EP 2 705 313 B1 relates to a household refrigeration appliance with a heat insulation wall.
The patent is in German and the term at stake reads “übereinanderliegend”, i.e. lying on top of each other. In the English version of the claims, the term “superimposed layers” was used.
Brief outline of the procedure
The patent was maintained according a new MR which corresponded to AR2 before the OD.
The opponent appealed.
What is at stake is the feature “superimposed layers” in the first cladding element , on the narrow side. At least locally, the element has a plurality of superimposed layers of the flat material.
The board held that claim 1 of the MR, i.e. as maintained, lacked N over D2=WO 2009/065683.
In its communication under Art 15(1) RPBA, the board had noted that the opponent did not raise any objections against AR1 before the OD. The opponent’s late filed objections against AR1 by the opponent were thus not admitted. They were also not convincing.
The patent was thus maintained according to AR1.
The opponent’s point of view
The opponent argued that the feature “superimposed layers”’ did not require direct contact or mutual contact between the layers, but merely a local arrangement in which the layers are superimposed one on top of the other.
This would be the case, for example, where layers lie parallel to one another and overlap locally, with or without a gap between them.
The proprietor’s point of view
According to common linguistic understanding, the term “superimposed” means that the layers are in direct contact with one another and touch each other. Crucial to this understanding is the meaning of the verb “to lie”, which implies support by or resting upon a base.
Consequently, “layers lying one on top of the other” require not merely a spatial arrangement, but a static stacking. When lying, the support is not provided independently of the spatial arrangement one on top of the other by another component, but precisely by this arrangement, that is, by the layers themselves.
In the patent, too, the term “superimposed” is consistently understood to mean that the layers are at least in contact with one another at certain points and rest on one another.
The apparent absence of contact points visible in some figures can be explained in particular by the fact that the profile in these figures has not yet been foamed, cf. § [0031]. However, due to the pressure during foaming, the layers would also come into contact with one another in these figures.
In Figures 6 and 7 a groove or joint between two of the layers is indeed shown, but in the relevant passages these layers are not described as “superimposed layers” but only as “arranged one above the other” or “bent one above the other”. The aspect of “lying”, which expresses the idea of supporting one another, is therefore absent.
Thus, when taking into account the disclosure of the patent in dispute in its entirety, as required by G 1/24, it follows throughout that “superimposed layers” must be in direct contact with one another and must rest on one another.
The opponent’s differing interpretation of the term, however, is based solely on a reference in claim 7. According to R 43(7), reference signs shall not be construed as limiting the claim.
The three reference signs (40, 41, 49) inserted after the phrase “several superimposed layers” in claim 7 would otherwise limit the subject-matter to three superimposed layers rather than ‘several’ (i.e. at least two).
In the example shown in Figure 6, to which claim 7 relates, there are indeed two “superimposed layers” (40 and 41). However, the layer designated by reference number 49 is, however, spaced apart from the other layers by the inner lining element 24 and is therefore not “superimposed”
The board’s decision
Literal meaning
For the board, the “lying” of the layers must be such that several layers are at least locally superimposed in the vertical direction, but not that they support “one another”.
The wording, and any reference to support, does not preclude the possibility that something else may be situated between the layers. In particular, it is not required that the layers rest ‘on top of’ one another or lie ‘against’ one another.
Functional meaning
The same conclusion follows from consideration of the functional significance of the “superimposed layers” for the stiffening and dimensional stability of the thermally insulating wall, as explained in paragraphs [0010], [0015] and [0017], as well as the reference in paragraph [0010] regarding the formation of thermal bridges by these layers.
If, as argued by the proprietor, it can be inferred from this that the layers must support one another and that there is thermally conductive contact between them, this is also the case when a layer of another thermally conductive and supporting element is inserted between them, as in D2. Through the interlocking complementary hooks 14 and 15 in D2, these layers together form a thermal bridge, support one another and serve to stiffen the wall cladding against deformation during foaming, cf. D2, page 4, line 31 to page 5, line 15.
The disclosure in the patent
The patent does not contain any explanation or definition to the effect that ‘superimposed’ layers must always be in contact with one another. On the contrary, the embodiments shown in Figures 5, 6 and 7 contain indications that this is precisely not to be understood as such.
In relation to R 47(3), the board referred to a decision of the Federal Court of Justice, 15.112005, X ZR 17/02, Coke Oven Door, para. 11 with contains a reference to T 237/84.
For the board, the subject-matter of claim 7 is not limited by the interpretation adopted by the board to either exactly three layers or exactly the superimposed layers shown in Figure 6.
Figure 6 forms part of the description and drawings which must always be taken into account when interpreting the claims in accordance with G 1/24. The reference signs in claim 7 facilitate an understanding of the relationship between the embodiment shown in Figure 6 and claim 7.
The board considered Figure 7 as representative of the invention defined under claim 1 as granted. With reference to Figure 7, the board considered that the cladding element must comprise the multiple superimposed layers ‘on the narrow side’ of the heat-insulating core, and the layer designated by reference number 37 is the only layer that rests directly on the heat-insulating core. Consequently, the “superimposition” in the contested patent does not require direct contact.
Comments
Reference signs shall not be construed as limiting the claim, but they can be used to interpret the claims according to G 1/24. In Figure 7, also representing the invention as filed and as granted, it is manifest that the superimposed layers do not have to lie directly on top of each other, i.e. they are not superimposed layers.
The present decision actually confirms T 237/84. This is a very old decision, I was involved in. It is for the ED to decide whether reference signs are helping the understanding of the claim or not, but reference signs are not limiting.
Why it was then necessary to refer to a decision of the BGH remains a mystery?
On the procedure
D2=WO 2009/065683 bears the classification symbols F25D23/02 and F25D23/06.
EP 2 705 313 B1 bears the classification symbols F25D23/02 and F25D23/06. The ISR established by the EPO was carried out in F25D-
It is thus surprising that D2 has not been found during the search.
Comments
2 replies on “T 0903/24-Reference signs and claim interpretation”
I do not find this a helpful decision. As you say, this issue was settled long ago and the need to discuss the BGH is difficult to follow.
Rule 43 contains an instruction that reference numerals should not limit the claims and I do not see how G1/24 changed that. Furthermore I see problems in the (machine translated) statement ‘this indirectly results, via Figure 6, in an explanatory effect of the reference numerals on the understanding of the terms used in the claim’. A simpler way to address this would have been ‘the wording of the claims appears to relate to what is shown in Figure 6, so the disclosures in Figure 6 can impact the understanding of the terms used in the claim’. Mentioning reference numerals was not necessary in my opinion.
Following this decision, I expect to see parties arguing that their claim interpretation relies on an explanatory effect of reference numerals, which will complicate proceedings. One can imagine cases where an incorrect reference numeral is used to arrive at an interpretation that would not have been reached if the reference numeral had been omitted, which could lead to granted patents being invalidated due to their reference numerals. I would liken this decision to opening Pandora’s box.
@ Anonymous
I will not enter a discussion as to whether the decision is helpful or not.
I am of the opinion that G 1/24 has not changed the way reference signs are to be looked at. Reference signs do not limit the claim, but help to increase the intelligibility of it. I cannot see a problem when a reference sign is helping to understand the meaning of a claimed feature when applying G 1/24.
In the present case, the board merely used the reference sign to explain why the superimposed layers do not have to lie directly on top of each other. If, in line with other decisions applying G 1/24 (but T 439/22), the proprietor wanted to have this limitation to be taken into account, it should have amended the claim accordingly, provided this explanation was present in the originally filed disclosure. In the present case it would not have been possible, as it was manifestly not originally disclosed.
I did not mention it in the blog to shorten it, but in Reasons 2.8.2, the board referred to a decision of the Court of Appeal of England and Wales, 14.01.2022, [2022] EWCA Civ. 20, para. 39. In this paragraph the Court of Appeal of England and Wales noted: “there is nothing wrong with using the reference numerals to describe how the claim works by reference to the figure”. I can fully subscribe to this stance. This is exactly what the board did in the present situation.
If parties wish to argue that their claim interpretation relies on an explanatory effect of reference numerals, so let them do so. I doubt they will go far with such an argumentation.
I do fail to see how this should complicate matters. I have not yet seen a case in which a granted patent has been invalidated due to the reference numerals in the claim. It is clear that it is not for the applicant/proprietor to decide whether reference signs are increasing the intelligibility of the claim or not. It is for the ED or the OD to have the last say in the matter.
In T 237/84, the applicant accepted to add reference signs during prosecution, but wanted them deleted at grant stage. T 237/84 put an end to this practice.
If reference signs are not limiting but increasing the intelligibility of the claim, it is legitimate to take them into account when interpreting the claim. A claim has to be clear as such, irrespective of whether it comprises reference signs or not. Therefore, I do not think whatsoever that the present decision has opened the Pandora’s box.