamb » (I guess this is ironical). Martin Wilming’s submission includes valuable insights on the meaning of « inconsistency ».
ReplyDeleteJackson Lamb’s submission is unusual in that it deplores that the topic tends to inspire « heated, even vitriolic debates », and has sometimes included « accusations that proprietors and their representatives may have improper motivations if they believe that there is no legal basis for requiring description amendments ». I personally think that debates even heated are welcome but it is beyond the pale if they are vitriolic and include such accusations.
As to UPC decisions, Agfa v Gucci has been given a lot of attention. But another decision, Paris LD 13 November 2024 HP v Lama, point 41, is quite interesting. In this case, the defendant argued that Figures 4-8 should not be taken into account in the interpretation of claim 1 because they were explicitly mentioned as « unclaimed examples ». The court dismissed this argument, stating that figures 4-8 are part of the description and may when appropriate be used to interpret claim 1. This highlights the explanatory value of the description.
Just because the period for filing Amicus briefs has closed does not mean that the flow of third party observations has stopped. See the one dated yesterday from Ilya Kazi, for example. He suggests that a not-invalid claim reciting "up" could nevertheless be invalid if i) down is not patentable and ii) the description defines "up" as "up or down". To me, that is an argument for requiring descriptions to be "conformed" every time i) the claim is narrowed such that the description of at least one illustrated embodiment of "the invention" fails to recite explicitly the prresence in that illustration of every single feature of the claim. other feature recited in the claim.
DeleteNow, in chemistry it is (I guess) relatively easy to decide whether any given embodiment exhibits the claimed feature combination. In mechanical engineering, however, it is not always easy. Take the feature "ratchet", for example, added during prosecution. Fig 1 shows a pawl engaging a toothed wheel, that is, a classic "ratchet". Fig 2 shows an arrangement of two tubes, telescoped, one inside the other, as in the devices useful as a curtain rod for a curtain spanning the open fourth wall of a shower cubicle. The skilled reader grasps that they can pull the two tubes to extend their combined length but not push them to make their combined length shorter. This is what I would call a ratcheting effect.
To satisfy the EPO must I cancel Fig 2 or declare it to be NOT the claimed invention, just because the description falls short of stating explicitly that the two tubes provide a "ratchet"? Like so many of the Amici, I think leaving Fig 2 unamended helps rather than hinders Art 69 EPC considerations of "infringement by equivalent" but I can imagine an EPO Examiner seeing the issue differently. I hope the EBA will help to reduce tensions between Applicants and Exam Divisions, over the question when I have to declare any given disclosure as inside or outside the scope of protection which the applicant is seeking. As Mr Kazi observes, it is not just the time it takes to "conform" the description to the satisfaction of the EPO. It is also the time it takes to argue what "conformity" means. With words it is never possible to achieve 100% clarity. A level of clarity that is "fit for purpose" should be enough. Likewise with "consistency".
Max, I don't see Ilya's argument as pointing to any need for a description to be "conformed". Quite the opposite.
DeleteTo illustrate my thinking, I would ask you to consider the following question: faced with the same description and claim set as described in Ilya's amicus, how would a national court (such as a court in the UK) interpret the claim?
Would the court not, in view of Article 69 EPC and its Protocol, interpret the term "up" according to the description definition? And, if they would, then why - in view of Reasons 19 of G 1/24 - would the EPO interpret the claim without reference to the description definition? Is that not the core of the problem that the EPO has created for itself?
Sorry about the lack of clarity in that last post. Some of my words disappeared and I didn't spot it before I hit the send button. Overnight, it occurs to me though that assessing what is an Art 69 EPC "equivalent" might be just as tricky in chemistry as it is in mech eng. Pemetrexed disodium, for example. Imposing on pre-grant patentability examination and claim clarification the full force of Art 69 EPC strikes me as disproportionate over-egging of the pudding being cooked within the walls of the EPO.
DeleteOK, Pudding, I dare say that I have mischaracterized Ilya Kazi's observations. Nevertheless, I am still not understanding what he is urging the EPO to do when his applicant, during prosecution, imports "up" into claim 1, leaves in the description the definition "up includes down" and contends that the application is good to grant because all EPC Articles are satisfied, patentability, clarity and support. If "up" is clear, supported and patentable, must the "up includes down" statement be excised, or can it be allowed to remain? Reasonable minds differ on this issue, right?
ReplyDeleteMax, it seems to me that Ilya's example assumes a post-grant scenario, as all of the remedies proposed flow from the (broad) interpretation of the claim. In a pre-grant scenario, my view is that the EPO would be justified in raising an objection under Art 84 EPC (whether a lack of clarity or a lack of essential features). That is, the applicant should clarify in the claim a definition of "up" that:
Delete- is clearly and unambiguously derivable from the disclosures of the application as filed; and
- provides, together with the other features of the claims, a (patentable) distinction over the disclosures of the prior art.
In Ilya's example, it is unclear whether it would be possible for the applicant to define "up" in a manner that satisfies both of these requirements. This is because the generic term ("up") is defined using precisely the same word ("up") as one of the specific alternatives falling under the genus. Really, a case of (hypothetically) very poor drafting. Nevertheless, I am sure that a solution can be found in most real-life cases.
Once a solution is found, then:
- the amended claim, as interpreted in the light of the description, will meet the requirements of Article 84 EPC; and
- allegedly "inconsistent" disclosures of the description will therefore have been determined to not give rise to a lack of clarity.
In other words, where a claim truly satisfies the requirements of Art 84 EPC, any deciding body should have no difficulty in identifying the description disclosures that are "inconsistent" with the claim (and that therefore fall outside of the scope of the claim).
Be that as it may, Pudding, I am still hoping for an answer to my hypo about the word "ratchet" in the claim. Fig 1 shows a toothed wheel and pawl and the written description identifies this as a ratchet. Fig 2 shows two tubes one sliding inside the other. You can extend the combined length but not shorten it, once extended. You know, the typical shower curtain rail. It delivers a ratcheting effect, but the decription of Fig 2 fails to include the word ratchet. The EPO demands excision of Fig 2, or else the insertion of wording that Fig 2 is NOT an embodiment of "the invention".
DeleteArt 69 EPC contemplates equivalents as falling within the scope of protection. Fig 2 meets the function-way- result test for equivalence. Therefore, I say, to require its excision, or labelling that it is NOT an embodiment of the subject matter for which protection is sought, is unjust, unfair to the inventor, and discouraging to those who would include in their patent applications a greater rather than lesser degree of disclosure.
Am I here being unfair to the EPO? How many ED's would allow me to go to grant with Fig 2 present and described as an embodiment of the invention, and how many would require its excision? Is it clear or does it need the EBA, in G1/25, to give the Examiners more guidance?
Max, it is a little disappointing to make a carefully reasoned point only to receive a “Be that as it may” response and an abrupt change of topic. It rather suggests that you cannot be bothered to think about it.
DeleteIn any event, you are not being unfair to the EPO in connection with your hypo. Instead, you have merely recognised an issue behind one of the points that I made in a response to you on 2 February, namely the challenge of determining the reference point, ie the precise subject matter with which the disclosures of the description should be consistent.
@MaxDrei and Proof of the Pudding
ReplyDeleteRegarding the “up” vs. “up or down” scenario mentioned by Ilya Kazi, I note a similarity with the post-G 1/24 proceedings before the Board of the referring decision T 439/220. The owner of the patent-in-suit EP3076804 filed in July 2025 an auxiliary request to delete from the description para [0035] which sets out three different meanings for the term “gathered sheet”.
The minutes of the oral proceedings held on 17 December mention that the Board is minded to hold the request inadmissible for violation of Art 123(3) but provide no reasoning to support this position. This suggests that the deletion of “down” in Ilya Kazi’s scenario might be held inadmissible as well.
Francis, the minutes refer to "Article 123(3) EPC (removal of a former restriction IMPLICITLY contained in the claim)". I think that this provides a pretty clear signal about how the Board applied the ruling in G 1/24 to the case at hand. Nevertheless, it will be interesting to see the Board's detailed reasoning on this point.
DeleteRe: the deletion of "down" in Ilya's hypothetical example, that would indeed appear to be problematic from an Art 123(2) EPC perspective. On the other hand, inserting "wherein up means down" would appear to be a potentially permissible amendment (ie deletion from a single list of alternatives) that arguably - even without amendment of the description - satisfies the Art 84 EPC clarity requirement.
@Proof of the pudding
DeleteYou are right, the minutes refer to the removal of a former restriction, but I do not find any restriction where the deletion rel
@Proof of the Pudding
DeleteYou are right that the minutes refer to the "removal of a former restriction implicitly contained in the claim" but para [0035] of EP3076804 sets out an expansion vs. the ordinary meaning of "gathered sheet", not a restriction.
As to a hypothetical "up means down", it sounds like typical Orwell's novspeak "war is peace" or "freedom is slavery"" and I feel it should be dismissed as self-contradictory, regardless of whether it may be arguably permissible.
Yes, I too noticed the expansion vs. restriction point. However, the Board's "restriction implicitly contained" comment would seem impossible to explain if the claims have not been interpreted in the light of the description (and, in particular, the definition whose deletion would allegedly cause an Art 123(3) EPC issue).
DeleteAs for "up means down", that is merely the consequence of the unfortunate language selected by the (hypothetical) patent applicant. Also, once you get over the echoes of Orwell, the meaning of the language is clear enough.
Patent applicants should be free to describe their inventions in the terms that they believe are the most appropriate. Also, "unusual" definitions used by real-life applicants are rarely (if ever) quite so self-contradictory at first blush. For these reasons, it is important to recognise when - despite the presence of an "unusual" definition - a claim is clear enough.
It is always fascinating to read these threads where DX Thomas spends so much time attempting to undermine well-founded and logical arguments.
ReplyDeleteRegarding the function of Art 84 EPC: The requirement for 'support' is there to ensure that the subject matter of the claim has a basis in the specification—essentially, to ensure that what is claimed is actually disclosed. The intent was never to turn the description into a mirror image of the claims by deleting everything that is no longer encompassed by the allowable claims.
If the Enlarged Board of Appeal eventually decides to justify why there is supposedly legal support to force these amendments, I sincerely hope they take the time to go argument-by-argument through why T 56/21 is wrong. The Board in T 56/21 took the time to consider the arguments raised and provided distinct legal reasoning for each one. Pushing that decision to the side as if it does not exist is not a strong argument.
We all know that 'democracy' is not what drives EBA decisions, but if we look at the numbers, the message from the end-users is clear: there is no support for this practice. Furthermore, it is clear that in other major jurisdictions, no such requirement exists.
As mentioned in multiple briefs and as is well known to everyone familiar with the Travaux Préparatoires, the only purpose of Art 84 at the time of drafting was to ensure that the claims are clear in themselves and that corresponding passages could be found in the text as filed. I personally struggle to understand how a simple English sentence has been twisted to mean something completely different than what was clearly intended.
I am hopeful that the EBoA will not be influenced by the EPO President’s position and that the historical basis for Art. 84 EPC will prevail. Once the decision is out and the initial discussions subside, things will likely get very quiet on the blogs... As controversy will be back to normal...
It is always fascinating to read these threads where DX Thomas spends so much time attempting to undermine well-founded and logical arguments.
ReplyDeleteRegarding the function of Art 84 EPC: The requirement for 'support' is there to ensure that the subject matter of the claim has a basis in the specification—essentially, to ensure that what is claimed is actually disclosed. The intent was never to turn the description into a mirror image of the claims by deleting everything that is no longer encompassed by the allowable claims.
If the Enlarged Board of Appeal eventually decides to justify why there is supposedly legal support to force these amendments, I sincerely hope they take the time to go argument-by-argument through why T 56/21 is wrong. The Board in T 56/21 took the time to consider the arguments raised and provided distinct legal reasoning for each one. Pushing that decision to the side as if it does not exist is not a strong argument.
We all know that 'democracy' is not what drives EBA decisions, but if we look at the numbers, the message from the end-users is clear: there is no support for this practice. Furthermore, it is clear that in other major jurisdictions, no such requirement exists.
As mentioned in multiple briefs and as is well known to everyone familiar with the Travaux Préparatoires, the only purpose of Art 84 at the time of drafting was to ensure that the claims are clear in themselves and that corresponding passages could be found in the text as filed. I personally struggle to understand how a simple English sentence has been twisted to mean something completely different than what was clearly intended.
I am hopeful that the EBoA will not be influenced by the EPO President’s position and that the historical basis for Art. 84 EPC will prevail. Once the decision is out and the initial discussions subside, things will likely get very quiet on the blogs... As controversy will be back to normal...
@ whoknows=Anonymous of 09.02.2026 at 18:10:00 GMT
DeleteIf I would have found the arguments well-founded and logical, I would not spend a single second contradicting them.
T 56/21 is one decision which has failed to convince me. As the decision is at odds to the views of the EPO and the boards in matter of adaptation of the description, the mere decency would have required from that board to address a referral to the EBA as it had announced previously. It rather chickened out and decided to ignore the importance of the question and placed the EBA before a fait accompli. It decided ex-cathedra, for whatever reason.
Where do you take that “The intent was never to turn the description into a mirror image of the claims by deleting everything that is no longer encompassed by the allowable claims”? Any source other than IA?
It is your interpretation, and that of your colleagues, to claim that Art 84 is unidirectional. To be clear, if the subject-matter of the claims could not be found in the description, it would be ridiculous. Seen in this way, Art 84 boils down to crash an open door. Support goes in both directions, whether you like it or not. It goes in both directions for the simple reason that it has also for effect that the description cannot contain statements at odds with the claims. on the other hand, if everything is to be found in the description, why bother with claims? Nothing has been twisted when it comes to Art 84. It is rather a unidirectional interpretation of Art 84 from yourself and your colleagues.
It should also not be forgotten that Art 84 stems from Art 6 PCT which goes further as it added that the claims should be “fully supported”. It is not a question of having drawings or not to be part of the description, but really to find an adequation between claims and description. This is the work already done in decisions following G 1/24. If you want a interpretation found in the description but different than that derivable by the skilled person from the claim wording, then bring it in the claims.
For the surplus, I refer to my comments to “The Irony” as far as the history of the EPC is concerned. In the same reply, I added that it is not the number of briefs against adaptation of the description which should prevail.
I hope that the EBA will not give a political reply to the questions referred as it was the case in G 1/21 and G 3/19. The EBA is perfectly in a position to look at the historical basis for Art 84. I further hope that the EBA will have recourse to the interpretation rules set out in the VCLT.
interesting to mention Article 6 of the PCT here as basis for Art. 84 EPC... how comes that the wording "claims to be fully supported by the description" of Art. 6 PCT has not been interpreted in the same bidirectional way by most other major jurisdictions? at the very least, this is a strong indication that the bidirectional interpretation of the wording of Art.84 EPC would NOT be the only reasonable interpretation. Article 6 PCT is certainly considered and interpreted by the USPTO, UKIPO, JPO and KIPO. In view of the equivalent wording of Article 6 PCT and Art. 84 EPC (although the EPC's provision lacks the "fully" of "fully supported", so Art.84 EPC appears a bit weaker when compared with Article 6 PCT), this means nothing else that all of these patent offices (USPTO, UKIPO, JPO and KIPO) have considered that equivalent wording of Article 84 is not to be understood bidirectional. The irony being that it is probably primarily the EPO's (unmatched) strict approach towards added-subject matter that would prompt any cautious applicant to include a plethora of fallback positions (including e.g. claim-like clauses) in the first place! @DXThomas: can you acknowledge that the EPO's practice has increasingly become more restrictive on this topic or do we live in different universes?
Delete@ not convinced-12.02.2026-17.20.00 GMT
DeleteI had overlooked your question to myself and I hope you can accept my apologies.
My response is simple: different granting authorities may have a different interpretation of one and the same legal rule. The USPTO, UKIPO, JPO and KIPO may have their own interpretation of the notion of support, and the EPO is entitled to have its own. Or are we here to say that weight is to be given to an interpretation that is valid in the highest number of granting authorities?
At this rate, why not weigh up Amicus Curiae, which say NO to Q1 for 72,5% against the 17,5% having answered YES to Q1? The decision would be much easier and quicker to take.
In a more recent comment I have acknowledged that the more restrictive approach on the adaptation of the description has been provoked by the famous “pemetrexed” case. Please refer to it.
Claim like clauses are an invention of US applicants, quickly taken over non-US ones, but I fail to see any reference to claim like clauses in the EPC or its implementing Regulations.
The EPO’s stance on added matter might look severe at a glance, and is certainly not to the liking of applicants/proprietors, especially those from the US, but the coherent body of case law relating to novelty is such that any user of the European patent system knows where he stands. To me added matter and novelty are the two faces of the same coin.
@ not convinced-12.02.2026-17.20.00 GMT
DeleteI had overlooked your question and I hope you can accept my apologies for this late reply.
My response is simple: different granting authorities may have different interpretations of one and the same legal rule. The USPTO, UKIPO, JPO and KIPO may have their own interpretation of the notion of clarity and/or support, and the EPO is entitled to have its own. Or are we here to say that weight is to be given to an interpretation that is valid in the highest number of granting authorities?
At this rate it could be envisaged to weigh up Amicus Curiae, which say NO to Q1 for 72,5% against the 17,5% having answered YES to Q1? The decision would be much easier and quicker to take.
In a more recent comment I have acknowledged that the more restrictive approach on the adaptation of the description has been provoked by the famous “pemetrexed” case. Please refer to it.
To my knowledge, claim like clauses are an invention of US applicants, quickly taken over non-US ones, but I fail to see any reference to claim like clauses in the EPC or its implementing Regulations.
The EPO’s stance on added matter might look severe at a glance, and certainly not to the liking of all applicants/proprietors, especially those from the US, but the coherent body of case law relating to novelty is such that any user of the European patent system knows where he stands. To me added matter and novelty are the two faces of the same coin.
I fed the URL of this page into an AI (Claude) with the prompt:
ReplyDelete"Analyse this discussion and tell me what you infer about the relative strengths of arguments and about the personalities of the commenters."
This is what it came up with, with no further prompting from me. Part 1; other parts to follow.
This is a fascinating and quite heated debate about European patent law. Let me analyze the arguments and commenters' personalities:
Strength of Arguments
Arguments AGAINST mandatory description adaptation (FICPI, Schmid, Roelands, "Proof of the pudding", Max Drei, Anonymous):
Strongest points:
Legal basis: Article 84 EPC is unidirectional - claims must be supported by description, but there's no reciprocal requirement
Practical risks: Forced deletion of embodiments can create Article 83 (sufficiency) and Article 123(2) (added matter) violations
Unnecessary: Courts like Hamburg in Agfa v. Gucci can already handle inconsistencies without prior EPO intervention
International anomaly: No other major patent office requires this
Cost burden: Creates unnecessary billable hours for attorneys
Arguments FOR mandatory description adaptation (Kortekaas, DXThomas):
Strongest points:
Legal certainty: Prevents "Angora cats" where proprietors use narrow interpretation for validity but broad for infringement
G 1/24 alignment: The requirement to "consult" the description means inconsistencies must be removed
Article 84 support: Claims cannot be truly "supported" by a description containing contradictory information
Harmonization: Ensures EPO and national courts/UPC reach consistent interpretations
Argument Quality Assessment
The anti-adaptation camp has stronger legal textual arguments - they can point to the absence of explicit EPC provisions requiring adaptation. Their risk-based arguments (sufficiency, added matter) are particularly compelling.
The pro-adaptation camp relies more on teleological interpretation and policy goals (legal certainty, preventing gaming). Their arguments are philosophically coherent but struggle to identify clear EPC article support.
The referring Board's skepticism in T 697/22 (point 16.8, noting courts weren't "fooled" by unadapted descriptions) suggests even the EPO recognizes adaptation may be solving a non-problem.
@ Odo Ital-10.02.2026-00.13.00 GMT
DeleteI read with great interest the result of the question you put to an IA.
I do not know how your IA has been trained, so that for this reason alone, its reply is subject to caution. In particular, there are other Amicus briefs which are in favour of adaptation of the description than that of Mr. Kortekaas and myself which have apparently have been ignored by the IA.
Look at the analysis in my blog and you will find a few other: Mr. P. de Lange, Mr A. Cossu and the German Chamber of Patent Representatives. Those do not appear to have been taken into account in the reply of your AI. Their argumentation is, at least for me, much more convincing than that of the anti-adaptation group.
The same goes for the anti-adaptation group, which is much larger as your IA appears to know. Not that their argumentation is different, it is a mere repetition. A patent does not become novel if the proprietor repeats that it is novel like a Tibetan prayer mill. The argumentation is as unconvincing as that your IA has selected.
I do not agree that the anti-adaptation camp has stronger legal textual arguments. The EPC has to be read and interpretated according to the interpretation rules set out in the Vienna Convention on the Law of Treaties. Your IA seems to ignore the VCLT.
When interpreting the EPC in the light of the VCLT and of the Travaux préparatoires, other conclusions are possible, not to say more convincing.
When you mention Reasons 16.8 of T 697/22, you or your IA should have mentioned Reasons 16.6: “The local division further held that the deletion of the feature "chromatic colour" in claim 1 during examination proceedings while the description remained the same had created an inconsistency between the granted claims and the description. The passages in the description mentioning "chromatic" should therefore have been deleted”. This is a direct plea for the adaptation of the description. Your IA seems to suffer from some inconsistencies.
Mr Thomas: You appear to have missed that the commenter above says that they gave the AI a link to "this page" and asked it to analyse "this discussion". Any absences in references to amicus briefs or to aspects of the T 697/22 would therefore be reflections of what is on this page itself.
DeletePerhaps someone can do an experiment by pointing an AI to the EPO's page containing all the amicus briefs, and asking the AI to analyse that instead?
What good is it to only have some amicus briefs being looked at by an IA. Either all or none.
DeleteI have not missed anything. Just looking at the present page gives a distorted view.
DeleteIf somebody feels the necessity to call for an IA, then do it on all amicus briefs and not just about the odd comments on this page.
Part 2:
ReplyDeleteAccording to my AI - again, I emphasise, with absolutely no guidance, this is what it came up with after being prompted by this very webpage:
Commenter Personalities
DXThomas (former EPO examiner):
Personality: Dogmatic, combative, defensive of EPO institutional authority
Style: Lengthy, repetitive, somewhat condescending ("you seem to carefully ignore")
Tells: Frequently invokes his EPO credentials; takes disagreement personally; misreads opponents (thought "Proof" was advocating description supremacy when they clearly weren't)
Weakness: Circular reasoning - often asserts EPO practice proves EPO practice is correct
Proof of the pudding:
Personality: Methodical, patient (initially), increasingly exasperated
Style: Socratic questioning, exposes logical contradictions
Strength: Identifies the core paradox - if EPO interprets claims in isolation (primacy), why does description content matter? If description matters, why not interpret claims accordingly?
Tells: Eventually taps out when DXThomas keeps repeating the same points
Max Drei:
Personality: Diplomatic elder statesman, somewhat ironic
Style: Concise, philosophical, focused on systemic issues
Insight: Suggests EPO's motivation is aesthetic ("good order") rather than legal necessity
Cultural intelligence: Graciously accepts DXThomas's explanation of "zeal" misunderstanding
Anonymous (since 1970s):
Personality: Pragmatic, experience-based
Style: Brief, cuts to practical consequences
Focus: Legal certainty in litigation, not theoretical purity
Anon Y. Mouse & spinning and purring:
Personality: Procedurally knowledgeable, detail-oriented
Contribution: Factual corrections, precedent citations
Jackson Lamb:
Personality: Sardonic, self-aware about using pseudonym
Style: Meta-commentary on the debate's temperature
The Irony
The debate itself proves the anti-adaptation camp's point: if experienced patent professionals can't agree on whether there's a legal basis for adaptation, how does forced adaptation enhance "legal certainty"? The Agfa v. Gucci case shows courts can navigate inconsistencies. Meanwhile, DXThomas's repeated invocation of that same case (claiming it supports his view because the court said "chromatic" should have been deleted) illustrates confirmation bias - he ignores that the court reached the correct result anyway.
The telling moment is DXThomas admitting "this Kat heard directly from a partner at a top-tier firm (which shall remain nameless) that the billable hours were one reason he remained in favour of the requirement." blogspot This suggests economic incentives may be driving policy more than legal necessity.
Could the glorious anonymous, The Irony, give the source of what he alleges with respect to billable hours. Do you really have to scrape the bottom of the barrel for lack of well-argued reasons?
DeleteComing now to something more serious, if you read carefully my analysis of the Amicus Curiae, you will see that some national courts, and even the UPC, are not so well able to navigate with inconsistencies. They can be used against a potential infringer, although I do not find this correct.
In all correctness and fairness, when you said that in the Agfa/Gucci case the court did not fall for the inconsistency with “chromatic”, you should have been compelled to add that the same court insisted on the fact that “chromatic” should have been deleted before grant. Half truth are always better for those who are afraid of the whole truth.
If experienced patent professionals can't agree on whether there's a legal basis for adaptation, this is not a reason to simply conclude that there I no need to adapt the description. Experienced patent professionals are not there to say what the law is. This is the role of the BA and of the EBA.
It is noticeable that the German Chamber of Patent Attorneys has come up with a well-reasoned statement in favour of the adaptation of the description, contrary to all other associations of representatives claiming that there is no basis in the EPC requiring the adaptation of the description. Even the GRUR Standing Committee on Patent Law, did not take position on Q1 in view of the varied position of its members, but made useful suggestions as how to handle the problem.
I doubt that the EBA will be impressed by the fact that the vast majority of the Amicus Curiae are in favour of not adapting the description. What is at stake is the law and the rule of law, not the opinion of representatives. I trust the EBA to say whether there is a basis or not in the EPC.
The Amicus Curiae in favour of the adaptation of the description have given good arguments, even going back to the Travaux Préparatoires to argue this necessity.
What is forgotten in all the Amicus Curiae refusing adaptation of the description, is the history of the EPC. Contrary to all national systems, in which national courts can iron out inconsistencies having slipped through the granting procedure, the EPO has to deliver a patent which should not allow the proprietor to interpret its patent as he thinks fit in post-grant procedures.
It should also not be forgotten, that there were no less than four drafts of a European patent in the framework of the the "Common Market", now the EU. Those drafts had foreseen an integrated system like the present national systems. They never materialised in view of the obstacles put in the way by a certain country. It is only the pressure of the PCT which forced the Europeans to come to an agreement, and this agreement was then opened to non-EU member states. In this respect the EPO cannot be compared to a national patent offices.
I never got a sensible reply on this point.
Daniel, you're shouting at an AI. As the author pointed out above, the posts from "Odo Ital" are an AI-generated summary of this blogpost and its comment thread.
DeleteIt looks like the section headed "The Irony" is the AI's summing-up. In that context it seems to have hallucinated that you made a comment about billable hours. Using CTRL+F I was able to locate those comments as being in the main post itself.
@ Odo Ital-10.02.2026-00.14.00 GMT
DeleteTo be honest I could not give a damn of what an unknown IA says about myself. Moreover, not knowing how this IA has been trained, its output is anyway subject to caution, even if it would say nice things about myself.
When I had enough, I make it clear, and this has nothing to do with being condescending. It is just an outlet of my exasperation.
My credentials go far beyond what I have done at the EPO. I was not only examiner, I was also in charge of various groups of examiners. What has been a great satisfaction of mine, is to see plenty of examiners I had in charge to be appointed as directors or as members of the boards. You can draw certain conclusions, but they cannot be negative as your IA claims.
For over 40 years, I have been training candidates to the EQE. If I would have been as bad as your IA makes out, I would not have survived as long as I did in this field.
I am also regularly consulted by industrial companies and firms of patent attorneys.
As former staff member of the EPO, my loyalty goes manifestly to the EPO, but this does not mean that I approve everything coming from it. I also criticise the EPO when I find that critic is due, be it about the work of first instance divisions, or about the way boards act when it comes to procedural aspects. Just have a look at my blog.
My conclusion is that your IA is rather ignorant. What comes out of it is nearer a bin than a pedestal.
Again, you are shouting at an AI.
DeleteIf (as the poster says) the AI was asked to analyse this IPKat page and its comments, then any conclusions the AI has drawn about any individual are probably based on what is on this page, no more and no less.
As you are anonymous, please refrain from any familiarity.
DeleteAs one would say in my home country, we have not looked after the cows together.
I am not shouting at the AI. I have justified reservations.
Simply analysing one page of the IPKat is not correct.
There is a well known meme in certain corners of the Internet which runs along the lines of "I'm not mad. Please don't put in the newspaper that I got mad."
DeleteIf DXThomas is so unbothered by an innocuous comment such as that above, why spend hundreds of words justifying his position and doing so based on a blatant appeal to authority?
Mr. Thomas,
DeleteYou note in reply to the commenters above that "we have not looked after the cows together," which I understand is a French idiom about social familiarity.
But this is an English-language forum. Enforcing French social hierarchies through literal translation of French idioms seems... out of place?
If the substance of someone's argument is sound, does it matter whether they
use your first name or not?
In English professional discourse, addressing someone by their first name is
not presumptuous—it's normal. Insisting on "Mr. Thomas" feels oddly formal.
I raise this because you've repeatedly objected to anonymous commenters,
pseudonymous commenters, and people using your first name. This suggests you're
more concerned with social hierarchy than with the substance of arguments.
Which brings me back to: does the validity of an argument depend on the
speaker's identity and status? Or should arguments be evaluated on their merits?
Because if the former, you're doing French academic hierarchy. If the latter,
why do you care so much about names and formality?
@ The FRANDalorian-11.02.2026 at 22.39.00 GMT
DeleteWho do you think you are to come up with such a comment? In French I would say you have really missed a good opportunity to keep quiet.
I am fully aware that IPKat is an English Forum. I do consider IPKat a forum based in the UK, but open to non-UK citizens and non-native English speakers. As being an English forum open to foreigners, I did not see in the rules for commenting that English etiquette should apply and everybody should be called by his first name.
In an open discussion, knowing who is my vis-à-vis, I have no problem to be called by my first name as I then know with whom I have to do. I have the courage of my opinions and act in the open and do not hide behind a pseudo.
I have never requested to be called Mr Thomas, but for people not having the courage to act in the open, I do not tolerate what I consider undue familiarities. It is my good right to do so and in an open English professional discourse, I would not bat an eyelid about being addressed by my first name. Here the discourse is not open and this makes a hell of a difference.
You are thus totally mistaken when you think that I am more concerned with social hierarchy than with the substance of arguments. If I find arguments not convincing, I do it in the open, even in reply to glorious anonymous writers, and I will explain why I find the arguments not convincing.
If the author of a comment is so sure about his arguments, why does he need to hide his identity? I would like to obtain from you a compelling argumentation while it is absolutely necessary to hide behind a pseudo when bringing forward legal or technical arguments. I do not see any and this is why I am acting in the open.
I know very well some colloquial English expressions, as contrary to what you might think, I have strong family bonds with the UK and I regularly fly over to visit my English relatives. As I am polite, I will refrain to use the one which comes to my mind when I see the arrogance with which you act hiding behind a pseudo.
Je n'ai pas de leçons à recevoir de votre part.
I am rather amused about all the examples flying around.
ReplyDeleteIf the claim is limited to “up”, but the description states “up and down”, what hinders the applicant/proprietor to write in his claim “up and down”, provided the same function is achieved with “up and down”. If the same function is not achieved with “down”, then “down” has to be deleted, as it introduces doubts about the scope of protection. I would even say that Art 83 is not fulfilled, when the function “down” is not disclosed and is not apparent from common general knowledge. When the disclosure explains that the same function is achieved "up" and "down", then limiting the claim to "up" makes no sense, other than trying to catch alleged infringers which use the "down" function and have only read the claim. G 1/24 will even bring good order in such a mess. This is also a form of adaptation of the description to the claims.
The same goes with the ratchet example. A pawl engaging a toothed wheel is a ratchet acting in rotation. Two tubes, telescoped, one inside the other, which can only be moved in one direction is a ratchet acting in translation. The problem is that in Max Drei’s example, a ratchet acting in rotation is compared with a ratchet acting in translation. The problem is that both are not compatible. If you bring both in the same disclosure and claims, you will earn an objection of lack of unity.
In one of the Amicus Curiae and in one contribution, the author explains that the claim is about a screw. claims a screw. The description mentions a rivet. According to the author, it would allow to cover equivalents and would also represent an infringement according to the German Federal Court. The problem is that a screw is a releasably means for fixing two pieces together, but a rivet is anything but a releasably means for fixing two pieces together. The rivet can only be removed by destroying it. We are thus far from equivalents…..
If you bring in examples, please insure some technical coherence.
By the way, I have published my own analysis of the Amicus Curiae. I did not use IA, as IA might give an answer, but one does not know on what data the answer is based and on what it has been trained.
Maybe, DXThomas, you need to update your blog to include Roche's contribution to the amicus brief. It may help with your stats.
DeleteLook at my blog. Max Drei drew my attention to my mishap.
DeleteThe blog has been completed by now.
People arguing in G 1/25 that description amendments have “no legal basis” are technically right, but that point is not as strong as it sounds. A lot of central EPO practice does not come directly from the EPC text either. The problem–solution approach is often linked to Rule 42(1)(c) EPC because it mentions stating the technical problem and solution, but even then this link is a stretch. The gold‑standard “direct and unambiguous” test for added matter comes from the Enlarged Board, not the EPC wording. The strict sufficiency approach also comes from case law rather than from Article 83 itself. The same pattern can be seen with the recent priority entitlement issue, which needed the Enlarged Board in G 1/22 and G 2/22 to resolve a long‑standing strict practice that was never clearly based on the EPC text.
ReplyDeleteSome say that “users do not like description amendments”, but that view comes mostly from a small group of patent attorneys. Real users include many more people, such as examiners who need clarity, competitors trying to understand what a patent covers, courts that interpret the claims, and the wider public who rely on accurate disclosures. Their needs are broader than the preferences of attorneys.
I like description alignment as a public policy tool because it makes the patent clearer for everyone. I accept the legal basis is thin, but that is not the fatal hurdle people make it out to be. The EPC is broad in places and the Enlarged Board has always stepped in to fill gaps.
On the other hand, references to "public policy" (or apparent "teleological" interpretations, as referenced in the AI summary) have one fundamental weakness: a complete absence of any hint in the TP - or other contemporaneous documents - of any kind of reference to the alleged policy or purpose.
DeleteThe EPC may be broad in places but prior interpretations of the EPC have at least been based upon something tangible - and not just invented in retrospect to suit a perceived need / desire of the EPO.
Thanks to Diana Taverner for a contribution I can subscribe to every word of it.
DeleteAs a further example of what the EBA has achieved are Swiss-type claims.
Without Swiss-type claims, no protection for second and further medical indications.
They are now obsolete as the EPC has been amended in 2000, but it needed the EBA to find a solution.
Well, looking at the actual amicus curiae briefs, the No comes not from a small group of patent attorneys. They mostly come from the applicants, i.e. real users of the system. It is actually mostly patent attorneys and the office that push for a yes regarding adaptation of the description.
DeleteAnd no, it does not necessarily make the a patent clearer, because by focusing on these inconsistencies, examiners have no time to really go into checking clarity of claims in light of the description. You will never ever get a "clear" patent, because, in particular for phamra applications, reaching this levels of perfection would amount to spending more time on description amendment than on the claims. And this for a useless task, because such a level of perfection will never be reached. This, imho, is rather like the punishment of Sisyphos on patent examiners, which is unnecessary, because courts can deal with it and harmonization should be up to the courts (and the national judges actually speak to each other and read the decision of others).
Regarding gaps in the law: maybe read the Roche amicus curiae at page 4 regarding G2/02. The EPO goes beyond filing gaps. They did not provide a test like the problem-solution approach for deciding on what is inventive (which btw is not binding! just as an example T 465/92 so it is not like requiring amending the description) or any of the test for added matter (which are, again, non-binding!, see G2/10), but the EPO refuses patent applications for not following their made up law.
Thanks for your reply, Proof of the pudding. I may be missing something, as I haven’t checked the TP in detail. If there are passages showing where the added-matter doctrine or sufficiency standards were actually derived, I’d be grateful to see them and happy to concede that as a difference.
DeleteCOMVIK just occurred to me. The brief “as such” wording in Art. 52 led to a long and complex line of case law, requiring two Enlarged Board decisions before the current framework settled, which seems another example of judicial development rather than a clear TP basis. My aim isn’t to take sides but to keep the discussion even-handed.
Thanks for explaining your view BF. I take the point that applicants feature strongly in the amicus briefs, even if my own experience of the debate, including on LinkedIn, has been a bit different. Without wider data it’s hard to be certain either way.
DeleteOn the legal side, most EPC requirements can’t really be applied without tests developed through case law. If those tests were treated as truly non-binding, the EPO would struggle to refuse applications at all, which doesn’t reflect how the system has worked in practice.
That doesn’t mean the current approach to description alignment is perfect, but improving consistency between claims and description can still be seen as a legitimate goal to work towards, and clearer patents are generally a good thing.
Diana, added matter and sufficiency are clear and unambiguous requirements of the EPC. What you are describing with respect to (EBA) case law is therefore not the creation of a legal requirement but instead the development of a test that can be used to assess compliance with that unarguable requirement.
DeleteBy way of contrast, what is at stake in G 1/25 is whether there is even any description adaptation requirement in the EPC at all. Perhaps the closest that the EBA has come in recent years to settling a debate on whether the EPC includes a specific requirement was in G 4/19 ("Double patenting"). In that case, the EBA assessed sources relevant under the VCLT and concluded as follows:
"the competent legislator, here the Diplomatic Conference, established that the prohibition on double patenting was a generally recognised principle as a question of fact, and in addition made it clear that this was a principle falling under Article 125 EPC as a matter of interpretation of the law, the Office was thereby not only empowered to apply this principle but effectively also duty-bound to do so".
Analysis of the sources cited in the pro-amendment amicus briefs reveals a complete absence of any that are both (a) relevant under the VCLT and (b) point to a "generally recognised principle" which supports adaptation of the description.
In view of these conclusions (as well as the interpretations of EPC provisions in the anti-adaptation case law and amicus briefs), it is pretty difficult to see where a legal basis for a purported description adaptation requirement could possibly be found.
Of course, if you believe that I missed anything relevant, I would be happy to hear what that is.
Diana I welcome your efforts to keep the discussion "even-handed". Of course the EPO, in pursuance of its duty to the public interest, strives to eliminate all inconsistencies. Fair enough. However, I am one of those drafting and prosecuting attorneys that can immediately relate to the well-rehearsed, widely felt and vigorously pushed arguments against insistence upon strict "conformity". The Amicus brief from CIPA alludes to a middle way and my instinct tells me that such a path is the one to pursue. But I myself lack the insight to see how to formulate this middle way in words that express it clearly. Instead, I live in hope of a miracle from the EBA.
DeleteFor the EPO to blithely insist of Applicants that they conform a mechanical engineering case to a claim narrowed in prosecution by deleting embodiments or declaring them to be "not" the invention is what upsets me. Whether the EPO likes it or not, Europe has a Doctrine of Equivalents. Acts outside the literal wording of the claim routinely infringe. So courts have to decide routinely what is or is not an "equivalent". What if the problematic embodiment that has to be deleted or labelled as "not" the invention is, in fact, an equivalent that, if coming from a competitor, would be an infringement? Is it fair to the inventor for the EPO to refuse to grant until the Applicant labels it as "not" the invention? My feeling is that the issue what is equivalent can best be addressed after consideration of what the inventor declared to be thier invention in their patent application in their text as originally filed. What do you say?
Thanks, Proof of the pudding, for your excellent points. I agree that added matter and sufficiency are clearly grounded in the EPC, with case law providing tests to assess compliance. In practice, the line between developing a test and creating a new requirement can be subtle.
DeleteThe question is whether Article 84 can be used to shape the description. Applications are often drafted with inventive step in mind, so the description is pre‑adapted to show how the claimed solution addresses the technical problem. COMVIK-type inventions are adapted to avoid excluded subject matter, Article 83 ensures enough detail is included to support the claims, and post-SCOTOS practice provides another example of a major patent office subtly prompting description adjustments to head off objections under Article 101. Much of this adaptation happens automatically, so we barely notice it when drafting.
Even G 4/19 shows that what looks like a simple legal point can require careful reasoning and substantial work to reach a final outcome.
Thanks, Max, I agree that seeking a middle ground makes sense, and I fully acknowledge the bind that equivalents can create for applicants. Cases like Ensygnia/Shell in the UK show why compromise is needed: the proprietor’s post-grant description amendments expanded scope and created real uncertainty for third parties. Description can thus be used unfairly, while it can also be used fairly to determine scope for equivalents.
DeleteA possible compromise could build on the existing strict literal compliance check, which currently applies in equivalents only for obviousness. If an applicant has limited the patent to a particular equivalent used in a specific combination to ensure validity for added matter reasons, they cannot later rely on other equivalents. This is similar in spirit to a Formstein-type safeguard: just as that defence prevents patentees from claiming embodiments that would be obvious over prior art, this approach prevents patentees from overreaching by later claiming equivalents they effectively ceded during prosecution.
This balances protecting inventors while avoiding unfair surprises for third parties.
It could be that I am the last to realise it, but I am now convinced that the contributor sailing under the flag "Diana Taverner" is an AI.
DeleteCoincidentally (or perhaps not) my UK newspaper carried an opinion piece yesterday lamenting the ever-increasing number of utterances of the words "thank you". My experience of AI's is that they always begin with a "thank you, Max". And sometimes go on to compliment me on the excellence of my comment. Thank you, Diana, for your courteous demeanour. As the Americans would say, appreciated.
After wading through all that though, these AI's can be useful, sometimes. They might get their points jumbled up but, often, they mention a case, or a legal provision, that one otherwise might have overlooked. Or they might reveal how one's point can be misunderstood. What say you, Pudding?
Hi Max. I can assure you I am not an AI. I am aiming for a polite good faith debate hence my use of thank you. Unfortunately, in my experience civil debate in our profession is under threat. I’ve been at oral proceedings where professional representatives have scoffed and tutted during oral submissions, not to mention the ad hominem attacks, something which I am keen to avoid. I can only imagine how this looks to OD/BoA but I suspect this might be part of win at all costs mentality by trying to get under your colleague’s skin.
DeleteThe views/points above are my own and my overt politeness might have gone too much the other way.
Nah.
DeleteTwo thoughts. Pudding's reply convinces me that he is not an AI, Diana's not Second, up to now in my life I have not suffered the unpleasant experience of having a bucket of slop poured over my head. But, sorry Diana, I have an uneasy feeling that it's happening now.
Delete@ Diana Taverner,
DeleteIt would never have come to my mind that you were a mere IA. What a strange idea.
I can however confirm that some representatives can become very aggressive, not only against the deciding body, but even more against the representative of the other party. I witnessed one representative literally insulting the other ones and disputing their professional capability. In view of the situation, I requested the badly behaving representative to express apologies towards his counterparts. He bluntly refused.
In view of this behaviour I informed the representative that the OP will end immediately and that we will reconvene later, when he had calmed down. This threat was enough to calm him down, as he knew very well that he would have to explain to his client why the latter would have to pay the costs for a second OP. The apology was very formal, but the trick worked. As he lost the case, he stormed out of the room without saying goodbye. He was customary of this. He only said goodbye when he won the case. Needless to say that he was well known for his temper in first instance and in appeal.
I also wondered whether Diana Taverner was an AI, or at least had AI assistance in formulating her posts. It is something that is difficult to put your finger on but there are certain turns of phrase that "feel" AI-like.
DeleteThe fact that we increasingly have trouble differentiating between humans and AIs is in itself very interesting. The Turing Test is well and truly dead.
PS - I realise this is purely a linguistic quirk, and I intend no criticism, but in English it is AI, not IA.
Hello Mouse. On the issue of AI assistance, I had occasion last week to ask Bing which actress played the role of mother to Beatrix Potter in the film biography. Amusingly, and perplexingly, first it answered with the name of the actress who played BP herself, and, then, on further prompting, at the second attempt, BR's BF.
DeleteSo I asked Co-Pilot why Bing kept on giving the wrong answer. Co-Pilot's reply was deeply informative. And it also gave me the correct answer to my question.
Increasingly, people wanting an answer to a question turn to an AI to answer it. No wonder that several humungeous new data processing facilities are being built, here in Germany. They are going to be needed, to handle all those increasingly complicated and sophisticated questions.
@Diana There is a reason why tests are non-binding, because they can fail, aka give a wrong positive/negative answer. And it is good that they are non-binding, because in these cases the EPO is still able to act, just read the case I cited for the PS approach or have a look here https://www.epo.org/en/legal/case-law/2025/clr_i_d_2_2.html . The same is true for added matte. Just think about the novelty test: https://www.epo.org/en/legal/case-law/2025/clr_ii_e_1_3_7.html
DeleteSo there is not a subtle distinction between requirements of the law/EPC or what a test is. A test will help give guidance if the requirements of the law are fulfilled, but it cannot substitute the assessment if this requirement is 100% fulfilled.
Yes, clearer patents are better, but a focus on real issues for clarity instead of unnecessary amendments that do not even lead to clarity: subject-matter disowned for formal reasons are not taken into account for assessing equivalence, see Pemetrexed judgement from the German Federal supreme court. So no, the EPO requirement to delete subject matter on formal grounds (Art. 84) has no impact on the DoE in some countries like Germany or Italy, while it might have in others. So great, the goal of harmonization between EPC countries cannot be achieved by the EPOs deletion requirements.
And there is a reason why patent attorneys are more out spoken in public: They have nothing or less to fear whether their personal opinion reflects on their company as an applicant, i.e. how the offices treat their application, and advertisement.
Small point Daniel, on my ratchet hypo. It comes from real life. A hand unit for a self expanding stent pulls proximally a tube surrounding the stent, relative to a tube that extends through the lumen of the stent. The invention was to include a ratchet in the hand unit, to prevent any residual distal movement of the outer. tube during release of the stent. So your distinction between rotation and translation for the movement of the ratchet components leaves me unpersuaded that the ratchet claim lacks unity.
ReplyDelete@ Anonymous-10.02.2026 at 15.45.00 GMT
DeletePlease abstain from familiarities. If you give your name, I might accept being called Daniel, but not from somebody hiding.
The example you give has nothing to do with Max Drei’s example. Max Drei put next to each other a classical ratchet like on a bicycle and a ratchet in a curtain supporting tube.
Your example is totally different. One way to deploy a stent is either to fix it on a balloon (not self-expanding stent) or to extract it from a sheath (self-expanding stent). If the stent at the distal end is pulled out from a tube, or a support tube is pulled out from the stent at the distal end, I can perfectly imagine that proximally a rotational ratchet mechanism combined with a thread transforms the rotation at the proximal end in a translation at the distal end. As you explain it, it does not associate a rotational ratchet with a translational ratchet as envisaged by Max Drei. You are actually comparing apples with pears. For your information, I have been dealing with stents for over 10 years and chaired numerous OP in opposition in this domain.
I can thus only repeat, come up with coherent examples.
Some misunderstanding here, DXT. My fault. Sorry.
DeleteOn the road when I posted my "Small point, Daniel" comment, I failed to insert my pseudonym before clicking on "send". I had thought that it would be obvious to everybody that the "Small point" comment was coming from MaxDrei.
As to the hand unit I had in mind, it is for a self-expanding percutaneous trans-luminal stent. You know, one which is released 130 cm away from the hand unit, by pulling back proximally the sleeve that surrounds it, while restraining from any proximal movement the inner of the two sleeves arranged telescopically. The inner tube extends along the lumen of the stent and all the way back to the hand unit, to stop the stent being carried proximally by the retreating outer sleeve, so that the stent deploys as desired, progressively, that deployment starting at the distal end of the stent.
One hand unit has within it a toothed wheel and pawl. Around the axle is a thread attached to the outer tube, to pull the tube proximally. But another way to achieve a ratcheting effect would be to place the retreating sleeve, within the hand unit, inside another tube that grips the radially outer surface of the retreating sleeve, to prevent any distal "spring-back" of the stent releasing sleeve, whenever the surgeon pauses the deployment of the stent. For me, those two ratcheting systems are "equivalent" on a "function/way/ result" enquiry. But: what if I narrow claim 1 to recite "ratchet" while the application as filed explicitly names the wheel/pawl as a ratchet but fails to recite the word "ratchet" in the other of the two illustrated embodiments? Must I delete the second embodiment or declare it to be not the "invention". That was my concern.
As to the Diana Taverner pseudonym, I have no doubt that a real person is posting under that name. My doubt is whether it is a real person who is composing the text that she is posting. Thus, assurances from Diana Taverner that she is a real person doing the posting do nothing to assuage my doubts about who (or what) is composing her text.
Isn't it immediately obvious from the context and from the phrasing above that our anonymous friend behind the post of 10 February at 15:45 is MaxDrei, who has simply failed to identify himself correctly?
DeleteIt is clear that the poster is talking about MaxDrei's example and with inside knowledge.
Your belligerent reaction to "anonymous" commenters is itself very interesting. Your reflexive distrust of pseudonymous posters is very well known from comments both here and on your own blog including today where you apparently deleted anonymous comments and announced an intention to do the same to anyone identifying themselves with a pseudonym. This is frankly bizarre behaviour and it is unbecoming of someone who leans so often on their credentials as a source of authority. Address the substance of the arguments, not their source.
@DXThomas
ReplyDeleteWhen you say that « the EPO has to deliver a patent which should not allow the proprietor to interpret its patent as he thinks fit in post-grant procedures » , you are not quite in line with your comments of 1 Sept and 4 Sept 2023 (which I came across by chance yesterday) on the IPKat thread relating to T 558/20. I quote :
» What the proprietor does with its patent is none of the business of the EPO ».
« The EPO is there to grant patents if the conditions set up by the EPC are met. The rest is indeed irrelevant and has nothing to do with EPO’s self-interests «.
I agree with these robust statements, which are all the more justified as patents involved in court proceedings are a minuscule percentage of granted patents. Issues specific to court proceedings have therefore no reason to significantly influence examination at the EPO. I do not understand why you changed your mind but it is of course your right.
I agree with you that it is for the BOAs and the EBA to say what the law is but If stakeholders are formally solicited for amicus curiae submissions, I can suppose the submissions are not worthless and will be given due consideration by the EBA.
I also point out that the authors of the submissions are not only « representatives » or their associations. Industry associations have also filed submissions. I am myself a professional in corporate practice.
@Mr Hagel, 11.02.2026-09.20.00 GMT
DeleteWhen entering T 558/20 in the search mask of IPKat, I get the following results:
- COCO CHANEL’s independent distinctive role.
- Mr Justice Arnold finds American Science's mobile X-ray scanner patent valid.
No trace of T 558/20.
If you think it through, it is true that “What the proprietor does with its patent is none of the business of the EPO”. He can hang the patent certificate in his lounge or attack an alleged infringer with it. He is not even obliged to pay annual fees. The EPO has no say in the matter.
It I also true that “The EPO is there to grant patents if the conditions set up by the EPC are met. The rest is indeed irrelevant and has nothing to do with EPO’s self-interests”. Or should the EPO grant patents which do not meet the conditions set up by the EPC? This happens, alas, quite regularly, as after appeal in opposition 40% of the patents are revoked, 35% are maintained in amended form and only 15% of patents come out of opposition not maimed.
Contrary to what you imply, I have never changed my mind. My opinion has always been, and still is, that the patent delivered by the EPO should not allow the proprietor to change the interpretation of his patent as he thinks fit. In order to achieve this aim, it is the EPO’s duty to insure that the proprietor cannot have his cake and eat it.
Board’s decisions following G 1/24 have already paved the way in this direction. I hope that the next step will be achieved with G 1/25. For example, it is not acceptable to leave as optional in the description features which have been incorporated in an independent claim.
Requiring alignment between claims and description I one of the duties of the EPO. Some courts might spot this inconsistency and ignore it, some might use it against a potential infringer. Look at the comments at the end of my analysis of the Amicus Curiae for G 1/25.
If patents involved in court proceedings are a minuscule percentage of granted patents, why then bother with carrying out a documentary search followed by a substantive examination? Why then not revert back to the old French system valid between 1844 and 1968? In this system, only formal aspects were checked. This allowed a well-known, late Franch litigation lawyer, to say that the proprietor discovers his patent on the day of the infringement. I am sure you have heard about Me Mathély.
The reason for this is rather simple. When an examiner gets an application on his desk, he is never sure whether this patent will be the pearl in the oyster, or will simply end up in the Bermuda triangle of patents never having made it to the marketplace. As young search examiner in the early 70ies, I got on my desk an application from a private inventor: Mr Ray Milton Dolby. Do I need to say more? To my knowledge, Mr Dolby only gave licenses. To me, this was an eye opener.
National legislators, even the French one, and European legislators, found that this is not a good situation. This is why a documentary search followed by substantive examination have been instituted. This is to give patents a presumption of validity.
That "issues specific to court proceedings have therefore no reason to significantly influence examination at the EPO" is a mere fallacy. The examiner does not know in advance whether the patent will end up in court or not.
You are certainly right to suppose that the submissions brought to the EBA will be given due consideration by the EBA. It might however not be because a majority of the submissions are against adaptation of the description that the EBA will take this as criterion to decide against adaptation. It is not a question of volume or weight, but a question of law.
It is not worth saying much about your last point. In my analysis of the Amicus Curiae in my blog, all submissions were looked at, whatever their origin could have been. I hope you can give me credit for this.
Here you go: https://ipkitten.blogspot.com/2023/08/make-no-bones-about-it-credibility-test.html
DeleteQuick question: what might be the impact of the EBA's ruling on the practice of national patent offices (and courts) and the UPC?
ReplyDeletePrevious EBA interpretations of the EPC have had downstream consequences for courts and patent offices. So might we expect effects here?
Continuing the AI theme, I ran OCR on all of the amicus curiae briefs and on the President's comments and the parties' submissions. I then uploaded them to an AI (Claude Opus 4.6 in extended reasoning mode) and asked it for its view of the most persuasive arguments either side. With no further prompting from me this is what it generated:
ReplyDeleteA. Most Persuasive Arguments Against Mandatory Adaptation
1. The "no legal basis" argument (T 56/21 reasoning, widely adopted)
This is the single most powerful argument against adaptation. Article 84 EPC says: "The claims shall define the matter for which protection is sought. They shall be clear and concise and be supported by the description." The "support" requirement, on its natural reading, means the claims must find support in the description — i.e., what is claimed must be disclosed. It does not say the description must be purged of anything that goes beyond the claims. The text imposes an obligation on the claims, not on the description.
This is persuasive because it starts from the statutory text and respects the principle that restrictions on parties must have a clear legal basis. No EPC article says "the description shall be consistent with the claims" or "the description shall be adapted to amended claims." The absence of such language is telling, particularly when the EPC is otherwise meticulous about specifying what is required.
2. The G 1/24 "solves the problem" argument (Roche, AIPLA, Canon, proprietor)
G 1/24 held that the description and drawings must always be consulted to interpret the claims, even where the claims are clear. If courts and the EPO must already look at the description to interpret claims, and if (as G 1/24 also confirms) the claims have primacy in defining the scope of protection, then a skilled reader encountering a description that discloses broader subject matter than the claims simply knows that the broader disclosure does not define the scope of protection. The inconsistency is visible and manageable; it does not need to be surgically removed.
This argument is persuasive because it shows that G 1/24 itself provides the interpretive toolkit that adaptation was supposedly designed to supply. The argument that adaptation is needed to prevent post-grant confusion has been undermined by the very decision the President relies on.
3. The Article 123(2) risk argument (AIPLA, BIA, Canon, EFPIA, Roche, epi)
Forcing adaptation creates a genuine risk that otherwise valid patents will be invalidated because an adaptation introduced added matter or removed disclosure necessary for sufficiency. Several amici cite concrete examples where deletion of "unclaimed embodiments" later proved problematic when the patent was enforced or when further amendments were needed. This argument is persuasive because it identifies a real and practical harm flowing directly from the mandatory adaptation requirement — it turns a formal exercise into a substantive trap.
4. The "EPO is not competent to decide scope of protection" argument (ROCKWOOL, Bardehle Pagenberg, Ericsson)
Perhaps the most elegant argument against adaptation comes, paradoxically, from the opponent. ROCKWOOL observes that the EPO is not competent to consider infringement. Adapting the description necessarily involves the EPO making judgments about what falls inside or outside the scope of the claims — judgments that properly belong to infringement courts operating under Article 69 EPC. When the EPO deletes or disclaims an embodiment from the description, it is implicitly ruling on scope of protection, which is not its function. This argument is compelling because it identifies a structural mismatch: the EPO is being asked to do something that falls outside its institutional competence.
@ Odo Ital-11.02.2026 at 13.15.00 GMT
DeleteI read with interest the arguments against adaptation as summarised by your IA. I am not criticising the IA as such, but some comments are necessary.
For obvious reasons, your IA has manifestly not looked at any board’s decisions following G 1/24. It was satisfied with what was written in the submissions.
All decisions following G 1/24 are abundantly clear: if the description contains statements allowing a broader or a more limited interpretation than that derivable by the skilled person from the claim wording, in order for this interpretation to be taken into account, the claim has to be amended accordingly. If this is not surgically removing the visible inconsistency, what else is this then? The inconsistency might be manageable, but not in the way thought by those in favour of this application of G 1/24. It has a direct impact on the prior art to be cited against the claim. This does not mean that the inconsistency should be left as it is.
To the contrary, this argument is NOT persuasive. As such, G 1/24 does not provide the interpretive toolkit that adaptation was supposedly designed to supply. The argument that adaptation is needed to prevent post-grant confusion has by no way undermined by the very decision the President relies on. The contrary is true.
To be honest, the corresponding submission surprised me, especially coming from a very reputable patent attorney firm.
I am still waiting to see a patent having been revoked due to the adaptation of the description. Carried out correctly, no risk exists, be it for added matter or for sufficiency. If the applicant does not agree with the decision to grant for this reason, he can always appeal.
It can be agreed that the "EPO is not competent to decide scope of protection". The EPO is however competent to decide on the validity of the patent at grant or after opposition. It is also the duty of the EPO is to insure that the proprietor cannot change the interpretation of the claim as he thinks fit. For this reason, The adaptation exercise is by no way formal and is an absolute necessity.
With respect, Mr Thomas, you are moving the goalposts.
DeleteG 1/24 and its progeny were not within the scope of what the AI was asked to review. As stated above. But it *did* review the characterisation of that decision and its progeny in the submissions from both sides of the debate. Both the pro- and anti- camps have relied on G 1/24. That gives a balanced picture, does it not? Bear in mind here that the EBA is being asked to rule on the law itself and not on how it has been applied, so G 1/24 carries more weight than any line of case law deriving from it.
Don't you also see an inherent contradiction between your agreement that the "EPO is not competent to decide scope of protection" but also at the same time that it has a duty "to insure that the proprietor cannot change the interpretation of the claim as he thinks fit"? Let's face reality. Patent proprietors and their opponents will virtually always dispute the interpretation of a claim if it is litigated. Doesn't this give the game away: your real concern is about defining the boundaries within which interpretation can happen? If so, how is that not intruding on the competence of the infringement courts - and ultra vires for the EPO?
@ Odo Ital-11.02.2026 – 18.52.00 GMT
DeleteI am fully aware that I have moved the goal posts. It is the way you first used AI to merely analyse comments on the present blog which lead to my reaction. What you should have done is to have your AI look at all the amicus curiae and then publish the conclusions on this platform. You did so, but too late.
When reading some of the amicus briefs claiming that G 1/24 was against adaptation of the description I was rather surprised. When in application of G 1/24 the boards require applicants/proprietors to amend the claim by bringing in it the interpretation found in the description in order for the latter to be taken into account, is this not a first step of adapting the claim to the description? To me both moves equivalent like Art 84 acts in both directions. G 1/24 has never said that the interpretation in the description should take precedence on that the skilled person gains from the wording of the claim, although this was the hope of many.
I fail to see any contradiction between my argument (not agreement) that the "EPO is not competent to decide scope of protection" but also at the same time that it has a duty "to insure that the proprietor cannot change the interpretation of the claim as he thinks fit"?
I have explained on this platform and in my blog why the EPO is necessarily an outlier in matters of adaptation of the description. The EPO is not part of an integrated national system associating a national granting authority and a national judicial system. In such an integrated system, a national court can iron out, or not, any inconsistencies left in the description by the granting authority. It is this kind of inconsistencies which allow a proprietor to change the interpretation of the claim and hence nurture what is known under the term Angora cat.
This is not possible for the EPO since its competence is limited to decide upon the validity at grant or after an opposition. This is due to historical reasons which are easily forgotten in those days. Have a look at how it came to the EPC. The original European patent system was a fully integrated system within the EU, like any national sytem, but it never came through.
When the EPO requires removal of inconsistencies, it does not hinder national courts or the UPC to decide on direct, indirect infringement or infringement by equivalents.
I thus fail to see any intrusion on the competence of the infringement courts. The EPO does not act ultra vires, it simply does what it has been asked to do.
"When the EPO requires removal of inconsistencies, it does not hinder national courts or the UPC to decide on direct, indirect infringement or infringement by equivalents".
DeleteStrictly speaking, this statement is only true if one takes the amended (ie adapted) patent specification as the document to be interpreted by the national courts or the UPC. However, it is most definitely not true if one instead considers the different conclusions that the national courts or the UPC might reach if asked to interpret the claims in the light of the description of the application as filed.
Indeed, it is only if your statement is not true in such circumstances that it is possible to discern any practical purpose to the EPO's description adaptation practice: if adaptation makes no difference to post-grant claim interpretation, then why bother?
Thus, whilst you may fail to see the intrusion on the competence of the national courts (to interpret the claims as granted, whether for infringement or validity), it is as plain as day to those of us who choose not to blithely ignore it.
@ Proof of the pudding-12.02.2026-14.40.00 GMT
DeleteIf I understand you well, examination of an application is just useless as national courts or the UPC should always look at the application as filed.
In other words, does the interpretation of the claims depend on the file history? The CFI-UPC Munich LD tried to do so in the “Hanshow” case. It was promptly call back by the CoA UPC.
I said it before, the day the last part of Art 84 will de removed from the EPC and file wrapper estoppel introduced, then I will agree that the description will not have to be adapted to the claims, as it is the case in the US. We are far from there.
If the description and the claims are aligned, it will avoid a proprietor to change the interpretation of the claims as he thinks fit. For the surplus, G 1/24 has not said, that the interpretation found in the description should take precedence over the interpretation of the claims which results from a skilled person reading the plain claim wording.
Are you therefore insinuating that boards intrude on the competence of the national courts or of the UPC when, after G 1/24, they decide that an interpretation found in the description will only be taken into account if the claim is amended accordingly? This is to me the first step of the adaptation of the description to the claims or vice-versa. What is it wrong with it, other than it does not fit with your views?
Should the EBA decide that adaptation of the description is necessary, do you then conclude that the EBA intrudes on the competence of the national courts or of the UPC? Give the EBA a bit more credit.
Not all decisions of the EBA are an example of well argued decisions, but I am confident that in the present case, the EBA will come to a result which will enhance harmonisation with national courts and the UPC, well knowing that the EPC system is not comparable to any national system.
It is plain as a day to me that you would rather blithely ignore the consequences of G 1/24 since G 1/24 has not gone in the direction you wished, but this is the reality. You are not deciding, the boards and the EBA are.
OK, so you either cannot or will not see why it is that the EPO's description adaptation practice intrudes on the competence of the downstream courts.
DeleteTo be absolutely clear, I am not insinuating that there is any intrusion on the competence of the downstream courts in situations where the EPO requires amendment of a claim to accord with a definition found only in the description. This can be necessary to ensure that the claim clearly reflects the sole interpretation of the claim wording that is directly and unambiguously derivable from the disclosures of the application as originally filed.
The fact that you struggle to understand the difference between that situation and adapting the description to the claims speaks volumes. That is, it demonstrates that, in your mind, the claims and the description of an application as originally filed can disclose different (and incompatible) meanings for the same terms / words. But how can that possibly be so, especially if the claims merely use the terms / words in question whereas the description explicitly defines them? Is the so-called uniform concept of disclosure not relevant to settling this question?
Also, if there truly are two different and incompatible definitions for the same terms / words, which of those definitions is clearly and unambiguously disclosed in the application as filed? Is it the “definition” in the claims? Or the definition in the description? It certainly cannot be both.
I have to admit a certain level of amusement at the audacity of your accusation that G 1/24 did not go in the direction that I wished – when my beef is very clearly not with G 1/24 itself but only (certain) subsequent BoA interpretations of that decision. Still, I suspect that I could repeat that point until I am blue in the face and you still would not concede it.
On the topic of points that you struggle to concede despite ample contrary evidence, I would like to draw your attention to two (more) of your statements which appear to contradict one another:
29 Jan 2026: “The EPO’s description adaptation practice really does not seek to achieve a poor man’s file wrapper estoppel”.
12 Feb 2026: “the day the last part of Art 84 will de removed from the EPC and file wrapper estoppel introduced, then I will agree that the description will not have to be adapted to the claims”.
To me, your 12 Feb comment makes it clear that you believe that current EPO practice based upon Art 84 EPC serves a similar purpose to file wrapper estoppel. So do you still stand by your assertion from 29 Jan 2026, or will you now concede the point that I made?
@ Proof of the pudding-13.02.2026-10.19.00 GMT
DeleteI have never alleged that the claims and the description of an application as originally filed can disclose different (and incompatible) meanings for the same terms / words.
The so-called uniform concept of disclosure does simply mean that once the claims have been limited with respect of the claims as filed, the independent claim might not be in adequation with what is left in the description or vice-versa.
If a claim is limited by incorporation of features of a dependent claim, or from the description, it cannot be left in the description a statement to the effect that the feature added in the independent claim is still optional. For the same token any embodiments in the original description of the invention claimed at filing, cannot be left as embodiments of the claimed invention at grant, when those embodiments do not fall under the claims as granted.
You might believe whatever you want, but I never said that I believe that the current EPO practice based upon Art 84 EPC serves a similar purpose to file wrapper estoppel.
Reading the sentence with a mind willing to understand, it simply means once a file wrapper estoppel is introduced in the EPC, the description need not any longer be adapted to the claim. There is no more to be interpreted in this statement of mine.
Whether you get blue in the face or not is the least of my concerns. If the interpretation in the description is brought in the in the independent claim as required by all decisions following G 1/24, then, at least on this point, claim and description are adapted to each other. Or would you claim that they are not?
When reading all your statements, especially the one about the "unitary legal title", cf. your input of 11.02.2026 at 11:05:00 GMT, it is abundantly clear that you would like the interpretation in the description to be taken into account next to that of the claim. This is in clear contradiction with all decisions following G 1/24.
Well, I must take my hat off to you for your skills at avoiding answering direct questions and acknowledging obvious conclusions!
DeleteFor example, I asked about the subject matter that is disclosed by an application as filed. Your response: commence a discussion of a completely different topic (ie what happens once an application is amended). Bravo! I suspect that you may have missed your calling as a public relations manager or what is known in the UK as a spin doctor.
@DXThomas:
DeleteMaybe I am missing your point, but if (in your view) the need to adapt the description would evaporate if "a file wrapper estoppel is introduced in the EPC", then to me this implies that you see description adaptation having a file-wrapper-estoppel-like function. In other words, description adaptation is performing the role of file wrapper estoppel. If file wrapper estoppel is introduced, then description adaptation becomes redundant because its current function would be served by other means. That suggests that - contrary to what you say in your comment above - you do see description adaptation and file wrapper estoppel as equivalent to one another.
If this is not what you mean, please help me to understand.
For the surplus, have a look at the new Guidelines F-IV, 4.1 and 4.2 to enter into force on April 1st 2026.
DeleteThe board's application of the G 1/24 has made its way in the Guidelines.
And here's Claude's take on the other side of the debate:
ReplyDeleteB. Most Persuasive Arguments In Favour of Mandatory Adaptation
1. The "unitary legal title" / legal certainty argument (President, Patentanwaltskammer)
The strongest argument for adaptation is the practical reality that European patents take effect across 40+ jurisdictions with different interpretive traditions. Some courts look at prosecution history; many do not. Some apply file wrapper estoppel; others do not. A German infringement court that reads a patent and sees the description disclosing broad embodiments alongside narrow claims without any indication that those embodiments are not part of the claimed invention may arrive at a different interpretation than the EPO intended. The Schacht quote in the President's submission — noting that a significant portion of interpretive difficulties in infringement proceedings results from descriptions not being adapted to amended claims — is telling.
This argument is persuasive because it identifies the real-world downstream consequence of non-adaptation: legal uncertainty for third parties trying to assess freedom to operate.
2. The Article 84 sentence 1 argument (President, T 3097/19)
The often-overlooked first sentence of Article 84 states: "The claims shall define the matter for which protection is sought." If the description actively contradicts the claims — for instance by stating that a feature is "optional" when the claim makes it mandatory — then arguably the claims fail to clearly "define the matter for which protection is sought," because a reader consulting the description (as G 1/24 requires them to do) receives a contradictory message. This argument is persuasive as a textual matter: the first sentence is about the function of the claims, and a contradictory description interferes with that function.
3. The G 1/24 "makes adaptation MORE important" argument (President, D.X. Thomas, Bugnion)
This is the mirror image of argument A.2 above. The reasoning: if G 1/24 means the description must always be consulted — even when claims are clear — then an inconsistent description will always be read alongside the claims. Before G 1/24, one could argue that a clear claim would simply be interpreted on its own terms, making the description irrelevant. Post-G 1/24, that argument is no longer available. Therefore, inconsistencies in the description are more dangerous now, not less, because they will always be read.
I was particularly struck by the "unitary legal title" argument - just not for the reasons that the President et al. were aiming for. The reason that the argument caught my attention was that it completely contradicts the "primacy of the claims principle", as applied by the EPO.
DeleteEither the patent (application) is a "unitary legal title" or it is not. The EPO cannot have it both ways. If it is a "unitary legal title", then the claims should be interpreted accordingly (including by adopting the usual canons of interpretation, such as the same word(s) / term(s) having the same meaning throughout the whole document unless the context indicates otherwise).
Current EPO practice is based upon an inconsistent (and logically indefensible) approach, where:
- despite the ruling in G 1/24, the claims are typically interpreted in isolation; but
- a "unitary legal title" approach is then applied to justify a purported requirement to conform the description to claims that have been interpreted under a contradictory approach.
Whilst it might not be reasonable to expect perfection from the EPO, I think that the least that users deserve is for the EPO to be consistent with regard to how it interprets and applies the law.
@ Proof of the pudding-11.02.2026-14.12.00 GMT
DeleteI allow myself to think that under the notion of "unitary legal title" you see something totally different than for instance the president of the EPO.
In Point 57 of his submission, the president noted that “The EPO grants patents which may take effect in over 40 contracting, extension and validation states”. In this respect the patent is indeed unitary, as it is the same patent for all states in which it will take effect.
Under "unitary legal title" you see something totally different, i.e. that the claims and the description form a "unitary legal title". Your conclusion is then that the claims should always be interpreted in the light of the description, i.e. that interpretations found in the description might supersede the interpretation the skilled person derives from the plain wording of the claims.
When you look at all board’s decisions following G 1/24, it is not possible to allege that the claims are always read in isolation. The claims are read with the eye of a skilled person and are compared with interpretations found in the description again read by a skilled person.
All decisions following G 1/24 are abundantly clear: if the description contains statements allowing a broader or a more limited interpretation than that derivable by the skilled person from the claim wording, in order for this interpretation to be taken into account, the claim has to be amended accordingly.
In other words, there exists an inconsistency between the claims and the description. This inconsistency has a direct impact on the prior art to be cited against the claim. This does however not mean that the inconsistency should be left as it is.
Two possibilities
- the claim is limited by bringing in the latter the limiting definition found in the description
- the limited claim is broadened claim by bringing in the latter the broadening definition found in the description
In both cases the claims have not been interpreted under a contradictory approach. On the contrary the approach is as coherent as possible.
In both cases, eventually claims and description are aligned with each other.
The EPO and the BAs are consistent with regard to how they interpret and apply the law as decided by the EBA.
You would have liked the EBA to have said: what is in the description is to be interpreted next to the own possible interpretation of the claims.
The EBA simply said that the description has to be consulted, but not that the description overrides the claims. The boards have thus rightly concluded that the possible interpretation from the description has to figure in the claims in order to be taken into account. It could not be more coherent.
This is the more rightly so, in that post-grant jurisdictions have interpreted inconsistencies between claims and description in a contradictory manner, in other words, in an inconsistent way.
Allow yourself to think whatever you wish. I was referring specifically to para 50 of the President's submission, which includes the following statement:
Delete"Claims and description are part of one "unitary legal title"".
I can see that there is another meaning that could be conveyed by the phrase "unitary legal title". However, in my view, arguments based upon that alternative meaning are fundamentally flawed on the grounds that the EPO has no institutional authority to police post-grant interpretations by national courts.
As for your argument about post-G 1/24 case law, this just circles back to an issue that has been flogged to death on this thread. My response: if the description really does provide basis for an "inconsistent" interpretation of the claims, then how on earth could the EPO possibly conclude that the claims satisfy the requirements of Article 84 EPC?
The EPO might not want to engage in "interpretative somersaults" but it worries about national courts doing precisely that. Why so, if G 1/24 is being interpreted and applied so assiduously?
@ Proof of the pudding-12.02.2026-11.05.00 GMT
DeleteWhen the president of the EPO was speaking about a “unitary legal title”, reference was made to the fact that a granted European patent can be valid and/or have effects in a vast number of countries. With the amendment of Art 54 in the EPC 2000, there are not any longer different claims and/or descriptions for different EPC contracting states. this reinforces the unitary character of a granted European patent.
If this is not characteristic of an “unitary legal title”, it is difficult to understand what a unitary legal title could well be.
The EPO being restricted to grant or opposition after grant has a duty to deliver a “unitary legal title” which is applicable as it stands in a large number of countries.
I can well agree with you that the EPO has no institutional authority to police post-grant interpretations by national courts, but this is no hindrance to the EPO to request removing any inconsistencies between claims and description. This is the duty which has been conferred to it by its contracting states when it was set up in the early seventies.
As explained in a later post, Art 84 is there to insure that the scope of protection is as clear as possible and cannot be put into question by statements contained in the description.
By applying G 1/24 and Art 84 as they do, the EPO and the boards of appeal are in no way hindering national courts to apply Art 69 and the Protocol so as to decide on direct/indirect infringement or infringement by equivalents.
It has become abundantly clear that the EPO and the boards of appeal will not any longer tolerate major inconsistencies between independent claims and the description.
The way the EPO has been set up by its founding countries, which limited its competence to grant and opposition, plays a major role. For this reason, the EPO is not comparable to a national granting authority in which a national jurisdiction can iron out inconsistencies which have been left at grant.
The dichotomy between grant/opposition and post-grant or post-opposition procedures at national level or at the UPC is a fundamental aspect of the European patent system which is never taken into account. I have regularly commented upon this specificity, but never got a reply. For this reason, the EPO cannot be compared with a national grnting authority.
So are you saying that the President was using the phrase "unitary legal title" to simultaneously convey two different meanings, ie claims + description as a unitary document and EP patent as having some sort of "unitary" effect across EPC Member States? If so, I think that you may be missing the President's point, as it appears to me that he is arguing that the description should be adapted precisely because it is "unitary" with the claims (and so, in his view, should be fully consistent with the claims).
DeleteAs for your insistence, despite a mountain of contrary evidence and well-reasoned logic, that the EPO's description adaptation practice does not intrude upon the interpretative competence of the downstream courts, I would invite you to consider how a national court might assess the provisional rights conferred by a published application under Article 67 EPC.
Which description should the downstream court use to interpret the claims of the published application? My understanding is that it would have no choice but to use the "unadapted" description of the published application. If subsequent adaptation of the description has an effect on claim interpretation (as is clearly the EPO's intention), then how should the downstream court assess the infringement of a claim that has not been amended during prosecution, but whose interpretation differs between the published application and the granted patent?
Let's face it, if the intention is to "steer" the downstream courts towards what the EPO perceives is the "correct" interpretation of the claims, that clearly intrudes upon the interpretative competence of a downstream court which might otherwise disagree with the EPO's interpretation of the claims (and hence also the EPO's reasoning for requiring certain adaptations to the description). Whilst I do not doubt your ability to continue to deny this, it nevertheless remains an inescapable conclusion.
@ Proof of the pudding-20.02.2026-11.12.00 GMT
DeleteIn my opinion the president of the EPO has made clear that the EPO grants a title which is the same for all contracting states of the EPC, validation or extension states. In this sense it is a unitary document, valid with the same content in all those states.
Since the EPC 2000 there are not any longer different descriptions and claims for different EPC contracting states. You ignored this comment, but you came with a legal aspect which is barely applied.
I further note that I did never use the term unitary effect, as the jurisdictions in all those states can interpret the patent as they think fit, and for the EPC contracting states according to Art 69 and the protocol.
How many times an applicant has requested provisional protection under Art 67? Taking an exception to transform it in a rule is far-fetched. In this situation it is manifest that a downstream court has only the application as filed to deal with. An application can be sold to a new legal entity, but only a patent is a true asset. An application is a mere hope of an asset. This potential asset becomes a real one, once the patent is granted.
You seem happily forget one fundamental aspect of patenting: after examination, the patent is as it should have been filed, and what was subject-matter of the claims might have changed totally. And it is there that the adaptation of a description comes into play.
Despite an alleged mountain of contrary evidence, which only looks well-reasoned for those who want to be convinced, I maintain that the description has to be adapted to the claims as granted. Any feature in the description having led to a limitation of the independent claim cannot be left as optional in the description at the moment of grant.
By doing so, the EPO does not interfere with the possibility of subsequent courts to decide upon direct/indirect infringement or infringement by equivalents. The EPO merely blocks the proprietor of changing the interpretation as he thinks fit. For validity, the limited claim matters, but for infringement with the optional character left in the description abroad interpretation matters. This what all the boards applying G 1/24 have rightly done.
If have thus good reasons not to deny anything as your reasoning is only convincing for those who want to be convinced. The inescapable conclusion is that by applying G 1/24, the description has already to be adapted, and from there we are not far from any further necessary adaptation of the description.
OK, let's agree to differ on the "unitary legal title" point. It very much feels like you are not engaging with the substance of the points that I have made, and that we are therefore talking past one another.
DeleteI have to say that I am a little surprised about your comments on Art 67 EPC. The extent to which that provision is used by applicants is not at all relevant to a discussion of what principles or information can be deduced from the wording and/or practical effects of that provision. I would also point out that a positive provision of the EPC cannot in any sense be reasonably described as "an exception".
In any event, it seems that we agree that a downstream court faced with an assertion of provisional protection under Art 67 EPC would need to assess the protection provisionally provided by reference to the claims and description of the application as filed. Thus, if post-filing amendments change the interpretation of terms used in the claims of the application as published, how might this affect the assessment of infringement of an unamended claim by an embodiment which falls within the scope of the claim as originally published but - due to description adaptations - not within the scope of the claim of the patent as granted?
I really would appreciate it if, for once, you could engage with the substance of the point that I am making. Frankly, you are not going to persuade anyone if you instead deploy tactics such as of refusing to address "hypotheticals", repeating "hypotheticals" of your own that you believe prove your point and/or dismissing or minimising the legal issues that my arguments engage.
Two observations on the above exchange. I will try not to repeat points already made.
Delete1. DXThomas now states that the EPO has "the duty which has been conferred to it by its contracting states when it was set up in the early seventies" to remove inconsistencies between claims and description. This is a remarkably specific historical claim. If such a duty was conferred at the Munich Diplomatic Conference of 1973, it will appear in the EPC, in the travaux préparatoires, or in the Conference minutes. I have been unable to identify any such reference. The travaux are silent on the point. If I am wrong, I invite DXThomas to identify the specific source. If I am right, then the "founding duty" is not a historical fact but a retrospective inference, which the founding documents do not support.
2. Proof of the Pudding's Article 67 point deserves more than the dismissal it received based purely on Article 67 only being invoked rarely. The argument is not that provisional protection is frequently invoked. The argument is that it exists, and that its existence creates a structural problem for mandatory adaptation. It exposes a dilemma. Either adaptation is interpretively significant, in which case it creates problems the EPC does not solve, or it is not, in which case there is no need for it. Responding that Article 67 is "barely applied" does not escape the dilemma. It merely declines to engage with it.
And here's Claude's take on the least persuasive arguments:
ReplyDeleteA. Least Persuasive Against Adaptation
1. "All other IP5 offices don't require it, so the EPO shouldn't either" (many amici)
While factually broadly correct, this argument from international harmonisation proves very little. The EPC is a self-contained legal system with its own provisions, and the fact that other offices with different statutory frameworks take a different approach is not particularly informative about what the EPC requires. The EPO also has numerous practices that differ from the USPTO or JPO. Moreover, as the President's survey shows, several EPC contracting states' national offices do require adaptation.
2. "Adaptation is expensive and burdensome" (many amici)
This is a policy argument, not a legal one. If the EPC requires adaptation, the cost of compliance is irrelevant to the legal question. Many EPC requirements are expensive. The practical burden argument should inform how adaptation is applied (proportionately, at the end of proceedings, etc.) but does not answer whether it is legally required. Several amici spend considerable space on this point when it is ultimately beside the question the EBA must answer.
3. "The referring board engineered the referral" (anonymous brief)
The anonymous amicus argues that the Board should have remitted to the Opposition Division rather than referring to the EBA, and that the referral was "contrived." While the procedural history is unusual, this argument essentially asks the EBA to decline a referral on a point of fundamental importance with clear diverging case law because the referring board could have taken a different procedural path. The EBA has repeatedly held that it will examine admissibility on its own terms. This argument is unlikely to find favour.
B. Least Persuasive In Favour of Adaptation
1. "Rule 42(1)(c) requires the description to disclose the invention as claimed" (various)
Rule 42(1)(c) says the description must "disclose the invention, as claimed, in such terms that the technical problem and its solution can be understood." This is about disclosure — ensuring the description contains enough to understand the claimed invention. It does not say "the description shall disclose only the invention as claimed." Reading it as a prohibition on disclosing more than what is claimed stretches the text beyond breaking point. Several pro-adaptation amici rely heavily on Rule 42 but the textual argument does not survive close scrutiny.
2. "Consistency as a general principle of law" (some first-line decisions)
Some decisions cite a general principle that the parts of a patent should be internally consistent. No such general principle is found in the EPC, and deriving one from the overall structure of the Convention is circular reasoning — it assumes the conclusion. The EPC tolerates many forms of "inconsistency" (e.g., dependent claims that are narrower than independent claims, description that includes prior art discussion not reflected in the claims). Elevating "consistency" to a free-standing legal requirement creates an untethered basis for objections.
3. "40 years of case law supports the requirement" (President)
The longevity of a practice does not establish its legal correctness. As several amici note, the early decisions (T 11/82, T 403/88, etc.) contained little or no reasoning for the adaptation requirement, typically stating it conclusorily. The first detailed examination of whether the requirement existed was T 1989/18, and the first comprehensive refutation was T 56/21. The argument from tradition is weak precisely because the tradition was never interrogated until recently.
@Proof of the pudding
ReplyDeleteMy comments on your question 1.
If we focus on the case law of the UPC, key principles are
1.Primacy of the claims : inconsistencies with the description do not affect the conclusions.
2.Interpretation relies on the explanatory value of the description. This implies that even embodiments disclaimed by description amendment may be helpful for understanding and should not be disregarded (see LD Paris HP vs Lama [358/2023] 13 Nov 2024, point 41). Hence the reliance on the description of the original application and the likely irrelevance of post-filing amendments.
3.The interpretation of claims is the same for validity and infringement assessments. Since compliance with the EPC conditions (novelty/IS/sufficiency/123(2) relies on the filing date and the content of the application as filed, the logic is to rely on the application as filed.
My conclusion is : no impact or very marginal.
On your question 2, it seems to me that previous EBA interpretations which had downstream impact related to the scope of patentable subject matter.
Francis, thank you very much for your thoughtful response.
DeleteThe reason that I asked is that interpretations of the EPC in EBA rulings are not only (effectively) binding on EPO Divisions and Boards of Appeal, they are also often viewed as persuasive (and occasionally even binding) for national patent offices and downstream courts.
Thus, for example, if the EBA interprets the EPC as lacking any requirement for description adaptation, this could well have an impact upon those (few) national offices which require at least some form of adaptation.
On the other hand, if the EBA decides that the EPC does require strict adaptation of the description, this could have a number of profound downstream consequences.
For example, might national patent offices (whose national laws include provisions that are intended to replicate the requirements of Art 84 EPC) feel obliged to start demanding "strict" description adaptations?
On the other hand, how will national courts square their Article 69 EPC and Protocol claim interpretation obligations with a national law provision that has been interpreted by the EBA to require adaptation of the very description that the court used to derive an interpretation of (amended) claims that is has found to be patentable? The potential for adaptations to shift the interpretation of the claims then becomes very real, potentially leading to an endless loop of amendment followed by re-adaptation of the description to the claims as interpreted in the light of the adapted description.
Thus, it is not so much that the national patent offices and downstream courts do not already have well-established case law. Instead, it is that there is a risk of the EBA ruling upsetting the apple cart for all concerned.
Me, I will be surprised if the EBA hands down an edict for "strict" conformity and equally surprised if it declares that no adaptation at all is required. Choosing one of those two options will have far too many unpleasant and likely unintended consequences. Requiring "strict" conformity is making a brutal rod for the EPO's own back, for example.
DeleteThe CIPA amicus brief suggests a way forward. And elsewhere in this thread I have sugested that, given a pragmatic decision by the EBA, applicants will henceforth become masters of their own fate, or authors of their own misfortune, if they are too clever by half and deliberately leave disconformities in their description, supposing that their carefully husbanded Angora cats will win a prize. More like come back to bite them, I fancy.
One issue which I have not heard any discussion about is the wording of Art. 84 in the languages other than English, and whether these have a different flavour to that of the English version.
ReplyDeleteIn French, we have "Elles doivent être claires et concises et se fonder sur la description." Not being an expert linguist, I put this into a translation machine and got back "They should be clear and concise and based on the description." Quite different to "supported by".
In German, we have "Sie müssen deutlich und knapp gefasst sein und von der Beschreibung gestützt werden." The same translation machine comes up with the same language as the English version (it could have been trained to do that, of course).
Does Art. 177 lead to the majority verdict winning?
@ Kant-11.02.2026-16.25.00 GMT
DeleteIrrespective of the fact that the three texts in the EPC are authentic, I just made a test with Deepl.com.
Deepl.com translated the sentence “They shall be clear and concise and be supported by the description” with
- “Sie müssen klar und prägnant sein und durch die Beschreibung untermauert werden“.
- “Elles doivent être claires et concises et être étayés par la description“
To me, “untermauert“ and “être êtayées“ is much stronger than merely “supported”.
Deepl.com translated the sentence “Elles doivent être claires et concises et se fonder sur la description“ in
- “They should be clear and concise and based on the description”.
- „Sie müssen klar und prägnant sein und sich auf die Beschreibung stützen“
I suspect that other translation machines will come up with other versions, but I would therefore not think that Art. 177 will necessarily lead to the majority verdict winning. Interpreting a rule of law has nothing to do with a majority of opinions.
It will entirely depend on how the EBA interprets “supported by the description”, “von der Beschreibung gestützt werden“ “se fonder sur la description“.
Thank you M. Thomas for your reply to my post.
DeleteI think the interpretation of a rule of law always will come down to the majority opinion in our legal systems, but I agree as each version is equally valid, one cannot be overruled by the other two. Nevertheless, the French version, with "se fonder sur" can be used to interpret "supported by" if necessary, especially as both also appear in the language of the PCT.
As a native French speaker, I would be interested in how you would understand the French wording. If it is like the English "shall be based on", this would give an strong indication to me of the drafters of the treaties implying a one way requirement, leading from the description to the claims and not vice-versa.
@ Kant-13.02.2026 at 10.38.00 GMT
DeleteWhen you think that the interpretation of a rule of law always will come down to the majority opinion in our legal systems, it does not mean the majority of those having an opinion on the topic, but the majority within the deciding body which will prevail. I see there a huge difference.
I do not which to overstress the argument, but people in a country might be for or against a certain piece law, but at the end of the day, it is the deciding body, for instance a parliament, which will decide on it.
As a French native speaker, I do not think that Art 84 is unidirectional. The locution "shall be based on" does not merely mean that what is in the claim has be found somewhere in the description. To me it applies in both directions. A feature added in an independent claim cannot be left as being optional in the description.
@DXThomas
ReplyDeleteI've read your exchanges across this thread and others with interest. Before addressing substance, I note with concern certain patterns in your argumentation that undermine productive dialogue.
You have repeatedly played the man and not the ball. These behaviors are inappropriate and contrary to the norms of professional discourse. If you have substantive responses, make them. If not, please refrain from these tactics.
You frequently cite cases without engaging the principles they establish. When pressed on logical gaps, you declare that your interlocutors "comfort your views" rather than addressing their arguments. This is not persuasive.
Now, to substance. I have four direct questions that go to the heart of the legal disagreement:
Question 1 (Statutory Interpretation):
Article 84 EPC states: "The claims... shall be... supported by the description."
Please explain why this phrase necessarily means:
(a) the description must be consistent with the claims; AND
(b) the description must not contain subject matter broader than the claims.
You have asserted this repeatedly but never explained the linguistic or logical path from the words "supported by" to this interpretation. The normal meaning of "support" is unidirectional - a foundation supports a building, but we don't say the building must exclude everything not load-bearing from the foundation.
Question 2 (The Agfa Contradiction):
You state adaptation is essential to prevent "Angora cats." You also state the Hamburg court "was not fooled" by the unadapted description and correctly concluded that "specifications in the description that are not consistent with the granted claims cannot serve as a basis for a broad interpretation."
If courts can correctly interpret unadapted descriptions (as Hamburg did), why is pre-grant adaptation necessary? You call this "two issues" but haven't explained how they're reconcilable.
Either:
Adaptation is necessary (courts can't handle it) - which contradicts Hamburg managing fine; OR
Adaptation is unnecessary (courts can handle it) - which undermines your position.
Which is it?
Question 3 (The Clarity Problem):
You maintain a claim can be "clear as such" while simultaneously requiring description adaptation because the description "allows a broader scope of protection than that which can be derived directly and unambiguously from the claim wording."
How can a claim meet the clarity requirement of Article 84 if the description creates doubt about its scope? Under your logic:
If the claim alone is clear then there is no description problem (courts just read the clear claim)
If the description creates doubt then the claim lacks clarity (remedy: amend the claim)
You've never explained what Article 84 violation exists when the claim itself is clear.
Question 4 (The EPO's Role):
You state: "The duty of the EPO is... to grant or maintain patents which are not open to multiple possible interpretations."
Please cite the EPC provision establishing this as the EPO's duty.
You acknowledge (correctly) that Article 69 and its Protocol govern scope of protection, which is a matter for courts. You acknowledge the EPO cannot guarantee courts will agree with its interpretation. Yet you maintain the EPO must nonetheless impose its interpretation via description adaptation.
This appears to be a claim that the EPO should control post-grant interpretation despite lacking post-grant jurisdiction.
Where in the EPC is the EPO given this role?
Please answer these questions directly, and not merely by citing Board decisions (which are interpretations, not the law itself) or by asserting that "this is what the Boards have decided."
If you cannot answer them, perhaps consider whether the positions you've staked out are as legally solid as you maintain.
While we wait for DXT I want to refer to a comment of his on 2nd Feb at 21:04. On G1/24 it commends to us T2027/23, the "somersaults" case, for its musings on Angora cats. Looking at the text, I find that it reminds patentees that they are the masters of their own fate. Perhaps a veiled threat from the TBA, that if they choose to go to grant with an inconsistent description they can't then complain when the inconsistency is used against them. All this reminds me of the EPO attitude on Art 123(2): any amendment is at your own risk. Tricky then, isn't it, for Applicants. Damned if you do (amend) but damned again if you don't.
Delete@ The FRANDalorian-11.02.2026-19.06.00 GMT-Reply part 1
DeleteFor a start I profoundly dislike your apodictic approach, culminating in me not being allowed to quote decisions. Who do you think you are? Where do you think you have got the legitimation to come up with such constraints? You would do well in moderating your tone and expectations. This is the more so that you come yourself with the Agfa decision.
I will reply as I think fit and if you do not like it, its your problem not mine. The absence of a reply does not mean that I agree with you, but the question is then simply too silly for me to spend time answering it.
Question 1
Compare the claims with the flat roof of a tunnel having vertical straight walls. At filling the vertical straight walls are wide apart. On the flat roof you can place a sheet of paper representing the description disclosing the invention as claimed. At filling the claims are fully supported by the sheet of paper which is as wide as the flat roof of the tunnel. Art 84 is fulfilled.
In view of the prior art revealed during examination, the vertical straight walls come then much nearer to each other at the time of grant. If now you put the same wide sheet of paper representing the invention on the narrow flat roof of the tunnel corresponding to the claims as granted, you will necessarily see that the original sheet of paper as wide as the original tunnel roof will hang down on each side of the narrow flat roof. The centre part of the description will still support what is claimed. This was always there and has not changed. However, it cannot be said that the sides hanging down support the claim as granted. They are simply flapping around and have no relation with the claims.
Either you tailor the actual description of the invention to the size of the narrow flat roof or you explain that the sides hanging down are not corresponding to what is actually claimed. This is to me an elegant way to look at the necessary adaptation of the description to the granted claims.
After grant, the proprietor cannot say that his claims are still as wide as originally filed and the description cannot be said to still cover the original wide flat roof. Otherwise the sides would not hang down. In other words, the original description of the invention does not any longer support the claims as granted, cf. Art 84.
Question 2
You have manifestly not read the Agfa decision. On the one hand, in the Agfa decision the court held that “specifications in the description that are not consistent with the granted claims cannot serve as a basis of a broad interpretation of a claim”, cf. Headnote 3.
On the other hand, I quote page 23: “There are several passages mentioning “chromatic” in the specification of the patent that should have been deleted”.
In the present case, the court has noted the inconsistencies, which should have been deleted, but at the same time it was not abused by the not adapted decision.
You should give more credit to courts, not only in the Agfa decision, than you do in your question.
My conclusion is that for the court, the description was not adapted, but the court was indeed not fooled by it.
This does by no way mean that the description is not to be adapted. If the court says that what is not falling under the claim should have been deleted, it means that the court is directly and unambiguously in favour of the adaptation of the description.
@ The FRANDalorian-11.02.2026-19.06.00 GMT-Reply part 2
DeleteQuestion 3
I maintain hat a claim can be "clear as such" while simultaneously requiring an description adaptation because the description "allows a broader scope of protection than that which can be derived directly and unambiguously from the claim wording."
Take a simple example: the claim defines a spade, which is an item every skilled person knows. In the description you find the statement that a spade is a earth moving implement. This implies that any earth moving implement may fall in the scope of protection of the claim. Under earth moving implement, the skilled person can understand various items like an excavator or a bulldozer.
Every skilled person will immediately recognise that the claimed spade has very little to do with an excavator or a bulldozer. The spade is actuated by hand, and the excavator and the bulldozer need an engine in order to function. There is a clear cut inconsistency between the claimed object and the scope of protection resulting from the statements in the description.
This allows the applicant/proprietor to fence off any prior art relating to earth moving implements other than a spade, for instance a rake, but to claim that the scope of protection extends to any earth moving implement like an excavator or a bulldozer. This is a perfect illustration of what is known in the profession as an Angora cat.
If the notion of earth moving implement is brought in the claim as required for instance in T 2027/23, the prior art that will be held against the claim will be totally different.
Needless to say, that it also works the other way round, with the earth moving implement in the claim and the spade in the description.
Question 4
By forcing the adaption of the description, be it under Art 84 alone, support, supplemented or not with R 42 or 48, the EPO will insure that the applicant/proprietor will not be allowed to change the interpretation as he thinks fit. I would think that, in general, the duty of any granting authority should be to avoid the creation of any possible Angora cat. The EPO is no exception.
This will not hinder any court acting in infringement to decide upon direct or indirect infringement or infringement by equivalents. This is what Art 69 and the Protocol on interpretation of Art 69 are for. It will certainly avoid the proprietor hiding in the description what he wants to see protected.
Closing comments
One aspect which has never been thought off, is that those so much in favour of not adapting the description are at one time acting as proprietors but at other times acting as third parties. What is positive as proprietor can be negative as third party. Or is it true, as I have been told, the more uncertainty and litigation possibilities exists, the higher the billable hours accrue?
The positions I have brought forward are legally solid and I maintain them, for all the reasons given above here.
For the surplus, I will now stop answering questions of yours which you could have answered yourself by thinking carefully about the situation.
@Proof of the Pudding & @Maxdrei
DeleteI focused on the UPC in my comment in response to Proof’s questions regarding downstream consequences of the EBA decision. As you know very well, the case law within Europe shows great differences, which will likely result in different consequences.
Take German case law. With the bifurcation of validity and infringement, the interpretation of claims by the court competent for infringement and by the court competent for validity may be different. In addition, German courts competent for infringement do not apply file wrapper estoppel. As a result, a patentee may choose to delete from the description an embodiment no longer covered by amended claims, in order to maximise their freedom to argue about equivalents. This may explain the cautious answers provided in GRUR’s amicus.
Because fh mentions me I feel that I should reply.
DeleteYes, of course, with bifurcation the infringement court and the validity court can interpret the claim differently. But I was told several decades ago that the respective judges are well-aware of the embarrassment that can follow when they differ. So, in the real world, they are careful not to disagree.
As I mentioned in a post earlier this week, an inconsistency is a 2-edged sword. Perhaps it allows a patent owner to contend for an interpretation that delivers infringement without falling into invalidity. But then it will also allow the infringer to contend for both non-infringement and invalidity. Just as an ill-judged prosecution amendment puts a patent owner into the 123-2-3 inescapable trap, so a retention of an inconsistency can have the same disastrous outcome.
As the EPO observes, applicants are masters of their own fate and the courts have enough wiggle room to punish those applicants who choose deliberately to breed Angora cats. Not many patent owners these days find themselves in the 123-2-3 inescapable trap invented by the EPO. They have mastered that fate. They know what to do to avoid it. Likewise, the way I see it, the problem of egregious inconsistencies (up vs down, or chromatic vs achromatic) will disappear once the courts start to lay down robust case law that allows EPA prosecutors to counsel their US clients on the inadvisability of following the urgings of their US litigator colleagues to cultivate in the EPO description as many inconsistencies as can be preserved beyond grant.
I am surely not the only one who remembers US attorney clients refusing to believe that any jurisdiction could deliberately create an inescapable trap. But they believe it now.
It will take quite some time, and robust decisions form the UPC, to bring US-based patent attorneys round to the mindset that going to grant in the EPO with a specification riddled with grotesque inconsistencies really is NOT a good idea, so bad that it could expose them to a malpractice suit brought by their client. We can all look forward with pleasure to the day when such gross contradictions are purged by the prosecuting attorneys themselves, prior to grant, in the interests of their client. It all depends on the UPC.
Overnight, another thought occurs, on bifurcation.
DeleteIsaac Newton observed that every force has an equal and opposite reaction. The Doctrine of Equivalents is a serious force. Its opposite we can call the Reverse Doctrine of Equivalents. The applicant who goes to grant with a narrowed claim (say "up") has chosen to confine their scope of protection accordingly. But in taking to grant a decription that includes "up includes down" they have deliberately stated more widely what is their invention, their contribution to the art.
Art 69 EPC has done away with the notion that what comes before and anticipates will, if coming later, infringe. Under Art 69 EPC the courts can say that validity is determined on the basis that up includes down whereas the scope of protection goes no further than up.
But what about the reverse case, I hear you say, a description that states "up means up" with a claim that states "up includes down"? The courts would surely find a way to deny a claim that down infringes, right? That is, if the EPO didn't get on to it first, already, prior to grant.
So, can we not hope for a G1/25 Decision that trusts to the good sense of the EPO and the courts, to cope with any remaining inconsistencies in such a way that, from now on, applicants and their advisers will be increasingly inclined to eliminate them all, prior to grant, in their own interests?
@DXThomas:
ReplyDeleteThank you for responding. Let me address the substance. I'll leave the ad hominems where they fell.
Question 1. I asked for the linguistic path from "supported by" to your interpretation. You gave me a tunnel. The analogy is vivid but self-defeating: paper overhanging the walls is still supported by the walls where it contacts them. The overhang doesn't undermine the structure, it's simply surplus. Article 84 says claims shall be "supported by the description." It does not say "the description shall contain nothing beyond the claims." Your analogy assumes congruence where the statute requires only support. That's the gap I asked you to bridge, and the tunnel doesn't bridge it.
Question 2. You say the Hamburg court wasn't fooled by the unadapted description, AND that the description should have been adapted. You appear not to see that these statements concede my point. If the court reached the correct interpretation despite the lack of adaptation, the adaptation wasn't necessary for correct interpretation. You say the court's statement that material "should have been deleted" proves it favours adaptation, but a court expressing a preference for tidiness is not the same as establishing a legal requirement. The dilemma stands. You've chosen both horns of the dilemma simultaneously, which is not resolution but contradiction.
Question 3. Your spade example is helpful as it proves my case. You say every skilled person will "immediately recognise" that a spade is not a bulldozer. Exactly. So the claim is clear, the skilled person isn't confused, and no Angora cat actually materialises in interpretation. If the claim says "spade" and the skilled reader knows what a spade is, the description's reference to "earth-moving implements" doesn't change the scope of protection. No court is going to read "earth-moving implement" in the description and conclude the patent covers excavators. Under Article 69, claims define that scope; the description aids interpretation. Your example demonstrates that clear claims do their job without modifying the description by surgery.
Question 4. I asked you to cite the EPC provision. You replied: "I would think that, in general, the duty of any granting authority should be to avoid..." "I would think" is not a legal citation. It is a policy preference. Your entire answer to Question 4 contains not a single reference to a specific EPC article, rule, or provision. This is remarkable for someone who claims his positions are "legally solid." You have confirmed that no such provision exists by being unable to cite one. That the EPO *should* do something (in your view) is not the same as the EPC *requiring* it. The distinction matters when we are discussing what the law actually says, which is what G1/25 must decide.
On tone. I opened my previous comment by asking you to engage arguments rather than impugn motives. Your response opened by attacking me personally, continued through suggesting that I've not read decisions or am too stupid to understand your position, and closed by suggesting my position is motivated by billable hours. I note this not to complain but to observe that it rather proves the point that Jackson Lamb and others have made. It's also logically confused: if uncertainty generates fees, then practitioners who oppose adaptation (and thus supposedly benefit from uncertainty) would be acting against their interests by advocating for clear rules through the G1/25 proceedings. The argument cuts both ways or not at all. The questions remain. Your answers so far have not met them.
@ FRANDalorian-12.02.2026 at 21.31.00 GMT
DeleteI have spent enough time on my replies. They are manifestly not to your liking. I do therefore not intend to pursue the discussion as it is to no avail.
Whatever you might think or say, Art 84 is bidirectional. In the Agfa case, the UPC-CFI LD Hamburg made clear that the description should have been amended. If the claims says spade you cannot leave in the description that the spade is any earth moving implement. The scope of protection has to be commensurate with the contribution of the invention to the state of the art. This is the basis of every patent system. The description is not there to hide something which is not in the claims.
To use your own words, your counter comments seem rather to prove my point.
As far as the tone is concerned, I would simply come up with the following statement: “What goes around comes around”.
My replies are manifestly not to your liking. I do therefore not intend to pursue the discussion as it is to no avail. This does not mean that I would not like to answer or that I have no answer, but I will not waste any more time.
DeleteFor the surplus I invite you to look at the new Guidelines F-IV, 4.1 and 4.2. They have taken over the board’s way to interpret G 1/24.
Needless to say, that I am in full accordance with the way the boards interpret G 1/24 and with the new Guidelines.
I allow myself to add that the aim of G 1/24 was harmonisation with judicial bodies acting in post-grant or post-opposition procedures. Some decisions applying G 1/24 have expressly taken the same stance.
I maintain that judicial bodies acting in post-grant or post-opposition procedures will still be able to decide upon direct or indirect infringement or infringement by equivalents.
They will simply not have to come to interpretative summersaults when looking at claims and description which are not aligned one with the other. This is the positive result of G 1/24.
When looking at the amendments proposed in the Guidelines F-IV, 4.1 (Clarity) and 4.2 (Interpretation of the claims) to enter into force on April 1st, a great part of the discussions above appear surreal.
DeleteI quote from new Guidelines 4.1
- “their clarity [of the claims] is of the utmost importance for reasons of legal certainty”
- “There should be no doubt as to which subject-matter is covered by the claims”
-
I quote from new Guidelines 4.2
- “the description and drawings cannot be relied on to read into the claim a restrictive feature not suggested by the wording of the claim”
- “If the description provides a special broad definition of a term used in a claim, the claim must be interpreted in the light of that broad definition when assessing patentability”
I conclude directly and unambiguously that the new Guidelines alone will force applicants/proprietors to delete before grant or maintenance in amended form, the optional character of a feature incorporated as a limitation in an independent claim.
If the claims says “spade”, the description cannot say “spade=any earth moving implement”, or vice-versa.
I have published the full wording of the new Guidelines on my blog, with a corresponding comment.
The Guidelines are, however, not binding on the Technical Boards of Appeal - still less the Enlarged Board of Appeal.
DeleteMr Thomas, in lieu of an answer to the logical gaps which I pointed out above, you have instead repeatedly pointed to the new Guidelines as though they settle the matter, going so far as to describe the preceding discussion as "surreal" in the light of their content.
DeleteWith respect, this argument has the matter precisely backwards. The Guidelines for Examination are an internal administrative instrument issued under the authority of the President of the EPO. They are not a source of law. They do not bind the Boards of Appeal. This is not a controversial proposition — it is one that the Boards themselves have affirmed on numerous occasions, and one that you yourself have acknowledged in other contexts.
Indeed, the very reason that G 1/25 is pending before the Enlarged Board is that there is a fundamental disagreement — among the Boards of Appeal, among practitioners, and among the users of the patent system — as to whether the EPC requires what the Guidelines purport to require, at least with regard to description amendments. If the Guidelines were determinative of the legal question, there would be no need for the referral at all. The fact that the referral exists is itself proof that the Guidelines cannot be treated as authority.
It is also worth recalling the procedural sequence. The President of the EPO has filed submissions in G 1/25 advocating for mandatory adaptation of the description. The same President is ultimately responsible for the content of the Guidelines. It would be remarkable if the Guidelines did not reflect the President's position. But the question before the EBA is whether that position is correct as a matter of law — not whether it has been administratively promulgated.
To cite the Guidelines as support for the President's legal position is, with respect, entirely circular: the President says X in his submissions; the President has also written X into the Guidelines; therefore X must be correct.
You are, of course, well aware that the Boards of Appeal are not bound by the Guidelines. This distinction is fundamental to the independence of the Boards: an independence whose importance you have yourself emphasised in other discussions.
There is a further irony. Several of the amicus briefs filed in G 1/25 — including those from epi, CIPA, AIPLA, and numerous practitioners — have drawn attention to the fact that the Guidelines have been progressively amended over the years to reflect an increasingly strict approach to description adaptation, without any corresponding change in the law itself. The question of whether the Guidelines are a faithful reflection of the EPC, or whether they have drifted beyond what the EPC requires, is effectively one of the central issues in G 1/25.
To rely on those same Guidelines as evidence that the law requires what they say it requires is to assume the very conclusion that the EBA has been asked to examine.
None of this is to say that the Guidelines are irrelevant. They reflect the considered view of the EPO's administration on how the EPC should be applied in practice, and the EBA will no doubt give them appropriate weight. But appropriate weight is not the same as dispositive authority. The EBA's task is to interpret the EPC, not to rubber-stamp the Guidelines.
I would therefore respectfully suggest that, until the EBA has spoken, it would be prudent to refrain from treating the Guidelines as though they have already answered the question that the EBA has been asked to decide.
@ Anon Y. Mouse-16.02.2026-17.10.00 GMT
Delete@ The FRANDalorian-16.02.2026 at 17.25.00 GMT
Reply part 1
Having worked at the EPO, I know too well that the Guidelines are not binding to the boards and certainly not to the EBA. The only point I wanted to make is merely that the official policy of the EPO as of April 1st 2026 will be to follow the board’s decisions applying G 1/24.
Applying G 1/24 means amending the claims, if the interpretation found in the description is to be taken into account. A different interpretation in the claim vs. the interpretation in the description represents an inconsistency.
A logical conclusion is that, by bringing in the claim the interpretation found in the description, one big source of inconsistency thus will be removed quasi automatically.
G 1/25 is also about removing inconsistencies and has therefore a direct link to G 1/24.
Applying G 1/24, should the applicant/proprietor refuse to amend the claim accordingly, the prior art against the claim will either be
- the prior art needed to deal with the broad claim, and not with the limited interpretation found in the description, or
- the prior art needed to deal with the broad interpretation found in the description, and not be limited to prior art relevant only to the limiting feature in the claim.
If the applicant/proprietor refuses to amend the claim accordingly, the free disposition opened to applicant/proprietor [Art 113(2)], will not lead the EPO to ignore the inconsistency, and the EPO will not deal with the file as the applicant/proprietor wishes, i.e. with two concomitant but not converging interpretations.
I have noted that G 1/24 already implies removing one fundamental inconsistency, and G 1/25, should, in my opinion remove the remaining ones.
@ Anon Y. Mouse-16.02.2026-17.10.00 GMT
Delete@ The FRANDalorian-16.02.2026 at 17.25.00 GMT
Reply part 2
As explained, and not coming as a surprise for people willing to understand, already G 1/24 requires removing the optional character in the description of a feature once this feature has been used to limit an independent claim.
For instance in the famous Agfa/Gucci case, G 1/24, would have necessarily brought the ED to request removal at grant of the feature “chromatic” from the description, as the claim had been limited to “achromatic”. From the file, it is however manifest that the ED did not bat an eyelid at the not adapted description and waived the file through to grant as the claim had been limited.
Indeed, several Amicus Curiae have drawn attention to the fact that the Guidelines have been progressively amended over the years to reflect an increasingly strict approach to description adaptation.
I do not intend to reopen a old can of worms, but if you have followed the evolution, the strict approach has been originally provoked by the attitude of the applicant in the famous “pemetrexed” case, refusing bluntly to adapt the description to the only example given which was the disodium salt. With G 1/24 this attitude would have led to a refusal of the application, as the applicant at the time insisted on maintaining a broadening statement in the description.
When later, with due respect to their function and person, both Lord Neuberger and Sir Robin Jacob were of the opinion that the examiner was wrong in raising an objection under Art 123(2) against a claim only referring to pemetrexed in general, they made clear that they ignored the fact that at the EPO equivalents do not fall under novelty but under inventive step. To asses added matter it is the novelty criterion which applies. Even the UPC has adopted this point of view.
Coming back to the original topic, I would therefore respectfully argue that in my last comments, and in the present one, I did not treat the Guidelines as though they have already answered the question that the EBA has been asked to decide.
The Guidelines as of April 1st 2026 do clearly and unambiguously answer what the EBA said in G 1/24, as applied by the boards.
With due respect to the EBA, I still maintain, that G 1/24, dealing with one inconsistency, has paved the way for G 1/25 dealing with the remaining ones.
@DXThomas
DeleteI have one question regarding regarding the "logical conclusion" from 18:57:
Why does it have to be an amendment of the claim? G1/24 only states that a response to any genuine finding of lack of clarity is amendment. Not solely an amendment of the claim. If the claim discloses something different from the description, the amendment can also be an amendment of the description, be it in deleting parts of definitions that are broader than the claim or deleting parts that are more limiting than the claim. If the claim discloses something broader than the description there should be no Art. 123(2) problem. Shouldn't all three options be possible and should it not be up to the applicant to decide which way to go?
PS: I think such a clarification should also be in the Guidelines. Including a clarification that not all objections as to a lack of clarity are genuine. I can already foresee how examiners may now use the new Guidelines to block any argument that they may be wrong and that no lack of clarity exists in the first place, but they just misunderstood something.
@ Kant, Max Drei, Mr Hagel, Proof of the pudding, Anon Y. Mouse, Anonymous, Odo Ital, FRANDalorian and Co,
ReplyDeleteAll decisions following G 1/24 have made clear that, if the scope of protection/interpretation derivable from the plain claim wording is different from the scope of protection/interpretation derivable from the description, or vice-versa, the claim has to be amended, so that only one single scope of protection/interpretation remains.
The applicant/proprietor has the choice of the scope of protection/interpretation he wants to be taken into account by bringing it into the claim. A different scope of protection/interpretation derivable from the claim or from the description will not any longer be possible, hence the end of a patent granted or maintained in amended from by the EPO being its own dictionary.
Thereby claim and description will be aligned with each other. This is one direct form of adaptation of the description. In other words, a broadening statement in the description will have to be deleted if the claim has been limited. Conversely, a limiting statement in the description will have to be deleted if the applicant/proprietor claims a broad scope of protection/interpretation.
This is exactly what G 1/25 is about.
The above considerations are clearly confirmed by the Guidelines F-IV, 4.1 and 4.2 applicable as of April 1st 2026. The new Guidelines are fully taking into account, the decisions applying G 1/24. In my opinion G 1/25 will do the rest.
This is why I think that G 1/24 has paved the way for G 1/25. I cannot imagine the EBA coming up with an interpretation different from that of the boards applying G 1/24, but we never know.
With respect, I think the discussion is now at a slightly different stage and we may be talking past one another.
DeleteYou say that all post-G 1/24 decisions have made clear that only one single scope of protection/interpretation may remain, and that the claim must therefore be amended wherever the description permits a different interpretation.
That may indeed describe the direction taken in several recent Board decisions. But that still leaves open the central question for G 1/25:
Where in the EPC does it follow that the existence of two potential interpretive routes — one from the plain wording of the claim and one from the description — automatically renders the patent non-compliant unless the description is purged?
G 1/24 established that the description must always be consulted when interpreting the claims. It did not state that any interpretive tension must be eliminated by amendment of the description prior to grant or maintenance. Consulting the description and deleting parts of it are conceptually distinct steps.
If the argument is that post-G 1/24 Boards have mostly converged on their interpretive approach, that may be descriptively accurate. But convergence in case law does not itself answer whether that direction of convergence is textually compelled by the EPC, or whether it reflects a policy preference for administrative clarity.
The referral exists precisely because the legal basis for mandatory adaptation remains debated even after G 1/24. If the matter were already settled as a necessary consequence of G 1/24, G 1/25 would be redundant.
@ The FRANDalorian-16.02.2026- 20.48.00 GM
DeleteYou might not have noticed it, but I did. It is noticeable that The FRANDalorian and Proof of the pudding are the only commenters on IPKat who use various fonts, like bold (or italics), when commenting on IPKat.
This, combined with the constant insistence on the role of the description in the interpretation of the claim, irrespective of the claim wording and its interpretation by the skilled person, leads to the necessary conclusion that The FRANDalorian and Proof of the pudding, are one and the same commenter. It will be easy to deny it, as neither The FRANDalorian nor Proof of the pudding act in the open. Any denegation will be of little value.
You and/or Proof have manifestly not accepted the way the boards apply G 1/24. You still cling to your idea that the interpretation found in the description should prevail over that resulting from the plain wording of the claim.
As far as I am concerned there are two compelling reasons for adapting the description in case of an inconsistency: G 1/24 and Art 84.
The first, G 1/24, is that, when assessing patentability under Art 52-57, the prior art to be taken into account is either that resulting from
- the broad interpretation in the description in spite of a limited claim, or
- the broad interpretation of the claim in spite of a limitation in the description.
Consulting the description and deleting parts of it are indeed conceptually distinct steps, but if the applicant/proprietor wants the definition of the description to prevail, then the claim has to be amended accordingly.
This direction of convergence might not textually compelled by the EPC, but the boards are there to apply decisions of the EBA, and have made clear what is expected from the applicant/proprietor in their decisions following G 1/24.
All decisions applying G 1/24, do not deny the existence of two potential interpretive routes, one from the plain wording of the claim and one from the description. They simply refuse to be forced to do “interpretative summersaults”. It results that for the boards, in application of G 1/24, only one interpretative route is allowable, and this is that resulting from the claim wording.
Or do you claim that the boards do not know how to apply decisions of the EBA and that they are plainly wrong in their decisions?
In Reasons 2.2.3 of T 0981/23, the board reminded that Art 84 requires claims to be clear, and the meaning of a claim feature should be clear for the person skilled in the art from the wording of the claim alone, and referred to G 1/04, Reasons 6.2. The correct response to any lack of clarity in a claim is amendment, cf. G 1/24, Reasons 20, second sentence.
The board added that G 1/24 otherwise confirms most of the established case law, Reasons 10, including the principle of the primacy of the claims and referred to G 1/24, order, first sentence: "the claims are the starting point and the basis".
If the scope of protection, determined by the claim, cf. Art 84, is rendered unclear by the description allowing a different interpretation, hence a different scope of protection, the second compelling reason for adapting the description is Art 84.
The direct consequence is that the patent is non-compliant unless the description is purged and claim and description are on a par.
A great deal of the necessity of the adaptation of the description has thus already been dealt with by application of G 1/24. G 1/25 is there to possibly give indications as to how to remove remaining inconsistencies, like embodiments not any longer covered by amended, i.e. limited claims.
I therefore expect in all logic that T 56/21 and the like will end up in the Bermuda triangle of decisions which have become obsolete following a decision of the EBA.
@ The FRANDalorian-16.02.2026- 20.48.00 GMT
DeleteYou and/or Proof of the pudding are manifestly not prepared to accept the way the boards apply G 1/24. You still cling to your idea that the interpretation found in the description should be taken into account, if not prevail, prevail over that resulting from the plain wording of the claim.
As far as I am concerned there are two compelling reasons for adapting the description in case of an inconsistency: G 1/24 and Art 84.
The first is that, when assessing patentability under Art 52-57, cf. G 1/24, the prior art to be taken into account is either that resulting from
- the broad interpretation in the description in spite of a limited claim, or
- the broad interpretation of the claim in spite of a limitation in the description.
Consulting the description and deleting parts of it are indeed conceptually distinct steps, but if the applicant/proprietor wants the definition of the description to prevail, then the claim has to be amended accordingly.
This direction of convergence might not textually compelled by the EPC, but the boards are there to apply decisions of the EBA, and have made clear what is expected from the applicant/proprietor in their decisions following G 1/24.
All decisions applying G 1/24, do not deny the existence of two potential interpretive routes, one from the plain wording of the claim and one from the description. They simply refuse to be forced to do “interpretative summersaults”. It results that for the boards, in application of G 1/24, only one interpretative route is allowable, and this is that resulting from the claim wording.
The boards do not well know how to apply decisions of the EBA and that they are anything but plainly wrong in their decisions.
In Reasons 2.2.3 of T 0981/23, the board reminded that Art 84 requires claims to be clear, and added that the meaning of a claim feature should be clear for the person skilled in the art from the wording of the claim alone, and referred to G 1/04, Reasons 6.2. The correct response to any lack of clarity in a claim is amendment, cf. G 1/24, Reasons 20, second sentence.
In T 0981/23, the board added that G 1/24 otherwise confirms most of the established case law, Reasons 10, including the principle of the primacy of the claims and referred to G 1/24, order, first sentence: "the claims are the starting point and the basis".
If the scope of protection, determined by the claim, cf. Art 84, is rendered unclear by the description allowing a different interpretation, hence a different scope of protection, this represents the second compelling reason for adapting the description is Art 84.
The direct consequence is that the patent is non-compliant unless the description is purged and claim and hence description are on a par.
A great deal of the necessity of the adaptation of the description has thus already been dealt with by application of G 1/24. G 1/25 is there to possibly give indications as to how to remove remaining inconsistencies, like embodiments not any longer covered by amended, i.e. limited claims.
I therefore expect in all logic that T 56/21 and the like will not be the leading decisions in matters of adaptation of the description. I trust that the long line of case law requiring the adaptation of the description will be confirmed by the EBA.
Glad to see that your logic engines are up to their normal capacity: similar views + knowledge of bold / italics = definitely the same person.
DeleteOf course, you will have carefully considered and then ruled out as impossible the chances that:
- the similarity of views could be the natural result of the logicality and persuasiveness of the "anti-adaptation" arguments; and
- two commentators using bold / italics could be explained by each of them independently learning how to use such emphases in their comments (eg by looking as explanations such as in https://hamlette.blogspot.com/2015/08/how-to-use-italics-bold-underline-and.html).
For the record, I can confirm that I am most definitely not The FRANDalorian. I am not even sure that they and I share precisely the same views on claim interpretation. My own views align very closely with the approach to claim interpretation applied by the UK courts. Whilst you try to malign my views by equating them with a preference for a description definition prevailing over the claims, it is clear to me that you simply cannot be bothered to pay attention to why that represents a gross mischaracterisation of the views that I have actually expressed.
Trawling through the by now 174 comments it occurs to me that most are coming from the patent right-owning community and its lobbyists, and that perhaps the strongest argument in favour of mandatory conformity between what the claim defines as the invention and what the description describes as the invention is the public interest in having patent owners say not only how to work their invention but also declare, with all possible precision, what they hold out to the public to be their patent-worthy invention, and how far they think their monopoly rights should extend.
ReplyDeleteBut there's a problem with that, namely, that it is the courts that will decide ultimately what is the public interest, not the EPO. I seriously doubt that the UPC will give the public interest as much air time as it will to the court's user community and financiers.
Up to now, the EPO has been unique in having no court overseeing it and telling it where its duty lies. That has had many benefits, not least the world's best approach to obviousness. I must say, in relation to G1/25, that here is another opportunity to set a world-class standard on definition of the scope of monopoly rights, but will the EBA be able to seize the opportunity. I mean, how much influence over the upcoming division the UPC and its sponsors will have. Perhaps the upcoming oral proceedings in early May will give us some clues.
Dear Max Drei-18.02.2026-11.00.00 GMT
DeleteReply part 1
Your comment brings me once more to stress the specific position of the EPC and the EPO in the patent world.
The EPC has been built on the ruins of four earlier European patent convention drafts under what was called then the European Community, now the EU. Those four drafts would have led to a closed convention as it was only opened to EC contracting states.
The emerging PCT was seen as a threat by some European countries, and discussions about a European patent began anew. The breakthrough came with the proposal to have two treaties. It is known under the Haertel proposal.
The first, was meant to go up to grant and opposition, and ended up with the EPC and the EPO. Post-grant and post-opposition dealings of patents granted by the EPO, i.e. infringement and validity, were to be dealt with by national jurisdictions in the EPC contracting states.
The EPC was conceived as an open convention as it included many countries which were at the time not member of the EC, e.g. SE and AT. It was also open to CH and TR. TR was a contracting state of the Institut International des Brevets (IIB), the predecessor of DG1 at the opening of the EPO.
The second was dead borne. The Luxemburg Convention of 1975, was to become a unitary patent with a unitary effect in all EC/EU contracting states. It ended like the previous four drafts in oblivion as the EC/EU contracting states could not agree on the language regimen and forum shopping.
The UPC can actually be seen as the successor of the dead Luxemburg Convention, with the proviso that not all EU contracting states signed and/or ratified the UPCA. The UPC does however do everything to extent its reach to non-UPCA contracting states.
Dear Max Drei-18.02.2026-11.00.00 GMT
DeleteReply part 2
You are right to notice that “the EPO has been unique in having no court overseeing it and telling it where its duty lies”. Neither national jurisdictions nor the UPC have any right to tell the EPO and its boards of appeal how to apply and interpret the EPC. This is the exclusive right of the EBA. In this respect, the UPC has no better position than national jurisdictions dealing with validity and infringement in their respective territories.
In a fully integrated system like it exists in the contracting states of the EPC, a national court can iron out, inconsistencies which might have subsisted at grant. The EPO is responsible towards its contracting states as a whole, and neither towards national jurisdictions nor towards the UPC. The UPC cannot impose its views on the EPO although all its promoters and supporters would like it to be so.
When in lots of Amicus Curiae for G 1/25 it was found that the EPO is an outlier, this is absolutely correct, but they grossly ignore the facts which have been shaped when the EPC and the EPO were set up.
It is the duty of the EPO to grant a patent which has to be clear, novel and inventive, and in which the patent as granted does not allow the proprietor to change its interpretation as he thinks fit.
With G 1/24, and the way the boards apply it, a first series of fundamental inconsistencies, has been, or is on the way of getting removed. G 1/25 should be there to remove the remaining ones.
It is clear that different interpretations of the validity as well as of the scope of protection should not occur between the EPO and the UPC. G 1/24 and the way the boards apply it, a first step has been done in this direction. G 1/25 should be the continuation of the move. By delivering a standard product free of inconsistencies, the EPO does in no way hinder national jurisdictions nor the UPC to apply Art 69 and the Protocol.
You are right to “doubt that the UPC will give the public interest as much air time as it will to the court's user community and financiers”. Be it only the different fee for revocation or a counter-claim for revocation compared to that for infringement, makes abundantly clear who should be the beneficiaries of the UPC.
The more different interpretations possible of a granted patent, the more possibilities to create Angora cats are opened. This is to the great benefit of all the internationally active lawyer firms profiting directly from the UPC, as it means more billable hours.
Daniel, on reading your thoughts in reply to mine the one that jumped out at me was the choice the UPC made, to charge a higher fee to enquire into validity than into accusations of infringement.
DeleteThe proponents of the UPC will presumably retort that an enquiry into validity takes more time than one into infringement. In the great majority of cases, that will be true, right?
But another argument occurs to me. If the UPC sees enquiries into validity as a costly and lengthy nuisance, then it is signalling that the EPO should shoulder the burden of checking for claims that are not in good shape for asserting against infringers.
In other words, the UPC has chosen to declare and confirm that it should be the EPO that should take the leading and primary role for making sure patents in Europe are in optimal shape for the court to decide, fast and clean, what is an infringement of a claim, unencumbered by the burden of an enquiry into validity (for which its non-technical judges are ill-equipped anyway). In G1/25, the EBA has a chance to show the world that the EPO is not frightened to "step up" and carry that responsibility.
Dear Max Drei-22.02.2026-11.50.00 GMT
DeleteI am not convinced that an enquiry into validity takes more time or is more difficult than one into infringement. Be it in infringement and/or in validity before a civil court, first the skilled person has to be determined and the claim has to be interpreted. Why should this be more complicated in terms of validity than in terms of infringement? One the claim is interpreted, the infringement has to be decided with the same accuracy.
To me the reason why the UPC adopted this difference in fees, is that the UPC was first conceived to help proprietors. This is confirmed when one sees how far the long arm of the UPC tries to extend.
When you see what happens for instance at the CFI-CD Milan, not allowing requests in view of new objections, and requiring the counterclaimant in validity to limit its attacks, see the corresponding publications on my blog, there is no reason why the fee for validity should be nearly double than that for infringement.
In this respect, I agree with you that with G1/25, the EBA has a chance to show the world that the EPO is not frightened to "step up" and carry its responsibility. By doing so it will just do what the EPC requires it and the EPO to do.
When applying G 1/24 by removing any interpretative inconsistency between claim and description (in a bidirectional way), I hope that the EBA will continue on the route it took, by requesting all remaining inconsistencies between claims and description to be removed under G 1/25.
Max Drei: your contribution here is, characteristically, the most intellectually honest articulation of the pro-adaptation case. You have identified what seems to be the strongest argument (the public interest in clarity) and immediately acknowledged its limitation: it is the role of the courts, not the EPO, to determine the public interest having regard to the full real-world context reaching far beyond issues confined to the single issue of patentability.
DeleteYour observation about the UPC fee structure, and what it might say about implicit delegation of quality control to the EPO, is astute. But the EBA's task in G 1/25 is not to decide what the EPO should do (in the minds of at least some observers) as a matter of institutional design. It is to determine what the EPC requires the EPO to do. With respect, the legislative history of previous failed attempts to create a unified system is of no more than historical relevance when considering how to interpret what was actually enacted.
DXThomas is right that the bifurcated structure of the European patent system is unique and consequential. But the consequence is that the EPO's mandate is defined by the EPC, and only by the EPC. Any institutional gaps between the EPO and the courts are there by design. If the contracting states wish to confer a broader mandate to close the gap, they may amend the Convention. The EBA cannot and should not do it for them.
@DXThomas (18 February, 14:44 and 19 February, 17:32)
ReplyDeleteI am not Proof of the Pudding. We are different people. Your alleged « evidence » for this theory (that I and Proof both know how to use bold text on Blogger and both find your arguments unpersuasive) would not survive even the most generous application of the "directly and unambiguously derivable" standard.
But, anyway, the identity question is a distraction and I'm not going to stoop to your level of playing the man and not the ball. What matters is the substance, and on substance your recent posts contain a conflation that goes to the heart of G 1/25. For the sake of the wider readership of this thread we should not let this pass unchallenged, even if you remain vociferously unconvinced and prefer to repeat your conclusions rather than defend them.
You state that "G 1/24, dealing with one inconsistency, has paved the way for G 1/25 dealing with the remaining ones." This elides a critical distinction. G 1/24 addresses a claim interpretation problem: where the description provides an interpretation that diverges from the plain meaning of the claim, the claim must be amended if the applicant wishes that interpretation to be taken into account. The operative direction of amendment is into the claim. This is a normal application of Article 84's clarity requirement, directed at the claims.
G 1/25 asks a different question entirely: whether the description must be purged of material that does not correspond to the claims as amended. These are conceptually distinct operations. Consulting the description (G 1/24) and deleting from the description or marking aspects as "not claimed" or "not part of the invention" (which is what G 1/25 is asked to decide) are not two steps in a single logical sequence. If they were, the referring Board would not have needed to make the referral.
G 1/24 already requires the claims to be amended wherever the description would alter their interpretation. Once that requirement is satisfied, any remaining difference between description and claims is, necessarily, a difference that does not alter claim interpretation. The skilled person can identify it and set it aside. Put simply: G 1/24 already provides the interpretive tool to resolve any inconsistency that affects meaning. What G 1/25 is asked about is the residue: inconsistencies that, by definition, do not alter interpretation. G 1/24 demonstrates that only those divergences affecting claim interpretation require intervention; and that intervention is directed at the claims, not the description.
(continued in a second comment)
@ The FRANDalorian-20.02.2026 at 22.41.00 GMT-Part 1
DeleteWhether you and/or Proof of the Pudding are one or the same person is actually irrelevant. Hiding behind a pseudo allows you to pretend whatever you want. I noticed a similarity and we can leave it at this.
For the sake of the wider readership of this thread, I will not (I will also avoid using the royal “we”) not let this pass unchallenged. What I have said is logical and I will not get tired to repeat it. May be one day the penny will drop.
It is correct that G 1/24 and the decisions applying it, say that there cannot be two interpretations, one based on the claim wording and one based in the description. There is nothing vociferous about repeating this statement. Boards have said that the claim has to be amended accordingly, but the description can be amended as well. At the end of the day, both are then on a par, and there is no wriggle room left for the proprietor in post-grant or post-opposition procedures. .
If this does not mean adaptation of the description to the claim or the other way around, what else could be understood under alignment of the claims to the description or vice versa? The only adaptation left, is then to decide which parts of the description are "not part of the invention" (which is indeed what G 1/25 is asked to decide).
You claim that specifying which parts of the description are not any longer covered by the claims does not alter the interpretation of the claims and that the skilled person can identify it and set it aside. I would strongly disagree with this statement. Why would a proprietor insist on keeping an unamended description by letting the impression survive that parts of the original disclosure are still covered by the claims if it would not think it could gain some benefits from it? This is why my position is that what has been started with G 1/24 ought to be achieved by G 1/25.
You allege that such inconsistencies, by definition, do not alter interpretation. I cannot share this point of view for the reasons given above. If you said A you have to continue and say B, i.e. remove any inconsistencies left between claims and description.
@ The FRANDalorian-20.02.2026 at 22.41.00 GMT-Part 2
DeleteQuestion: is there a file wrapper estoppel in the European patent system? The answer is plainly no. If this should be wished by the contracting states, this would indeed need a Diplomatic Conference. This is an occurrence I cannot see coming in the next future as Art 4a EPC is a dead borne ruling. We never saw a conference of ministers of the contracting EPC states.
Art 84 is as it stands and will not be changed for a while. Next to the aspect clarity there is the notion of support. How can something not covered by the claims support the claims? I fail to see it and I never got a sensible reply on this point.
You and your colleagues might call the adaptation of the description a poor man’s file wrapper estoppel, but this is your interpretation and you are free to think it. For my part, I do not have to discuss a notion which is foreign to the letter and the spirit of the EPC, but an interpretation of the EPC by a limited circle of persons.
The only body allowed to interpret the EPC and the EBA, and I cannot see the latter coming up with such an interpretation in the next future, even when dealing with G 1/25.
To me Art 84 and the respect of it, is by no means a purely formal or procedural ruling. Considering Art 84 to be a purely procedural ruling shows a profound misunderstanding of its function and purpose. I have always been of the opinion that Art 84 is as fundamental a ruling as novelty and inventive step are. Art 84 is there to require that the scope of protection to be defined as clearly as possible, and not there to leave wriggle room to patent holders, and a maximum of billable hours to lawyers acting for them in post-grant or post-opposition procedures.
Art 84 is there to be respected by the users of the European patent system and by the EPO acting in the name of the EPC contracting states. The statutory text of Art 84 has seen its application be subject of a policy decision, provoked by the pemetrexed case. This is quite normal, or would like that Art 84 to be merely ignored?
It is only if the EBA decides that the policy decision of the EPO is manifestly wrong, that it will be changed. It will certainly not be changed due to some vociferous opinions touted by people wanting the scope of protection to be as fuzzy as possible.
Comment 2 of 2
ReplyDeleteYou have not answered the file wrapper estoppel point either. Anon Y. Mouse put to you on 15 February a precise question arising from your own words. On 29 January you stated: "The EPO's description adaptation practice really does not seek to achieve a poor man's file wrapper estoppel." On 12 February you stated: "the day the last part of Art 84 will be removed from the EPC and file wrapper estoppel introduced, then I will agree that the description will not have to be adapted to the claims." As Anon Y. Mouse observed, if the introduction of file wrapper estoppel would render adaptation redundant, the two must serve the same function. If two mechanisms are mutually substitutable — if either one alone is sufficient and both together are redundant — then they serve the same function. This is elementary logic. Adaptation is therefore de facto prosecution history estoppel imposed through Article 84. This is precisely the "poor man's file wrapper estoppel" that you denied. You have not reconciled these statements.
The logical consequence is significant: the European patent system deliberately omits prosecution history estoppel. German case law has expressly held that statements made during prosecution are not binding on claim interpretation. Introducing such a mechanism would require legislative action. This is a point you yourself acknowledged when you referred to the need for a diplomatic conference. The EPO cannot achieve through procedural requirements under Article 84 what the contracting states chose not to enact.
On Pemetrexed, you stated on 18 February that "the strict approach has been originally provoked by the attitude of the applicant in the famous 'pemetrexed' case." This is a significant concession. It frames the current practice not as an interpretation compelled by the statutory text of Article 84 but as a reactive policy response to a single applicant's conduct. The EBA is being asked what the actual law requires. It is not being asked what policy-level response a particular case may have "provoked."
The Pemetrexed litigation also illustrates the futility of the adaptation exercise. The same patent specification produced contradictory outcomes across Europe at first instance. The decisions upon appeal in GB, NL, FR, DE, IT, etc. tended to converge, but for very different reasons. Those divergences reflect structural differences in national practice, not differences in the content of the description. It is far from clear how a "strict" adaptation could have harmonised their reasoning. And if it would have, doesn't this concede the point that the real goal is to function as a form of file wrapper estoppel?
Finally, I asked you on 11 February to cite the EPC provision establishing the EPO's duty to "grant or maintain patents which are not open to multiple possible interpretations." You responded with policy arguments but no citation. The new Guidelines do not fill this gap — as you have yourself acknowledged, they are not a source of law and do not bind the Boards of Appeal, still less the EBA. The question remains open. If the duty you describe exists, it should be possible to point to where the EPC establishes it. Please identify the specific provision of the EPC — not Guidelines, not trends in Technical Board decisions, not policy considerations — that establishes the EPO’s obligation to refuse or revoke a patent solely because the description contains embodiments outside the scope of an amended claim. If it cannot be pointed to, then we are in the territory of policy preference, not legal requirement. The EBA's task is to interpret law, not to set policy.
Excellent point about the change of EPO practice. If the interpretation of Art 84 EPC was always as the EPO now insists, why did EPO practice look so different prior to 2018?
DeleteArguments which assert that the change was a reaction to the pemetrexed case are based upon a fundamental misconception. That is, such arguments miss the point that precisely zero national courts found that the different (salt) forms of pemetrexed infringed the claim in a literal sense. Thus, the manner in which the downstream courts interpreted the claim was entirely consistent with the EPO's interpretation.
The different results in pemetrexed therefore all stem from the application of Article 2 of the Protocol to Art 69 EPC, i.e. to a provision that all Boards of Appeal have consistently agreed is simply not the EPO's concern. Thus, if there is any truth to the assertion that the change of policy was a reaction to the pemetrexed case, then this represents a backhanded acknowledgement that the EPO is not only overstepping the limits of its authority but is also seeking to directly intrude on the assessment of equivalents (ie a matter falling squarely within the exclusive competence of the downstream courts).
If this is a stupid question, Pudding, prompted by the comments that have come in over the week-end, you (and others) will I am sure be quick to tell me. When, post grant, an EPO panel is obliged to decide under Art 123(3), does it need to decide what is an equivalent under Art 69, EPC? Is Art 69 EPC none of the EPO's business? Or is it most definitely the EPO's business and if so, does it determine how the claim shall be construed?
DeletePut another way, shall it construe a claim not only i) as if the infringer had never been born, but also ii) taking no notice whatsoever of what Art 69 EPC commands?
@ Proof of the pudding-23-02.2026 at 12.18.00 GMT
DeleteYou seem to confuse cause and effect. Independently of national courts having decided infringement by equivalents according to Art 2 of the Protocol, this does not change the fact that the applicant at the time tried to give its claim a broader interpretation than that resulting from the plain wording of the claim witch was limited to pemetrexed DISODIUM. No other salt has ever been envisaged at the time of filng.
Under G 1/24, such an inconsistency would not have survived. Here we have not even yet arrived at G 1/25. The examiner had made clear that the description should be adapted, but in the end he gave up. In the last applicant's entry before the actual grant one reads:
“The Applicant seeks to draw a distinction between subject matter which is relevant to the invention which is indicated as being that to which "the present invention generally relates" and "the subject matter provided by the invention" which is the subject matter claimed.”
In the description one reads “Thus, the present invention generally relates to a method of reducing the toxicity associated with the administration of an antifolate to a mammal comprising administering to said mammal an effective amount of said antifolate in combination with a methylmalonic acid lowering agent”.
The claim reads: “Use of pemetrexed DISODIUM in the manufacture of a medicament for se in combination therapy….”
The claim was thus limited to pemetrexed disodium, but the applicant considered without any proof that a much broader interpretation was justified. He did this without any proof that the invention was generally applicable and not limited to pemetrexed disodium. At the time of filing, the applicant had no proof of the general application of its invention.
Whether national courts concluded to infringement by equivalents, has thus nothing to do with the inconsistency between the generalising statement in the description and the limitation in the claim.
Contrary to your allegation, the EPO is thus in no way overstepping the limits of its authority and by no means did the EPO directly intrude on the assessment of equivalents.
With the generalising statement in the description there was unmistakably an inconsistency between the interpretation resulting from the plain wording of the claim and that suggested in the description. Applying G 1/24 (and possibly later G 1/25) has thus the purpose of avoiding the proprietor to later come with an Agora cat.
The EPO was thus right to tighten the screws after this incident. The generalising statement in the description was not supported by the core of the description and the claim limited to pemetrexed disodium.
This was even more justified as the examiner was accused of misinterpreting Art 123(2). At the EPO added matter is considered under novelty and equivalents belong to inventive step.
I have a question about the EPO’s post-G 1/24 case law on claim interpretation. That is, with which of the following principles from UK case law is that EPO case law consistent?
ReplyDelete“(i) The first overarching principle is that contained in Article 69 of the European Patent Convention.
(ii) Article 69 says that the extent of protection is determined by the claims. It goes on to say that the description and drawings shall be used to interpret the claims. In short the claims are to be construed in context.
(iii) It follows that the claims are to be construed purposively – the inventor's purpose being ascertained from the description and drawings.
(iv) It further follows that the claims must not be construed as if they stood alone – the drawings and description only being used to resolve any ambiguity. Purpose is vital to the construction of claims.
(v) When ascertaining the inventor's purpose, it must be remembered that he may have several purposes depending on the level of generality of his invention. Typically, for instance, an inventor may have one, generally more than one, specific embodiment as well as a generalised concept. But there is no presumption that the patentee necessarily intended the widest possible meaning consistent with his purpose be given to the words that he used: purpose and meaning are different.
(vi) Thus purpose is not the be-all and end-all. One is still at the end of the day concerned with the meaning of the language used. Hence the other extreme of the Protocol – a mere guideline – is also ruled out by Article 69 itself. It is the terms of the claims which delineate the patentee's territory.
(vii) It follows that 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.
(viii) It also follows that where a patentee has used a word or phrase which, acontextually, might have a particular meaning (narrow or wide) it does not necessarily have that meaning in context.
(ix) [Omitted in view of the Supreme Court’s judgment in Actavis v Lilly [2017] UKSC 48]
(x) On the other hand purposive construction can lead to the conclusion that a technically trivial or minor difference between an element of a claim and the corresponding element of the alleged infringement nonetheless falls within the meaning of the element when read purposively. This is not because there is a doctrine of equivalents: it is because that is the fair way to read the claim in context.
(xi) Finally purposive construction leads one to eschew the kind of meticulous verbal analysis which lawyers are too often tempted by their training to indulge".
My current feeling is that the EPO’s post-G 1/24 case law is arguably not consistent with principles (iii), (iv), (viii), (x) and (xi) (and perhaps also not quite as firm as it should be on principle (vii)). But what do others here think?
Another question: is it acceptable in view of Reasons 16 of G 1/24 for the EPO’s claim interpretation approach to diverge from any of the above principles? If so, which of those principles can safely be discarded (and why)?
@ Proof of the pudding-23.02.2026-17.22.00 GMT
DeleteFor a start, there is no reason why G 1/24 and the way it is applied by the boards ought to be consistent with any of principles from UK case law on interpretation of the claims. UK case law is merely one type of case law among the 38 types of case law applicable in the EPC contracting states.
The purposive construction of the claims in the “Catnik” style, is one way to interpret claims. This is manifestly not the way that the boards applying G 1/24 interpret claims. Whether the description allows a more limited or a broader interpretation compared to that resulting from the plain claim wording, the boards have made clear that what matters is what the independent claims says.
The scope of protection should not depend from what the applicant/proprietor has hidden in the description, but should be derivable directly and unambiguously for the skilled person from the plain claim wording. Hence the necessity to align claims and description.
It should not be forgotten that in G 1/24, the EBA made clear that Art 69 does provide a legal basis for interpreting the claims as far as validity is concerned.
EPO’s post-G 1/24 case law does not have to be consistent with principles (iii), (iv), (vii), (viii), (x) and (xi) ) when it comes to validity. This is right so, as UK case law is not the leading case law as far the EPO is concerned.
Purposive construction of the claim is to be applied in infringement when applying Art 69 and the Protocol. When it comes to validity the situation has to be clear, hence purposive construction is not to be applied. This is the domain of post-grant or post-opposition courts.
You seem still not having accepted the way the boards apply G 1/24. The difference is that they decide in matters of validity, not you or your colleagues.