Setting the record straight concerning the right to dependent claims under the EPC
by Thomas Wyder and Stephan Steinmüller (Hoffmann Eitle)
EPO Examiners sometimes urge Applicants to delete dependent claims considered to relate to “unsearched subject-matter” after having raised a unity a posteriori objection. We are investigating in how far the EPC justifies forcing an Applicant to delete dependent claims as well as possible counter-arguments for defending them.
Legal basis
Rule 164(2)(c) EPC authorizes the Examiner to force an Applicant to excise subject-matter and to focus on one invention for Euro-PCT applications. If an Applicant was invited to but did not pay further search fees, the searched invention is considered elected. An analogous provision does not exist for “direct” European patent applications. However, according to G 2/92, if no additional search fees are paid for an EP application, then the searched invention is considered elected, and the Applicant must excise the remaining inventions. One might thus state that Rule 164(2)(c) EPC may be applied analogously to direct European applications. Rule 63(3) EPC sees to a case in which an Applicant is forced to excise subject-matter after receiving an incomplete (a partial) search report because a meaningful search cannot be carried out. Thus, one might argue that G 2/92 implies that Rule 63(3) EPC is to be applied analogously to direct applications for which a unity objection a posteriori was raised when no additional search fees were paid.
Coming from a different angle, Rule 137(5) EPC stipulates:
“Amended claims may not relate to unsearched subject-matter which does not combine with the originally claimed invention or group of inventions to form a single general inventive concept. Nor may they relate to subject-matter not searched in accordance with Rule 62a or Rule 63”.
It is to be emphasized that this provision requires two criteria to be fulfilled if subject-matter is to be forbidden in an amended claim. The subject-matter must be unsearched as well as not forming a single inventive concept with the originally claimed invention. Stating that an amended claim contains unsearched subject-matter is not sufficient for an Examiner to justify insisting on a deletion of the claim based on Rule 137(5) EPC.
“Searched” and “unsearched” dependent claims
Diversities in dependent claims are unproblematic with regard to lack of unity a priori. The Guidelines for Examination, Part F, V, 9, state that “no objection on account of lack of unity a priori is justified in respect of a dependent claim and the claim on which it depends, on the ground that the general concept they have in common is the subject-matter of the independent claim, which is also contained in the dependent claim”. Claims depending on a novel and inventive claim are linked by a single inventive concept. This may serve as a first guiding principle when dealing with a lack of unity a posteriori objection.
When an independent claim is considered novel and inventive, there is no need to direct a (further) search at the dependent claims (Guidelines for Examination, Part B, III, 3.7). In other words, although no specific actual search has been carried out for the dependent claims, they are usually deemed to have been searched, i.e., “searched by fiction”. This may serve as a second guiding principle for a scenario involving a lack of unity a posteriori objection.
It follows that the issue of whether a particular dependent claim is considered searched or not does not necessarily depend on whether an actual search has been performed for the subject-matter of the claim. Whether a claim is searched is primarily a legal and not a factual question.
When a lack of unity a posteriori objection is raised and an Applicant does not pay all additional search fees, the Examiner will often request the Applicant to fully delete those dependent claims that he/she thinks have not been searched. However, by simply following this request, the Applicant may unnecessarily surrender subject-matter, as will be explained by means of the following example.
An application as filed contains independent claim 1 and dependent claims 2 and 3, the latter depending on claims 1 and 2. A lack of unity a posteriori objection is raised since claim 1 is not considered patentable and dependent claims 2 and 3 are deemed not to relate to the same inventive concept. No further search fee is paid by the Applicant, and the search is restricted to claims 1 and 2. The subject-matter of claim 2 is held by the Examiner to be novel and inventive, while the subject-matter of claim 3 is deemed to constitute unsearched matter. Thus, the Examiner requests the Applicant to excise this latter subject-matter from the application.
If Applicant follows this request, he may later be barred from amending claim 1 by including therein the feature(s) of original claim 3, even if there is a basis for such a combination in the description. Such a feature combination would constitute unsearched subject-matter which does not combine with the originally claimed invention (feature combination of original claims 1 and 2) to form a single general inventive concept, thus violating Rule 137(5) EPC. Hence, the dependency of claim 3 on claim 1 can be coined an “unsearched dependency”. However, the Applicant may amend claim 1 by including therein the feature(s) of original claim 2, thereby arriving at an amended claim 1 whose subject-matter is novel and inventive. In this case, original claim 3 can be maintained in the application as a new dependent claim 2 which is dependent on amended claim 1. Hence, the dependency of claim 3 on claim 2 can be termed a “searched dependency”.
Keeping the “searched dependency” does not violate Rule 137(5) EPC. As previously explained, this provision requires two criteria to be fulfilled to warrant the deletion of a dependent claim. The requirement of not combining with amended claim 1 to form a single inventive concept is not met for amended claim 2 in our example. Specifically, amended claims 1 and 2 have in common the subject-matter of amended claim 1. Further, this approach is also in line with the search principles of the Guidelines for Examination, Part B, III, 3.7 (see the above-mentioned second guiding principle). Since the subject-matter of (amended) independent claim 1 is novel and inventive over the prior art, the subject-matter of (amended) dependent claim 2 should be considered searched, although no actual search may have been carried out for the feature(s) of original claim 3.
Defending “unsearched” claims
A central underlying aim of the requirement of unity lies in preventing grant of a single patent (separately) protecting two (or more) inventions. Keeping the two identified guiding principles in mind, one may conclude that claims depending on a novel and inventive claim share a common inventive concept, and that a dependent claim, although no specific actual search has been carried out for that claim, may at least be deemed to have been searched.
The question whether a claim has been actually searched or not should not be crucial in deciding whether a claim ought to be deleted. Considering the legal situation as a whole, rather than focusing on isolated provisions, one may conclude that an unsearched claim should be deleted only when there is no connection to the single inventive concept addressed by the originally searched claims.
Moreover, the question whether a claim is to be considered searched is a legal rather than a factual question. This should open up the possibility to argue that claims should be considered searched by fiction, i.e. that they are deemed to have been “searched”.
________________________
To make sure you do not miss out on regular updates from the Kluwer Patent Blog, please subscribe here.
It is nothing new that application of R 137(5), in the past R 137(4), is a regular source of problems between applicants and examining divisions.
Either examining divisions have not been correctly trained in the matter, which is difficult to believe, or they are simply misunderstanding, not to say misusing, R 137(5).
It is perfectly correct to say: “Stating that an amended claim contains unsearched subject-matter is not sufficient for an Examiner to justify insisting on a deletion of the claim based on Rule 137(5) EPC.”
If in the presence of a set of claims 1+2 and 1+3, if claim 1 is not allowable for lack of novelty or inventive step, it might be possible to claim that the remaining claims 1+2 and 1+3 lack unity, for lack of special technical features making a contribution over the existing prior art.
If claim 3 was as well depending of claim 2, then the applicant is perfectly entitled for claim 1+2+3 to be retained in a set of amended claims. There is no legal basis in the EPC for requesting the features of claim 3 to be deleted from the set of claims, even if indeed claim 3 has been unsearched “as such”.
If instead, during examination, the applicant would continue with one of the two claims 1+2 or 1+3, the latter is perfectly entitled to file a new dependent claim 3’ replacing the original dependent claim 3 in which feature F is taken over from the description. It is true that the subject-matter of claim 3’ will be “unsearched” “as such”, but the applicant cannot be requested to delete this new claim 3’ and forbidden to continue with claims 1+2+F or 1+3+F. if such a claim is indeed later filed, it does not matter that feature F has been “unsearched”.
The above reasoning also applies in case of non-unity a priori. Let’s say that the original disclosure relates to a locomotive and the after shave of the locomotive driver Both can be claimed in the form of independent claims, or in the form of a single independent claim in which a locomotive or the after shave of the locomotive driver are claimed as alternatives. Nobody would deny that there is a lack of unity between the locomotive and the after shave.
Should however the applicant file an amended claim comprising the locomotive and the after shave in a single independent claim, the examining division cannot say that this claim relates to “unsearched” subject-matter as a whole. In the absence of alternative subject-matter in a claim, it is not permitted to claim a lack of unity between all the features of an independent claim or between an independent and a dependent claim.
The problem is that in such situations, be it after a lack of unity a priori or a posteriori, the examiner in charge of the file may have to carry out a further search, but without getting a reward for such a further search. In case of a clear lack of unity, the examiner gets only rewarded for each search for which search fees have been paid.
It’s sad to say, but for the examiner, the only searches that matter and count are searches which bring points. Having to carry out a search for which he does not get rewarded is something which is not good for an examiner. He is thus minded to require deletion of any such “unsearched” matter, as this will avoid him to possibly have to carry out a search for which he will get no points!
There is a whole range of decisions of the Boards of Appeal in which subject-matter was indeed “unsearched”, but which had to nevertheless been to be taken into account and for which eventually a further search was due, even if it did not bring points.
The most recent one is T 1503/13. See also T 558/14, also quite recent. In T 789/07, the problem of unsearched matter which is nevertheless deemed to have been searched is clearly exemplified. In T 926/09, the size of the original description (900 pages) was not a reason to apply at the time R 137(4), now R 137(5). T 173/06 is also revealing in this respect.
Another form of abuse of R 137(5) can be found in the attitude of the examining division having led to T 359/11.
The conclusion reached at the end of the blog: “This should open up the possibility to argue that claims should be considered searched by fiction, i.e. that they are deemed to have been “searched”” is perfectly correct. It is high time that the quest for points should not end up in the type of “misinterpretation” of R 137(5) which has crept up in the mind of some examining divisions.
On the application of R 137(5), there is also interesting reading in epi information 2018/02:
https://information.patentepi.com/issue-2-2018/the-core-of-rule-137-5-epc.html
Decision T 820/15 illustrates perfectly the topic brought forward in the blog about the misuse of R 137(5).
https://www.epo.org/law-practice/case-law-appeals/recent/t150820eu1.html
In spite of a very long and detailed argumentation about the reasons why the application was to be rejected under
R 137(5), the Board of Appeal did very short of it and dismissed the argumentation of the examining division.
What the applicant did was indeed to limit his claim with further features. As those features had not been searched, the ED decided that R 137(5) applies. This is certainly not correct.
The examining division spent 8 pages in its decision on R 137(5). This is very unusual. If the situation had been so crystal clear as the examining division wanted the applicant to believe, it would not have needed 8 pages of drivel to come to a dodgy conclusion on the matter
The Board of Appeal was not very impressed and came to the conclusion that the feature added into claim 1 “had been appropriately reflected in dependent claims 3, 11, 19 and 27 as originally filed”. The BA did not say it, but the conclusion is clear: R 137(5) has been grossly misused.
Some refresher on the correct application of R 137(5) should be envisaged by EPO’s management.
Rather than spending time on an objection manifestly without merit, the examining division would have been wiser and carried out a further search.
Incidentally, the board quoted two further pieces of prior art, and arrived on the basis of these documents to a lack of inventive step.
Without checking in great detail, it seems that the examining division could have found those documents, rather than spending its efforts in denying what could not be denied.
I am not saying it is good or bad that the boards come up with new prior art, but it happens regularly on appeal in ex-parte procedures. One might consider that it is not the job of a board to do the job of a search division, but it shows that sometimes the quality of the search is not of the standard it should be. Art 111 gives the boards the competence to do so.
In appeal on inter-partes procedures, the boards never bring in new documents.
Thank you for both of your comments, which I found very useful.
The contribution of Mr. Bausch, Mr. Wyder and Mr. Steinmuller is very interesting and highly relevant. I recently wrote an article about Rule 17(5) first sentence (epi Information 2018/2). I realize very well my article was published only after the article of Bausch et al. in Patents & Licensing, February 2018 (http://www.iplcom.jp/back_issues.html), I don’t have access to that journal. Rule 137(5) EPC is apparently a hot topic.
I would say that legal basis is a bit different. GL H-II 7.2 says that “If the application is restricted to an unsearched but originally claimed invention, it can be refused under Rule 64 in line with G 2/92 […]. Rule 137(5) cannot be invoked. It does not apply when the applicant has not paid the search fee in respect of a non-unitary invention relating to the originally filed claims.” The key word is “cannot”. Hence, Rule 137(5) EPC does not apply in case “when the applicant has not paid the search fee” even though he was invited to do so. In my view, Rule 137(5) EPC is even more clearly not applicable when the EPO refrained from requesting additional search fees to be paid. The scope of Rule 137(5) first sentence EPC is restricted to features taken from the description (or the drawings), as is precisely the title of GL H-II 6.2. discussing Rule 137(5) (s1).
As I wrote in my (posterior) article, I think that: “Rule 137(5)(s1) was intended to complement G 2/92, not to replace it. [see e.g. T708/00, r.7] Accordingly, Rule 137(5)(s1) does not cover the cases that are already dealt with by G 2/92. Hence, Rule 137(5)(s1) is not the legal basis for objecting to claims, if the objection is based on the non-payment of a requested additional search fee. This is confirmed by established case law (Case Law of the Boards of Appeal 2016 (CLBA) IV.B.5.3.1, T 1285/11, r. 2; T 442/11, r. 2.2.4) and is followed in the Guidelines paragraph H-II 7.2 where G 2/92 is discussed.”
I would be glad to learn if the authors have a different view.
So for the authors’ example with claims 1, 2 and 3: I understand that these were all in the claims as filed. Therefore, Rule 137(5)(s1) does seem to be applicable at all. This follows also from the wording of the rule, because unity of invention of any one or more of the claims as filed (for a Euro-direct application) is sufficient.
In other words, I think that the statement “only when there is no connection to the single inventive concept addressed by the originally searched claims” is not entirely accurate. For the “unity” part of Rule 137(5)9s1), the original claims are by no way discounted if the EPO decides to not search them.
I would say that “one may conclude that an unsearched claim should be deleted under Rule 137(5)(1) only when there is no connection to the any one or more of the inventive concept(s) addressed by any one or more of the original claims” (for Euro-direct applications, for Euro-PCT applications the claims pending at the time of the search by EPO are decisive).
My conclusion is as of yet that Rule 137(5)(s1) aims for nothing more than preventing applicants from hiding additional inventions in the descriptions to prevent the Examiner for requesting additional search fees for them. However, I am absolutely open to learning about different views. Rule 137(5)(s1) is certainly not an easy rule to interpret.
For the example with claims 1, 2 and 3: I think that the Examiner/search division could have asked for additional search fees because the claims were pending at the time of the search. For the case wherein an additional search fee was requested and the applicant decided not to pay it, G2/92 decided that “cannot pursue that application for the subject-matter in respect of which no search fees have been paid”. Whether or this includes “pursue that application for the subject-matter in the dependent claims” is an interesting question which I think is still open. For the case that the search division refrained from asking additional search fees, I think that the EPO simply has no legal basis for objecting to the subject-matter during examination. The logic of G2/92 seems to be that the applicant must live with the consequences of his decision to not pay the additional search fee when asked to do so (or file a divisional), this does not support sanctions for applicants who are not even given the opportunity to pay additional search fees. Even if the search division did not ask for additional search fees for dependent claims which go in completely different directions because such claims need not be searched according to GL B-III 3.7. The EPO in my view has the possibility to search dependent claims if it minds doing so, even if claim 1 is novel and inventive. If the features in the dependent claims address completely different technical problems than the features of claim 1, I think the case can be made that payment of additional search fees can sometimes be justified but legal basis is not so clear.