The comments space for the last IPKat article on claim interpretation/description amendments appears to have maxed out. Sorry everyone, but maybe it's for the best. However, fear not, we now have a new decision to restart the debate! The decision in T 439/22 takes us back to our old friend, the definition of "gathered". Applying G 1/24, the Board of Appeal in the case is the second to find that a definition in the description can be used to broaden the scope of the claims (compare T 1849/23). The Board of Appeal in this case also goes one step further and, contrary to all previous EPO interpretations of G1/24, reasoned that description definitions can also narrow the claims. Even more significantly, the Board of Appeal also found that deleting the definition would unallowably extend protection under Article 123(3) EPC, pouring lighter fluid on the description amendment debate.
G1/24 and claim interpretation case catch-up
The appeal in T 439/22 concerned the Philip Morris patent EP 3076804, relating to a heat-not-burn device. The "aerosol-forming substrate" (e.g. tobacco) in the device was defined as comprising a "gathered" sheet of aerosol-forming material. The Opposition Division (OD) rejected the opposition, finding the invention novel over the cited prior art. The cited prior art disclosed a tobacco sheet that was spirally wound. The OD found that a gathered sheet would normally be understood as folded and convoluted and specifically disregarded a broader definition found in the patent description. This definition stated that the term "gathered" denoted that the sheet is "convoluted, folded, or otherwise compressed or constricted substantially transversely to the cylindrical axis of the rod". Under this definition, the term would have encompassed the prior art.
 |
| Contemplating definitions |
On appeal, the Board of Appeal thus had to determine if the spirally wound tobacco sheet in the prior art anticipated the gathered sheet of claim 1. The Opponent argued that the broad definition in the description could not be ignored and clearly encompassed the wound sheet of the prior art. The Patentee submitted that the wording of the claims should have primacy and that the term gathered should be interpreted according to its common technical meaning in the tobacco industry, which excludes spirally wound sheets.
The subsequent referral G 1/24 aimed to resolve the question of claim interpretation in view of the description. The EBA ordered that the description should always be consulted. However, as subsequent interpretations of G1/24 from the Boards of Appeal have found, "consulting" definitions in the description does not require the definitions to be actually "used" to interpret the claims. Indeed, up until the present case, the Boards of Appeal have universally found that the G 1/24 order to "always consult" the description does not require adopting a restrictive definition that narrows otherwise clear claim language (IPKat). Notably however, the Board of Appeal in T 1849/23 found that the description can be used to broaden the meaning of a claim beyond its literal reading. The present case is a second test of what G1/24 means for broadening as opposed to narrowing definitions.
Interpretation of the claims (G 1/24 and Article 69 EPC)
The Board of Appeal in T 439/22 found that the solely claim-based interpretation adopted by the Opposition Division was incorrect in light of G 1/24. The Board of Appeal particularly emphasised that claim interpretation is a unitary process involving both the claims and the description. In its reasoning, the Board of Appeal stated:
"in interpreting the language used in a claim, 'consulting', 'referring to', 'using' and 'taking into account' the description and figures are synonyms for the act of deriving the necessary information from the patent as a whole to understand what meaning a person skilled in the art would attribute to the terms used in the claim".
Interestingly, the Board of Appeal also explicitly disagrees with the reasoning in all of the Board of Appeal decisions relating to narrowing definitions that found that "consulting" does not equate to use:
"The Board sees no reason to assume that the expression "to consult" used in the order of the Enlarged Board should be given a meaning different from the expression "to use" found in Article 69 EPC. The term "consulted" is used by the referring Board itself in the second question referred to the Enlarged Board and was also previously widely used in the relevant case law (see e.g. T 400/94, Reasons 3.3.2; T 860/08, Reasons 1.2; T 2284/09, Reasons 2.3; T 1964/10, Reasons 3.2; T 2758/18, Reasons 3.9; T 1735/19, Reasons 2.3)" (r. 2.3)
Compare, for example, the following Board of Appeal decisions:
- T 1561/23: "G 1/24 does not even explicitly require that the definition of a term from the description must be used for the interpretation of a claim" (r. 2.10, machine translation from the German) (IPKat)
- T 2027/23: "a claim should not be interpreted, based on features set out in embodiments of an invention, as having a meaning narrower than the wording of the claim as understood by the person skilled in the art" (r. 3.5.6) (IPKat)
- T 412/24: "a sensible claim construction cannot be reinterpreted on account of the description being 'consulted' or 'referred to'" (r. 3.1.1)
- T 2034/23: "consultation of the description does not, however, mean that claimed features must always be construed as narrowly as any specific embodiments of the invention found in the description" (r. 3.6.2).
In the present case, the Board of Appeal, by contrast, found no reason to disregard the explicit definition provided by the Patentee in the specification. The Board of Appeal was therefore convinced that the spirally wound tobacco sheet of the cited prior art equated to the gathered sheet of claim 1 when interpreted according to the description. The Board of Appeal thus concluded that the subject matter of claim 1 as granted lacked novelty over the prior art.
Broadening, narrowing or contradictory?
We therefore have a second clear interpretation of G 1/24 from the Boards of Appeal that broadening definitions in the specification can be used to broaden the scope of the claim language. We also have divergence amongst the Boards of Appeal as to the correct interpretation of "consult". Instead, the Board of Appeal in the present case makes a conscious effort to align with the UPC (r. 2.3). However, significantly, the Board of Appeal in this case also moved away from the language of contradictory definitions, reasoning that "the definition of "gathered" given in paragraph [0035] does not contradict but rather encompasses the commonly accepted meaning of the term "gathered sheet", namely a sheet that is folded and convoluted to occupy a three-dimensional space" (r. 3.5.2). In other words, for the Board of Appeal, the description definition just added to the interpretative milieu in which the skilled person would understand the claims in view of the disclosure as a whole. The question therefore remains as to the effect of a clearly contradictory definition. Interestingly, a recent Board of Appeal decision found that contradictory definitions can both lead to a lack of clarity and sufficiency (T 878/23, IPKat).
Practical consequences
To be safe, therefore, the best way to broaden or narrow a claim has always been through use of the claim language itself, and not via the backdoor of broad description definitions (IPKat). This principle holds true for other jurisdictions outside of Europe, including the US, where narrowing definitions may also be read into the claims (IPKat). Why the patentee in the present case chose to include the definition of "gathered" is lost to the mists of time, as it doesn't appear to serve much purpose in the case and appears quite academic as a linguistic dispute. The Board of Appeal interestingly notes that "the fact that the patent defines this term in the description is a clear indication that, at least for the patent proprietor when drafting the application, 'gathered sheet' did not have a meaning that is so commonly accepted and well established that each further explanation would be superfluous" (r.3.5.1). More reasons, if any were needed, to strike out unnecessary definitions. It also saves you on excess page fees.
Description amendments extending protection
The issue of claim interpretation is of course highly related to the controversial topic of description amendments, now subject to its own referral in G 1/25 (IPKat). The present case adds an interesting flavour to the debate.
Faced with the possibility of revocation of the main request, the Patentee filed an auxiliary request where the paragraph containing the offending definition was deleted from the description. The deleted paragraph read as follows:
"As used herein, the term 'gathered' denotes that the sheet of tobacco material is convoluted, folded, or otherwise compressed or constricted substantially transversely to the cylindrical axis of the rod."
The Opponent argued that this amendment altered the interpretation of the claims and extended the scope of protection by removing language that narrowed the meaning of the claims. The Board of Appeal was not persuaded by the Patentee's argument in response that the rod-shaped geometry specified in the paragraph was implicit in the claims. The Board of Appeal particularly found that "Claim 1 contains neither explicit nor implicit features that necessarily limited the claimed aerosol-forming article to a rod-shaped configuration". Without the definition, the Board of Appeal reasoned, the claim no longer required the tobacco sheet to be "compressed or constricted substantially transversely to the cylindrical axis of the rod". Consequently, the claim could encompass articles that are not rod-shaped or sheets folded not transversely but along the axis, like a bellows. In other words, the Board of Appeal also considered it acceptable to use the description to narrow the meaning of the claims, contrary to all other Board of Appeal decisions to interpret G 1/24 thus far.
In its reasoning, the Board of Appeal further explained that the "skilled person will read terms in the claim in the sense of the definition, taking into account both the broadening and limiting aspects". By deleting the definition, the Board of Appeal was of the view that Patentee thus improperly sought to consider only the broadening aspects while disregarding the limiting ones. As such, the Board of Appeal reasoned, this removal of geometric constraints resulted in a scope of protection wider than that conferred by the patent as granted. The auxiliary request was therefore found to violate Article 123(3) EPC.
Final thoughts
Further reading
- Beware of boilerplate: Practical lessons for patent drafting from G 1/24 (Claim interpretation) (Apr 2025).
- EBA decides G 1/24 on claim interpretation: The description should always be consulted (Jun 2025).
- First interpretation of G 1/24 (the description must always be consulted) (T 1561/23) (Jul 2025).
- G 1/24 in the spotlight: Description definitions do not override clear claim language (T 1999/23) (Jul 2025).
- First use of G 1/24 to broaden clear claim language (T 1849/23) (Nov 2025)
Reviewed by Dr Rose Hughes
on
Saturday, February 28, 2026
Rating:
Follow the IPKat on LinkedIn 
![[Guest Post] Of Snakes and Artistic Value: Bologna IP Court Slithers Through Design, Shape Marks and Copyrights in a post-Cofemel landscape](https://blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEicgpLQWvJLkDaC4vfUlO-Ik8JCdDOCbRh2kN3Q4VOL28iO-_2pD0NkQcOpWwZlzn8w1orqz4c4-XvdGOxiL_gxlqlc-FkmnN6t1ifQz_C1XC3aiSOo5wsE_qC6vl173F4-c1307-ST2SEscrK9TBBJ0BhRLo9Yy7U9j41Ro2Hp6jbnn44oQhYI/s72-c/Bild1.jpg)
![[Guest post] A tale of David and Goliath in Gangnam style?](https://blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEi4KD1ViQQvuwd-EMW2xoCHKGROxBbTofQ6KO5mL6Wd4ymudDMHNN2IjtVodb0HrZ4Y86B8vkflbB5LN4_ECWDNfI8FzboPlLvLYN7aFKDtSw9UKQZmxLyEjrKWg5hRroPDNm15zsfKgcpcd4ARNpYHqKwGKS6QIUgLsL7noeMGDseE5RQxzsC4Ag/s72-c/IMG_0338.heic)
Posts
As what prompted the Board to issue this reasoning at this time, my suspicion is that it sees it as helpful to the EBA in its deliberations in G1/25. It seeks more guidance from the EBA, what "consult" means. The time window to influence the EBA is now quite tight . Its conscience is clear because the applicant here was manifestly the deliberate author of its own Angora cat, its own misfortune. We are seeing here the difference between a jurisdiction with binding precedent and one without.
ReplyDeleteAs to the other thread, with 209 comments, clicking on "Load more" can reveal the newest postings. Amongst them is a reference to T422/91 (Reasons 3) for its statement that the EPO is not going to get into the task of claim construction reserved for the courts. OK, even if the EPO is not going to do that, it still has to decide, from time to time, under Art 123(3), whether the scope of protection (are we referencing Art 69 EPC here?) has been expanded by a post-grant amendment.
ReplyDeleteI suppose that determining a "relative" quantity (which of two texts is "wider" than the other) is possible even if setting the "absolute" scope of protection for any given claim is not. But I remain sceptical, that even the "relative" assessment is something the EPO is willing to embark upon, or is competent to undertake. I mean, the scope includes equivalents, right, and what is "equivalent", in any given case, beforte or after the post-grant text change?
@ Anonymous-28.02.2026-18.08.00 GMT
DeleteAs search for T422/91 in the 209 comments on G 1/25 has not given any result. Independently of this, I fail to see in T422/91 (Reasons 3) either a direct or an implicit statement that the EPO is not going to get into the task of claim construction reserved for the courts.
Reasons 3 in T 422/91 relate actually to the clarity of functional features in a claim. A functional feature meets the requirements of Art 84 if the technical means capable of performing that function are readily available to the skilled person on the basis of his technical knowledge or in the patent specification, whereby it is sufficient that a single example is disclosed. This has nothing to do with getting into the task of claim construction reserved for the courts.
That the board admitted the opponent’s late objection under Art 123(3) is somehow surprising. I am of the opinion that deletion of § [0035] does actually boil down to an intermediate generalisation, as the claim is now covering items which are not any longer disclosed, and hence would also be objectionable under Art 123(2).
Having seen how previous decisions applying G 1/24 have interpreted the decision, I am not convinced that G 1/24 implies a holistic approach as asserted by the present board. The holistic approach adopted by the bord in T 439/22 allows exactly what other boards have queried: the possibility of an applicant/proprietor to change the interpretation of the claim as he thinks fit.
In accordance with previous decisions applying G 1/24, the remedy was not deletion of § [0035], but to bring the broad interpretation from the description into the claim. But this would have led to the same result, claim 1 as granted lacks novelty.
The scope of protection might encompass equivalents, but this does not mean that equivalents can be brought in the description during examination or opposition. If equivalents were not originally disclosed, their introduction after filing would lead to added matter.
Equivalents are found in Art 2 of the Protocol on interpretation of Art 69, but without any definition of what can be considered “equivalent”. During the Diplomatic Conference of 2000, the EPO had proposed a definition of “equivalents”. The participating states did not agree with it.
Very sorry, DXT. The case I intended to draw attention to is T422/19 (and NOT T422/91).
DeleteVerry sorry Anonymous-28.02.2026-18.08.00 GMT+02.03.2026-16.06.00 GMT, but a search in Advanced Search in board’s decisions under T 422/19 gives the following result: We found 0 results for "T422/19". It would however be interesting to get the correct number.
DeleteTo quote from T 0422/91, Reasons 3:
Delete"Mit dem Schutzbereich eines Patents befaßt sich das EPA im Einspruchsverfahren nur im Rahmen des Artikels 123 (3) EPÜ. Darüber hinaus sind Interpretationen des Schutzbereichs eines Patents nicht Aufgabe des EPA, sondern gemäß Artikel 64 und 69 EPÜ der für Patentverletzungsverfahren zuständigen nationalen Gerichte. Das EPA wird natürlich mit aller Sorgfalt die Fassung der gewährbaren Patentansprüche feststellen, da deren Inhalt den Schutzbereich des europäischen Patents nach Artikel 69 (1) EPÜ bestimmen, und - wenn erforderlich - klarstellen, wie in den Ansprüchen verwendete Begriffe verstanden werden sollen; darüber hinaus hat aber das EPA keine zusätzlichen Interpretationen über den zukünftigen Schutzbereich des erteilten oder aufrechterhaltenen Patents zu geben, an die im übrigen auch kein nationaler Verletzungsrichter gebunden wäre”.
My understanding of this quote is that the Board is saying that determining the extent of protection provided by a patent is not the responsibility of the EPO, but is instead the responsibility of the downstream courts.
@ Proof of the pudding-02.03.2026 at 17.15.00 GMT
DeleteReasons 3 of T 0422/91 reads
3. Klarheit des Anspruchs
Nach ständiger Rechtsprechung der Beschwerdekammer erfüllt die Definition eines technischen Merkmals durch seine Funktion die Erfordernisse des Artikels 84 EPÜ, wenn die für diese Funktion in Betracht kommenden technischen Realisierungsmittel dem Fachmann aufgrund seines Fachwissens oder der Patentbeschreibung ohne weiteres zur Verfügung stehen. Dabei genügt es normalerweise, wenn in der Patentbeschreibung ein einziges Beispiel offenbart ist. Da diese Bedingung im vorliegenden Fall erfüllt ist, genügt der Anspruch 1 den Erfordernissen des Artikels 84 EPÜ.
Die vorliegende Erfindung befaßt sich mit der Lösung einer bestimmten im Zusammenhang mit Fenstern auftretenden Aufgabe, die darin besteht, den Bereich der Spaltöffnungen eines Flügels zu erweitern. Es ist deshalb hinreichend, im Patentanspruch lediglich diejenigen Merkmale deutlich anzugeben, die für die Lösung dieser Aufgabe wesentlich sind. Daß die erfindungsgemäße Lösung an sich unter bestimmten Umständen zu einer Fehlbedienung des Fensters führen kann, ist somit als Argument gegen die Aufrechterhaltung des Streitpatents unbeachtlich.
Im Anspruch 1 ist von einer Rückhalteeinrichtung die Rede. Eine Einrichtung kann aus einem oder mehreren Teilen bestehen, so daß der Begriff "Rückhalteeinrichtung" nicht ausschließt, daß verschiedene Rückhalteteile dieser Einrichtung die beanspruchte Funktion erfüllen. Es besteht somit kein Widerspruch zwischen dem Anspruch 1 und der Beschreibung.
I fail to see here any reference to claim interpretation. Do you have another copy of T 0422/91 than that resulting from a query in “Advance search in bords of appeal decisions”?
Even if, arguendo the decision says that that claim interpretation in post-grant procedures is none of the business of the EPO, G 1/24 has made clear that Art 69 and the protocol cannot be used when assessing patentability under Art 52 to 57.
What the EPC requires, cf. Art 84, is that the scope of protection ought to be clear from the wording of the claim alone. This is nothing new under the sun and necessitates that the description does not allow another interpretation.
What decisions applying G 1/24 have made clear, is that the applicant/proprietor cannot vary in the interpretation of its claim as he thinks fit.
The only point of contact with Art 69 and the protocol at the EPO is the application of Art 123(3) in opposition. This means that the EPO cannot say for instance that the claim encompasses equivalents, if those equivalents have not been originally disclosed. This is indeed the domain of post grant courts.
However, if the prior art discloses equivalents, the corresponding claim might lack inventive step.
DXT, this thread is getting a bit disorientated. Not T422/91. Instead T422/19. Why we can't find it in the EPO Decisions database I do not know. so, I asked MS Co-Pilot. I copy below what the AI reported:
Delete"Die Entscheidung T 0422/19 lässt sich am sinnvollsten im Lichte der jüngsten Grundsatzentscheidungen der Großen Beschwerdekammer – insbesondere G 1/24 (Anspruchsauslegung) und G 1/25 (Beschreibung/Art. 84) – einordnen. Die Verbindung ist enger, als es auf den ersten Blick wirkt, weil T 0422/19 genau an der Schnittstelle von Anspruchsauslegung, impliziten Merkmalen und Art. 123(2)/(3)‑Einschränkungen operiert."
Max Drei,
DeleteI cannot be held responsible if “this thread is getting a bit disorientated”. I used the case numbers I found in “this thread”. What else could/should I have done?
It is a mystery why T 422/19 cannot be found in the board’s decisions data base. There are also no cases citing T 0422/19.
I hope that T 422/19 decision will soon be available in full.
The question is whether T 0422/19 has been issued before or after G 1/24. If it has been issued before G 1/24, I am inclined to consider that it has been superseded by G 1/24, as G 1/24 has made clear that Art 69 and the protocol cannot be used for interpreting claims when assessing patentability under Art 52-57.
What is said in what is apparently known from this decision is plainly obvious. The scope of protection only plays a role when it comes to applying Art 123(3) in opposition.
I can fully agree that, for instance, the role of the EPO is not to determine whether there are equivalents, but when assessing patentability under Art 52-57, it has to see whether the description allows a different interpretation of the claim compared to that derivable from the plain wording of the claim. G 1/24 has been issued for this reason.
In my reply of Monday, 2 March 2026 at 18:03:00 GMT, Part 2, I have shown that decisions applying G 1/24 can as well relate to added matter or clarity. Those decisions exist.
Those decisions can be summarised as follows: if any interpretation found in the description renders a claim unclear, or if one of the technically meaningful interpretations infringes Art 123(2), the applicant/proprietor cannot argue that it is the definition which does not lead to an infringement of Art 123(2) which has to be chosen. This would render any objection under Art 123(2) moot.
Since the appeal in T 0422/19 was withdrawn, there is no decision.
DeleteSee EP2871088
Daniel, I do not hold you responsible. I asked Co-Pilot for the A or B publication corresponding to T0422/19. Its reply: T0422/19 is not indexed in anything available on the net. I'm beginning to think that T0422/19 is an AI hallucination.
DeleteMeanwhile, harking back to my "Ratchet" versus "ratcheting means" hypothetical, in the context of pre-grant amendment, post-grant opposition, the Art 123(2)/(3) inescapable trap and construing the claim under Art 69, EPC, I am ever more curious what G1/25 will have to say about how strictly the A publication has to be conformed before the B publication is allowed to issue.
Unusually, I find myself in (partial) agreement with DXThomas - only insofar as I cannot find T0422/19 anywhere either and cannot find a reference to it in the comments on the prior blog post.
DeletePerhaps somebody could straighten this out to avoid further confusion here?
For what it's worth, going to the EPO Register's "Advanced Search" page and typing "T0422 19" into the Appeal Case Number box (the final one on the page) yields a file entry for EP2871088. However, the file wrapper for that application indicates that the appeal bearing that number was withdrawn without a substantive decision being taken.
It would therefore be incredibly helpful if someone could help straighten this out by clarifying (a) which decision they intended to refer to, (b) what that decision actually says and (c) where a reference may be found to it in the prior discussion. Otherwise I fear that this new discussion is going to get derailed before it even begins properly.
As to T 422/19, its first appearance was in a reply to my post of 24 Feb 2026 at 10:08. I did not note which pseudonym posted the reply but I remember that it began with words like "Max, in answer to your question you might find T 422/19, Reasons 3 helpful"
DeleteBut today, when I track back to my post, I see no such answer. Was it posted to the thread and then deleted by its author, I wonder. Who could that author be? Somebody inside the EPO, perhaps.
Leute, leute, have a look at T 442/91, Reasons 3. Could it be that the first poster of the reference to the Decision got the text right but the case reference wrong?
Delete@Max Drei
DeleteYou are correct. There was a typo in my previous message. The correct case is T442/91.
@DXThomas
You appear to be incapable of envisaging how claims can be interpreted without recourse to an EPO-style primacy of the claims principle. However, downstream courts have been doing precisely that for decades without perceiving any fundamental problems. It might therefore help to advance the debate here if you could learn to grasp (or demonstrate that you already grasp) the completely different philosophy behind the "court-like" approach to claim interpretation.
Whilst it is true that the scope of protection ideally ought to be clear from the wording of the claim alone, that does not mean that the claims must be interpreted as if they represent a stand-alone document, completely divorced from the description and drawings. I refer you to the principles from UK case law that are set out in my comment from 23 Feb 2026 on another thread. Principles (ii) and (vi) to (viii) should make it clear how it is possible to use the context of the description to interpret the claims whilst still anchoring the extent of protection to the wording of the claims.
As there is no decision available under T 0422/19, any considerations based on this non-existing decision are moot.
DeleteThank to Max Drei for finding the right decision T 0422/91. As this decision is rather old, it could be considered superseded by more recent decisions, like G 1/24.
However, G 1/24 has made clear that Art 69 and the Protocol are not be used for assessing patentability under Art 52-57.
For the surplus, what is said in this decision is plainly correct. It is not the role of the EPO to decide which equivalents could fall under the scope of protection of a granted claim, but if the prior art discloses equivalent features of those claimed, the claim might well lack inventive step.
,
Delete'What the EPC requires, cf. Art 84, is that the scope of protection ought to be clear from the wording of the claim alone. This is nothing new under the sun and necessitates that the description does not allow another interpretation.'
I am not clear where that comes from. Art 84 makes no mention of scope of protection. It says that the claims must define the matter for which protection is sought, which simply indicates the function of the claims. It says further that the claims should be clear and concise, which to me is no more than a structural/linguistic clarity requirement (ironically sometimes lost when the EPO requires a claim is rewritten in the two-part form).
If Art 84 was intended to mean that the scope of protection is clear from the claims alone, it has to be wondered why the legislators felt the need to have Art 69 state that the scope of protection is determined by the claims as interpreted by the description and drawings.
To my mind, Art 84 has been long used as a crutch by the EPO to make examination of claims easier for itself. A further indicator of the fallacy of the argument that the scope of protection should be clear from the claims alone is that so many claims simply cannot be understood until the description and drawings have been consulted.
Good comment. Let us all try to respect the difference between the subject matter for which protection is sought and the scope of protection under Art 69 EPC. Different words, to convey different concepts, I say
Delete@Anonymous
ReplyDeleteThis decision is a welcome illustration of the risk of revocation under Art 123(2)/123(3) which results from a deletion in the description.
Practitioners including myself in my G 1/25 amicus brief and user associations have been vocal about this risk. An article by Anja Schäfer and Andrej Majdic (epi information 4-2025) is entirely dedicated to this issue in the post-grant contexts.
Mr Hagel,
DeleteYou have never disguised your opposition to any amendment to the description, but in the present case there was no need to delete anything, but the to bring in the claim the broad definition found in the description.
I can thus not follow your general statement that the present decision is “a welcome illustration of the risk of revocation under Art 123(2)/123(3) which results from a deletion in the description”.
Nothing has to be deleted. It has just to be put in the right perspective, so that the applican/proprietor cannot change its interpretation as he sees fit.
When the broadening or limiting definition from the description is brought in the claim, claim and description will be on a par, without infringing Art 123(2)/Art 123(3).
For the surplus, it has just to be made clear that some original embodiments do not any longer fall under the claim wording as granted or maintained after opposition. There as well, no risk of any infringement of Art 123(2)/Art 123(3).
If done properly, there is no risk at all in spite of yourself and user associations having been vocal about a risk which does not exist.
The adaptation of the description is a necessity for the legal certainty of third parties. It is to be hoped that the EBA will not be impressed by the vocal statements of users or users associations.
It should never be forgotten, that in one situation a party is applicant/proprietor and in another situation the same party can be opponent/potential infringer. In the second situation the party can be glad to know where it stands. Why then refuse to clarify matters in the first situation? I fail to understand the logic behind this discrepancy.
"If done properly, there is no risk at all..."
DeletePlease, Mr DX Thomas, walk across this tightrope over the swamp filled with crocodiles. There is no risk at all if you put one foot in front of the other without error.
I note the Board's comment in point 3.4 of the decision:
ReplyDelete"This appears to be why the Enlarged Board saw the answer to question 3 ("May a definition or similar information on a term used in the claims which is explicitly given in the description be disregarded when interpreting the claims to assess patentability and, if so, under what conditions?") to be encompassed by the answer given to question 2 which is in essence: "The description and drawings shall always be consulted to interpret the claims when assessing the patentability of an invention under Articles 52 to 57 EPC" (see the order of G 1/24 and Reasons 1 and 21)."
This seems to be in line with what I and others have been arguing, most notably Thomas Körner in epi Information 3|25 (https://information.patentepi.org/issue-3-2025/g-1-24-clarity-of-claims.html), which I quote here:
"By considering Question 3 to be encompassed within Question 2, the EBoA seems to imply that a definition of or similar information on a term used in the claims which is explicitly given in the description shall always be considered. No indication is given of any conditions and/or circumstances under which such a definition or similar information may be disregarded."
However, the cases applying G 1/24 so far seem to have developed a condition for disregarding a definition - if the definition is narrowing, it can/should be disregarded.
A reminder of Q3:
"3. May a definition or similar information on a term used in the claims which is explicitly given in the description be disregarded when interpreting the claims to assess patentability and, if so, under what conditions?"
And G 1/24's comments on Q3:
"1. Question 3 is considered by the Enlarged Board to be encompassed within Question 2. As the Enlarged Board is going to give an answer to Question 2, the Enlarged Board considers that Question 3 is inadmissible as an answer to it is not necessary for the Referring Board to reach a decision in the case before it."
If G 1/24 intended the (or a) condition for disregarding a definition to be "is the definition narrowing?", then why did they not answer Q3 by setting out this condition?
The answer is because the boards are paying lip service to G1/24. G1/24 was explicit and said that deliberate divergence from downstream courts was to be avoided, but the boards have disingenuously reinterpreted the decision in an attempt to cling onto their old line of case law. In so doing, they seek divergence from the downstream courts which apply a more common sense approach to reading documents.
DeleteInterpreting documents is a standard procedure in law. It should not have taken 50 years to settle the debate. Consider the document as a whole and in context to figure out what it means, then deduce the consequences for the parties only after you have done that. It really is that simple. As soon as you try to factor in the consequences of the interpretation during the process of performing the interpretation, the result is an illogical, circular mess. You cause issues because words in claims mean different things depending on whether they are a patent application being applied for vs. when they are in a prior art document. Another consequence from rules that focus on primacy of the claims is that words take on new meanings when they are extracted from the description and moved into the claims. The claims are words that must be interpreted in context just like any other word in any other document.
The terms of a patent mean what they say regardless of whether the proprietor benefits or loses. If you invent rules such as “broadening definitions are taken into account when they hurt the proprietor on validity but narrowing definitions are ignored when they help the proprietor on validity”, you are not performing a true act of interpretation; you are conflating interpretation and adjudication. The EPO and its boards of appeal are wilfully blind and legally illiterate on this issue.
Could not agree more. The EPO appears to have persuaded itself that simply because claims should ideally convey a clear meaning on their own, they should be interpreted on their own (ie without any need to refer to the description) in all but the most exceptional circumstances. For this reason, concepts such as traditional rules / canons of document interpretation have been abandoned in favour of EPO-style primacy of the claims.
DeleteOf course, a side effect of the EPO's approach to claim interpretation is that, strictly speaking, examination of an application can be completed without any examiner ever needing to bother with the time-consuming task of reading the description. A cynic might view this as a strong incentive for the EPO to maintain their current (and, as you point out, legally illiterate) approach to claim interpretation. Such a cynic might also go on to point out that description adaptation is the EPO's attempt to limit the damage caused by their absurd approach to claim interpretation, with the consequence that the EPO also has a strong incentive to defend what is clearly an indefensible interpretation of Article 84 EPC.
Of course, I am nowhere near so cynical and so would not stoop as low as to suggest that all of the absurd interpretations of the EPC that have been (and are being) advanced by the EPO could be explained purely by financial motivations (ie minimising examiner time per case and thereby maximising net financial gain per case). This is because there must of course be a perfectly reasonable explanation for all of this ... I just cannot yet figure out what that explanation might be!
@ Anonymous-02.03.2026-09.58.00 GMT+ Anonymous-02.03.2026-13.03.00 GMT
DeleteReply part 1
When looking at board’s decisions applying G 1/24, next to the present one, it is difficult to say that the boards disregard what can be found in the description.
It is not under dispute that the description can contain both a broadening or a limiting interpretation of the plain wording of the claim.
The board’s simply say that if an applicant/proprietor wants the broadening or limiting interpretation to be taken into account, the claim has to be amended accordingly. To come to this conclusion, the boards have necessarily consulted the description.
The boards are neither paying lip service to G 1/24, nor have the boards disingenuously reinterpreted G 1/24. All board’s decisions applying G 1/24, even the present one, have made clear that if there is the description allows a different interpretation from that resulting from the plain wording of the claim, the applicant/proprietor cannot be left to decide as he thinks fit what the interpretation of the claims wording should be.
The legislator and G 1/24 have never said that what matters is what is in the description, be it broadening or limiting, and allows to supersede what is derivable from the plain wording of the claim. Most of the decisions applying G 1/24, deal with the situation where the description allows a broadening or a limiting interpretation of a claimed feature, but not only.
When following your reasoning, then claims are superfluous. According to the EPC, and independently of Art 65, and the London Agreement, only claims have to be translated into the other two official EPO languages in order to have a patent granted or maintained in amended form.
@ Anonymous-02.03.2026-09.58.00 GMT+ Anonymous-02.03.2026-13.03.00 GMT
DeleteReply part 2
In the following, readers will find decisions which show that the boards are anything but wilfully blind and legally illiterate on the issue of claim interpretation. Rather G 1/24, Reasons 10, confirms most of the established case law, including the principle of the primacy of the claims, cf. G 1/24, order, first sentence.
When a claim is unclear and the description does not allow to shed light into the matter, a board is not obliged to interpret an unclear claim by making assumptions and an examination under Art 56 cannot be carried out on that basis, see T 389/24. A similar result is found in T 981/23.
In T 981/23, the board even considered that G 1/24 has not only to be applied to assess the “patentability of an invention under Art 52 to 57″ but also for the issue of Art 123(2). The board referred to T 873/23, and T 2048/22.
In T 873/23, the board noted that, according to T 367/20, Reasons 1.3.2, a claim feature must not be interpreted in isolation but in the context of the whole document it forms part of. This is done, however, by taking due account of the primacy of the claims. This is also in line with G 1/24 and UPC_CoA_335/2023, second headnote, according to which the claims are the “decisive basis“. The board concluded that a new feature in the claim constituted an intermediate generalisation.
In T 1164/23, the board noted that the description and claims as filed express “the will” of the applicant for the very reason that they were drafted by the applicant himself, and thus, there would not be any need for further speculation on what else might be the subject-matter to be protected. Assuming arguendo that “the will” really was to be considered for examining the allowability of an amendment – which the board strongly rebuts – this concept would identify multiple different wills in the present case. In view of this and in application of G 1/24, claim 1 of AR1 was considered infringing Art 123(2).
In T 1588/23, the board held that unclear features cannot be ignored under G 1/24, and concluded that claim 1 was novel and inventive.
In T 405/24, the board noted first that, even if the Order of the 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).
T 669/23 shows that G 1/24 does not only apply to the patent as such, but can also be used when interpreting prior art. If the interpretation of features in the prior art and in the patent coincide, then the features are identical.
We are thus indeed far from the boards being disingenuous in their application of G 1/24 or being illiterate.
Jesus Christ... What an unholy mess the EPO is making of this.
ReplyDeleteI think that the EPO needs to swiftly disavow any approach that seeks a set of hard rules for using the description to interpret the claims. They should move towards a much fuzzier but more realistic approach: will all departments please just interpret each claim in light of the common general knowledge of the skilled person to arrive at an appropriate, technically sensible meaning of the claim.
If the claim says "The sky is blue" and the description says "blue includes green", the relevant question is not 'what effect does the description have on the claim?', but 'what does the skilled person understand by "The sky is blue?" and would the skilled person understand it to include green' Maybe it does include green, but maybe the skilled person would dismiss that out of hand as being nonsense. One should decide on a case-by-case basis.
For me, the hottest open issue in the EPO world of today is the degree to which the description must be conformed, by the ED, in the EPO, prior to grant. For every 100 cases suffering pre-grant narrowing amendment, only a very few percent are subject to post-grant enquiry.
DeleteSo, will the EBA dare to touch this very hot potato, which sets a huge challenge for the one who will draft the EBA's decision, or will it plead judicial economy and duck out of answering it despite the compelling and very long-felt-want of a solid answer?
One hopes that the EPO has chosen a suitable case, fact matrix, vehicle, on which to base a pivotal decision that is blessed with compelling logic. Will it be David Rogers who will write it? Is he by now near retirement? Might it be his landmark legacy decision?
Anonymous, I thoroughly agree with your assessment of the EPO's unholy mess exemplified by the blue sky claim! Let's just do away with all this amendment of the description nonsense, which is unique to the EPO.
Delete2nd attempt to post the following comment (which is not yet visible, so apologies if I end up repeating myself):
ReplyDeleteRose, I am a little taken aback by your comment that "Such an interpretation of G 1/24 has radical and unwelcome consequences for the potential impact of description amendments on claim interpretation post-grant".
It is only necessary to consider the implications of following the EBA's suggestion in G 1/24 (to harmonise with the claim interpretation practice of the downstream courts) to reach the conclusion that amendments to the description will of course have the potential of adding subject matter and/or extending the protection conferred by the claims. It would therefore be a stretch to describe that outcome as stemming from a "radical" interpretation of G 1/24.
Indeed, as Francis Hagel has pointed out, various professionals (including myself) have for a long time believed that EPO-style adaptations of the description are highly undesirable precisely because they give rise to grave (and potentially fatal) risks under Article 123(2) / 123(3) EPC. Thus, whilst such risks may come as unwelcome news to some, they are risks that have been hiding in plain sight for a long time.
It remains to be seen whether the EBA agree with the reasoning of the Board in T 439/22. However, it is hard to find any logical flaws in the Board's interpretation of the ruling in G 1/24. Other Boards have exploited arguable ambiguity in the meaning of "consult" to essentially carry on interpreting claims as they would have done prior to the EBA's ruling in G 1/24. However, it was the Board in T 439/22 which first used the word "consult" (in Q2 of the referral to the EBA). Thus, presumably, they would be best placed to determine what they meant by that word. Also, whilst it is perhaps possible that the EBA may have taken that word to mean something different, it seems difficult to fault the Board's conclusion that "consult" is essentially synonymous with "use(d)" in Article 69(1) EPC.
@ Proof of the pudding-02.03.2026-17.06.00 GMT
DeleteAs I explained in a previous comment, if done properly, there is no risk of infringement of Art 123(2)/123(3) when adapting the description to the claims.
The role of the EBA is by no way to agree or not “with the reasoning of the Board in T 439/22”. The EBA has no competence on its own to decide whether a board has correctly interpreted its decision or not.
This would only be possible, if there would be a new referral to the EBA, should it appear that boards give contradictory interpretations of G 1/24. We are far from this.
As in G 1/24 the EBA has made abundantly clear that neither Art 69 and the Protocol or Art 84 can be used as a basis for assessing patentability under Art 52-57, it is a daring conclusion to allege that “"consult" is essentially synonymous with "use(d)" in Article 69(1) EPC”. On the one hand you refuse that the EPO infers in the competence of post-grant courts, but you want to see Art 69 and its Protocol to be used by the EPO. I see here a fundamental contradiction.
It is clear that every board is free in its interpretation of a decision of the EBA. I am however not sure that the other boards are necessary pleased with T 439/22. All the previous decisions applying G 1/24 have been ignored by the board in T 439/22. Those decisions do not only deal with assessment of the patentability under Art 52-57, but also with clarity and added matter. The decisions bringing G 1/24 in relation with clarity should not be ignored.
What is to be avoided is that the applicant is allowed to chose the interpretation allowing him to escape an objection of lack of clarity and/or added matter. In T 1819/23, the board added that, in accordance with G 1/24, interpreting a claim means establishing its proper technical meaning. If one of the possible technical meanings is unclear or implies added matter, an objection under Art 84 or 123(2) is fully justified, even if the claim as such might at a glance appears clear or not adding matter.
You may see a fundamental contradiction. However, that is only because you are missing a crucial point.
DeleteYou are correct that neither Art 69 nor Art 84 EPC provides the rules for claim interpretation at the EPO. However, the principles underpinning claim interpretations according to G 1/24 are the same as those which underpin Art 69 EPC and its Protocol. If it were otherwise, it would have made no sense for the EBA to have advocated for alignment with claim interpretation practice of the downstream courts. Those courts rely solely upon Art 69 EPC and its Protocol for claim interpretations – so how could the EPO align with the downstream courts if they do not apply the same principles (if not, technically speaking, the same law)?
@Proof of the pudding
ReplyDeleteThe decision is not strictly speaking relevant to G 1/25, since the request for deleting paragraph [0035] was submitted by the patentee of their own motion. However, it will provide applicants a useful argument to resist a Rule 71(3) amendment by the ED or an ED’s request to amend, by illustrating the risk of Art 123(2)/ Art 123(2) issues arising from the amendment. Importantly, the decision explains that the deletion would alter the interpretation of the claims.
The argument based on this risk is in tune with the principle rooted in Art 113(2) that the applicant/patentee is solely responsible fo the patent as granted. The EPO incurs no liability even in the case of iatrogenic revocations, i.e. if the patent is held invalid as a result of an amendment entered by the ED or required by a communication.