On 28 December 2020, the Barcelona Court of Appeal (Section 15), one of Spain’s most experienced courts on patent matters, handed down a judgment which is interesting from a wide array of perspectives (how to assess inventive step, novelty, infringement, etc.). This blog will focus on the first aspect (i.e. inventive step) and, more specifically, on the dangers of combating inventive step applying the “problem-solution” approach if the court ultimately finds that the first step of that method has not been correctly addressed. The facts of the case can be summarized as follows:

The claimants, a French manufacturer of automotive parts and its Spanish subsidiary, were the holder and licensee of the Spanish validations of European patents EP 764.811 (“EP ‘811”) and EP 933.254 (“EP ‘254”), which protect vehicle lighting or signaling devices having specific features. They filed a patent infringement action against the defendant, a Spanish supplier of spare auto parts, which had sold spare headlamps and signal lights for cars that allegedly infringed the claimants’ patents. By way of counterclaim, the defendant filed a revocation action against various claims of the patents in suit, namely Claims 1–6 of EP ‘811 and Claims 1 and 18 and their respective dependent claims of EP ‘254. In a judgment dated 6 March 2019, Barcelona Commercial Court No. 1 dismissed the infringement action and partially upheld the revocation counterclaim, revoking certain claims of EP ‘254 on grounds of lack of novelty. However, the Court confirmed the validity (in particular, inventive step) of patent EP ‘811. The claimants lodged an appeal before the Barcelona Court of Appeal. They argued that the defendant’s spare parts did indeed infringe Claim 1 of EP ‘811 and disagreed with the first instance decision’s finding on the novelty requirement. The defendant, in turn, lodged a cross-appeal, disagreeing with the first instance findings on the validity (in particular, inventive step, of EP ‘811).

With regard to the topic of this blog (inventive step), in the judgment of 28 December 2020, the Barcelona Court of Appeal (Section 15) noted that Claim 1 of patent EP ‘811 could be defined by the following technical elements: C1.- Apparatus for lighting. indicating or signalling purposes in a motor vehicle, C2. – Including a ventilating device which consists of first ventilating means carried by a housing (100) of the unit, in combination with second ventilating means carried by a cover member, (200) mounted on the housing of the unit, C3.- So as to define together a sinuous path to the interior of the unit comprising a double air inlet in a lower part of the ventilating device, C4.- The sinuous ventilation path is defined partly by the housing or body of the apparatus and partly by a cover member which is carried on this housing or body, C5.- The double air inlet in a lower part of the ventilating device comprising two inlet apertures facing each other (216a, 216b), C6.- Together with an inlet passage (T1) which extends substantially transversely to the general direction defined from one said air inlet aperture to the other. C7.- The sinuous ventilation path constitutes a labyrinth comprising two chicanes giving two changes of direction (T1, T2, T3). C8.- The two changes of direction extend substantially upwards from the double air inlet, C9.- A passage of communication (T3, 1201) between the labyrinth and the interior of the unit C10. The passage being defined at least partially between lugs (210a, 210b) by which the cover member is fitted resiliently on the housing.

According to the defendant, the only difference between Claim 1 and the document asserted as the “closest prior art” (patent application WO 95/02783, “WO ‘783”)) was C10. Therefore, the defendant defined the “objective technical problem” taking into account this difference only. In contrast, the patent owner alleged that, in addition to C10, WO ‘783 did not disclose characteristics C3 or C5-C9 either. This opinion was shared by the Court of Appeal, which highlighted that one could only read characteristics C3 and C5-C9 assessing WO ‘783 with hindsight. In this regard, in paragraph 19 of the section of the judgment dealing with inventive step, the Court noted that:

19. In order for a document on the state of the art to disclose these characteristics, it must describe them, directly and unambiguously, so that a person skilled in the art who reads such documents on the application (or priority) date, with the necessary common general knowledge but without knowing the patented technical solution, could recognize those technical teachings. The skilled person should not seek in the state of the art the solution claimed, but seek there the solution that was offered at that time for the technical problem that the invention solves, so as to compare it to the one claimed. The skilled person, who the courts build, necessarily knows the claimed technical solution; but in analyzing the prior art and comparing it with the claim at issue, he must be able to justify the identification of the elements of the claim challenged in the teachings of the document itself, not in the teachings of the patent at issue. In other words, the skilled person must justify his arguments on prior art; otherwise, he would be making an ex post facto judgment and not analyzing what was known before the application (or priority) date.”

Building upon this premise, the Court arrived at the conclusion that, having failed to correctly establish the differences between claim 1 and the document alleged as the “closest prior art” by the party seeking the revocation of the patent, caused the inventive step attack to fail, making it unnecessary to consider the subsequent steps of the “problem-solution” approach. In particular, in paragraph 32 of the section addressing inventive step, the Court reached the following conclusion:

32. Consequently, the differences between D1 and the claim at issue are not limited to C10, not even to C5, but comprise technical characteristics C3 and C5 to C10. Therefore, we can no longer share the first step of the method followed by the defendant to assess the lack of inventive step. If we cannot start from that first stage in the application of the method followed to analyze inventiveness, we will also not be able to share its conclusions. Which leads us to the rejection of that claim of invalidity, without having to look at the following reasoning.”

All in all, this very interesting judgment contains a clear word of warning for patent revocation seafarers: if you do not set the first stone of the “problem-solution” approach in a way that the Court finds to be correct, you may sink.


To make sure you do not miss out on regular updates from the Kluwer Patent Blog, please subscribe here.

Kluwer IP Law

The 2022 Future Ready Lawyer survey showed that 79% of lawyers think that the importance of legal technology will increase for next year. With Kluwer IP Law you can navigate the increasingly global practice of IP law with specialized, local and cross-border information and tools from every preferred location. Are you, as an IP professional, ready for the future?

Learn how Kluwer IP Law can support you.

Kluwer IP Law
This page as PDF


  1. The novelty test “in order for a document on the state of the art to disclose these characteristics, it must describe them, directly and unambiguously, so that a person skilled in the art who reads such documents on the application (or priority) date, with the necessary common general knowledge but without knowing the patented technical solution, could recognize those technical teachings.” seems to be a bit at odds with the EPC as interpreted by the EPO, as this sounds more like the EPO test for original disclosure than the EPO test for novelty, which is wider in that it also takes into account implicit disclosure and independent from any hindsight considerations.

  2. That’s indeed the problem, with the Problem-Solution Approach (EPO-PSA) , isn’t it? The contrast between the use of EPO-PSA at the EPO and its use in the patents courts around Europe.

    At the EPO, PSA is set in stone. Nobody passes the written exam to qualify as a European Patent Attorney without satisfying the Examiners that one has a sure understanding and grasp of the Approach. No representative argues obviousness with anything other than EPO-PSA. No EPO Decision reasons obviousness with anything else but PSA. By now, nothing else is even imaginable.

    Which means that every patents court in Europe feels an urge to defer to it, even apply it themselves.

    Which would be very unfortunate, if it to be so, that very few litigators, and only a few judges, have any grasp or own experience of it at all.

    Adding to the problem is that EPO-PSA is so seductive. Its like one of those computer user interfaces that is so amazingly intuitive that every lawyer thinks they’ve grasped it, even when they haven’t any experience how to use it in a real case. Given these intuitive feelings, they see no need to learn, rehearse, practise how to operate under EPO-PSA. Their intuition will suffice, they erroneously suppose.

    Or do readers disagree?

  3. Just for comparative purpose:
    The BGH (and the BPatG) found the German part of EP ‘811 to be valid (BGH, judgement of 27.11.2018 – X ZR 16/17 – Scheinwerferbelüftungssystem
    https://juris.bundesgerichtshof.de/cgi-bin/rechtsprechung/document.py?Gericht=bgh&Art=en&sid=b2a2dfef0057a96644ed07dd34c84061&nr=93621&pos=0&anz=1 ).
    At a quick glance, the judgements seem to follow similar lines of thought (Rough translation of the catch-phrase of the BGH: During claim construction, the characterising features should usually be construed in such a way, that the patent proprietor wants to distinguish the claim from the prior art).
    NB, the German decisions found the claims valid compared with further prior art.

  4. @ Peter Parker

    In my humble opinion the “EPO test for original disclosure” is identical to the “EPO test for novelty”. What is valid for the application/patent is also valid for the prior art.

    I would just add that in the presence of a different terminology between the claim and the prior art demonstrating identity of features requires more than indicating bibliographic data. It requires an explanation as to why the features are identical in the limit of what the person skilled in the art derives directly and unambiguously from the prior art including his common general knowledge.

    @ The author:

    As far as the choice of the closest prior art is concerned, I would like to draw your attention to a few decisions:

    T1076/16. It is not because a document having a different aim than in the application/patent that it cannot be used as CPA, cf. Point 11 of the reasons.

    T 405/14: A document which comprises less features in common with the invention and not having the same aim can nevertheless be a correct starting point for dealing with IS, cf. Point 17-19 of the reasons.

    T 2057/12: The selection of closest prior art has consequences on the definition of the skilled person, cf. Point 3.1.4 cf. the reasons. The BA held that, it “can indeed not be excluded, beforehand, that real-world circumstances would have led a skilled person in a certain technical field to have given attention or even used an item of prior art from a completely different technical field”.

    T 1742/12: for the BA, it has been acknowledged that the choice of the closest prior art may not always be unambiguous and that, in such a case, the problem-solution approach may have to be repeated starting from other pieces of prior art, cf. Point 6.5-6.6 of the reasons.

    Whilst I would agree with the author that “if you do not set the first stone of the “problem-solution” approach in a way that the Court finds to be correct, you may sink”. I would however allow myself to add that there can be more than a single stone to start with when dealing with IS.

    It remains that once a document is considered as CPA, the differences have to be correctly assessed. If a further document can as well be considered as CPA, the differences are necessarily different (sorry for the pun) and the argumentation of lack of IS is necessarily different. Just have a look at T 591/04.

  5. As far as the German Federal Court is concerned (BGH) it is manifest that it is pro-patentee!
    Anything possible to find the patent valid will be used, even if it is in blatant contradiction with the case law of the BA, and often it will claim that it follows it very carefully!

    1. Dear Attentive,

      maybe you could explain where exactly the BGH is so pro-patentee.

      To me (examiner at the DPMA) they seem more strict than the BA (always considering that most of the cases before the BGH are concerning validity, where there is a different burden of proof compared to examination and opposition).

      Concerning this case of the car lights, I don’t like the wording of the “Leitsatz” – the reasons given for upholding validity are OK though, IMHO.

  6. @Attentive Observer: at the EPO, the tests for novelty and original disclosure are similar but not quite the same. For a feature to be “originally disclosed”, it needs to be disclosed in the original application documents AND be disclosed as “part of the invention”. Novelty is assesed based on the whole content and the question is whether a feature is directly and unambigiously disclosed, be it explicitly or implicitly. Further, novelty excludes unpublished abstracts whereas original disclosure excludes the abstract, published and unpublished.

  7. @ Peter Parker
    I think you misunderstood my comment and I agree with you..
    It is well established case law that part dealing with the prior art mentioned in the application/patent is not part of the original disclosure and cannot be taken to support amendments. Such amendments would contravene Art 123(2)!
    Very recent decisions: T 2241/15 or T 2359/16.
    Provided the PA is enabling, all parts of a document can be taken into consideration.
    In both cases the criterion is the same: what is directly and unambiguously disclosed taking into account common general knowledge!

Comments are closed.