Full decision of Enlarged Board of Appeal in partial priority / poisonous divisionals referral (G01/15) published

No more from divsionals
On 29 November 2016, the Enlarged Board of Appeal at the EPO had issued its order in case G 1/15, but not yet its decision. The Enlarged Board has said:

Under the EPC, entitlement to partial priority may not be refused for a claim encompassing alternative subject-matter by virtue of one or more generic expressions or otherwise (generic "OR"-claim) provided that said alternative subject-matter has been disclosed for the first time, directly, or at least implicitly, unambiguously and in an enabling manner in the priority document. No other substantive conditions or limitations apply in this respect.

A generic "OR"-claim is one that covers multiple alternatives without spelling them out. If such a claim can have partial priority to an earlier application that disclosed one of those alternatives, then the implication is that the publication of the earlier (priority) application cannot deprive the later application of novelty under Article 54(3) EPC.

The full decision with reasons has been published now. You should read the whole thing, here are a few key reasons:
5.1.3 If a claim in the later application is broader than an element disclosed in the priority document, then priority may be claimed for such element but not for all other embodiments encompassed by the claim or claims. This principle applies for each individual element disclosed in a priority document. It does not matter whether partial priority is claimed for one element in a priority document only, for a plurality of elements disclosed in one priority document (first Situation addressed in Article 88(3) EPC), for a plurality of elements disclosed in more than one priority document (second situation addressed in Article 88(3) EPC) or for a plurality of elements disclosed in a plurality of priority documents (situation addressed in Article 88(2), second sentence, EPC). It is also of no relevance whether one claim for which partial priority is claimed encompasses only one element disclosed in a priority document or a plurality of elements disclosed in one or more priority documents. The situation where one claim encompasses elements disclosed in a plurality of priority documents is specifically addressed in Article 88(2), second sentence, EPC.

The decision then goes on to quote extensively from the FICPI Memorandum on multiple priorities and endorses its approach and states in 5.3:
5.3 It is thus to be concluded that the EPC does not contain other requirements for recognising the right of priority beyond the same invention, whether for simple, multiple or partial priority, the last being regarded as a sub-group of multiple priorities.
As a consequence, the proviso laid down in G 2/98, (supra, Reasons point 6.7, last sentence), cannot be construed as implying a further limitation of the right of priority.

The state of poisonous divisionals
Reasons 6.4 and 6.5 explain how to assess whether subject matter within a generic "OR" claim may enjoy partial priority:
 6.4 In assessing whether a subject-matter within a generic OR" claim may enjoy partial priority, the first step is to determine the subject-matter disclosed in the priority document that is relevant, i.e. relevant in respect of prior art disclosed in the priority interval. This is to be done in accordance with the disclosure test laid down in the conclusion of G 2/98 and on the basis of explanations put forward by the applicant or patent proprietor to support his claim to priority, in order to show what the skilled person would have been able to derive from the priority document. The next step is to examine whether this subject-matter is encompassed by the claim of the application or patent claiming said priority. If the answer is yes, the claim is de facto conceptually divided into two parts, the first corresponding to the invention disclosed directly and unambiguously in the priority document, the second being the remaining part of the subsequent generic OR"-claim not enjoying this priority but itself giving rise to a right to priority, as laid down in Article 88(3) EPC.
6.5 This also corresponds, logically and exactly, to the scheme described in the Memorandum (see point 5.2 above): "If a first priority document discloses a feature A, and a second priority document discloses a feature B for use as an alternative to feature A, then a claim of the application directed to A or B will in fact consist of two distinct parts A and B respectively, each complete in itself...", and further: "... it would be appropriate to claim a partial priority in situations corresponding to the "OR"-situation under Multiple Priorities", the European patent application itself taking the place of the second priority document.
6.6 The task of determining what is the relevant disclosure of the priority document taken as a whole, and whether that subject-matter is encompassed by the claim in the subsequent application, is common practice in the EPO and among practitioners of the European patent system and as such should not pose any additional difficulty. Nor does it create uncertainty for third parties, as argued by the respondent and in some amicus curiae briefs. Although it can be a demanding intellectual exercise, the decisions reached in cases T 665/00, T 135/01, T 571/10 and T 1222/11 all show that it can be carried out without any need for additional tests or steps.
6.7 From this analysis it follows that the assessment of entitlement to partial priority right does not show that any additional requirements are needed.
From 6.4 it seems to follow that poisonous divisionals are truly dead. The floor (comment section) is yours.
Full decision of Enlarged Board of Appeal in partial priority / poisonous divisionals referral (G01/15) published Full decision of Enlarged Board of Appeal in partial priority / poisonous divisionals referral (G01/15) published Reviewed by Mark Schweizer on Thursday, February 02, 2017 Rating: 5


  1. Tufty's law recognised by the EBoA.

  2. Pseudononymouse

    Some sense at last!

  3. How many years for common (legal) sense to prevail? One despairs but at least they (appear) to have got there in the end.

  4. I cannot but help feel that the EBA has decided what should be the "right" answer and crushed the meaning of the EPC to fit.

  5. @Kant

    Surely that is preferable to a literal interpretation that led to a ridiculous result?

    Laws should be fit for purpose, there is no point treating the EPC as an infallible text

  6. Shame they dodged the divisional question. What happens if priority is withdrawn for the parent case? Can the div be cited?

  7. I didn't expect them to call it Tufty's law (I didn't call it that anyway), but it's nice to have been proven right. I've written a bit more about it here.

  8. @Kant

    I would instead say the EBoA have purposively construed the EPC, having due regard for the intentions of those drafting it (FICPI memo etc).

  9. I do not know whether the explanation in the paragraph 21-24 of FICPI Memorandum is consistent with the reasons for the decision or not. The explanation mentions not secret prior art but public us of the invention. Speaking precisely I do not know whether the EBoA had not found the explanation in FICPI Memorandum or disregard it intentionally. Everyone please read the paragraph 21-24 of FICPI Memorandum!

  10. What happens now if my first filing recites chlorine. I make a later filing directed to halides, claiming priority from the first filing, and I use the second filing as the priority filing for my EP application. Is the EP priority valid or is the priority invalid because of the first filing with the EP priority application not being the "first application"? Using the logic of G1/15, I would say I can notionally cut around the chlorine part and leave my priority claim intact.

  11. @Kant

    Valid for the non-Chlorine part. Invalid for the Chlorine part.

  12. Anon at 14:12 - If you assume that the FICPI Memo considered only novelty but not inventive step, then I don't see an inconsistency between the present decision and paragraphs 21-24 of the Memo. Novelty over a public use is the same as novelty over secret prior art.

    Under the present decision, the claim for a halogen would have the first priority for chlorine (because 'halogen' encompasses chlorine). And it would have the second priority for the other halogens (because 'halogen' encompasses those others). That is the same as the Memo, for a country which permits multiple priorities for one and the same claim.

    There could be a different end result if you also consider inventive step. If chlorine was publicly disclosed between the first and second priority dates, maybe that would render bromine, iodine or fluorine obvious (depending on the facts of the case).

    So there is a difference over section 52 of the UK Patents Act 1949. But we already knew that before the present decision. And I don't think the FICPI Memo considered inventive step.

  13. Kant at 14:56 - Using the logic of G1/15, I would say I can notionally cut around the chlorine part and leave my priority claim intact.

    I agree that you can notionally cut around the chlorine part. You have priority for other halogens but not for chlorine. However, since your priority claim does not cover chlorine, it is only partially intact.

  14. @Kant:
    I think so too, except that cutting around the chlorine part means a hole remains in your priority claim.

    Your claim to halides will be conceptually divided into "chlorine" (no valid priority) OR "halides minus chlorine" (priority from the second filing).

    If chlorine happens to be made obvious by an intermediate disclosure, your claim has a problem.

  15. Tim Jackson at 17:14 - It is so natural that the FICPI Memo considered inventive step, I think. We can find the phrase “There will be rejected as frivolous, as just being an exhaustive list of the halogens which any school boy can find in his elementary text book” in the 18th paragraph of the FICPI Memo.

    What is ground that you don't think the FICPI Memo considered inventive step?

    If the FICPI Memo had been drafted in 1940, I can understand. At that time even the UK, which established requirement of inventive step independently from novelty for the first time in the world, had no inventive step requirement. In 1968 France established requirement of “activité inventive,” which is equivalent to inventive step. When the FICPI Memo was drafted in 1973, almost all countries at least in Europe already had inventive step requirement, I think. So I think that the FICPI Memo considered inventive step.

  16. The "they will be rejected as frivolous" sentence in par. 18 is just a hypothetical lack of conciseness (Art. 84 EPC) objection that a patent office might make against a large number of independent claims, each claim directed to another halogen. The "if these sub-claims are so rejected" in par. 19 refers to such a lack of conciseness rejection.

    The point FICPI is making in pars. 17-20 is that having to draft a separate claim for each "alternative subject-matter" is not only inconvenient for the drafter but might also create problems with a patent office that does not see any point in a proliferation of independent claims (in the absence of multiple priorities for a single claim, the point would be to preserve priority rights, but there is always a chance that an examiner might be slow to appreciate that point).

    So the whole Memorandum is simply asking that applicants not be forced to draft a multitude of independent claims in order to preserve all their priority rights. They could of course draft such claims, but why should they if there is the simple option of allowing one priority per claimed alternative. I guess FICPI's real concern was whether post-grant amendment would be possible in all contracting states, see section III.

    It might be true, in practice, that if the chlorine alternative is obvious (at the filing date of the second priority document), then all other halogen alternatives are obvious too (par. 15 explains that the other alternatives were found only after further experiments, though), but that is not what the Memorandum is about. The example was merely intended to illustrate the general point FICPI was making (see my previous paragraph).

  17. Laws should be fit for purpose, there is no point treating the EPC as an infallible text

    So much for the rule of law...

  18. Having flicked back through Tufty's original post, I am at a loss to explain this sudden claim to the EPO applying Tufty's law. Tufty agreed with an expert witness who stated that a parent and divisional are one and the same application, but this is not a reason given by the Enlarged Board. Other commentators discussed the 'partial priority' solution in their suggestions over 'metal' being read as 'metal minus copper' for the purpose of assessing the priority claim.

    Let's not label this "Tufty's Law", but let's accept that no new law has been created.

  19. It would have been nice to have a comment on question 5, too.
    See the comment of Anon of Friday, 3 February 2017 at 07:32:00 GMT.
    More importantly, it seems that a priority document may disclose
    subject matter that is novelty destroying to and not encompassed
    by a claim of a subsequent patent/application. In such case the
    same problem would rise again.

  20. Anon at 07:24, can you give an example of subject matter that would be novelty destroying but not encompassed by the claim? It seems to me that "encompassed by" aligns the disclosure test for partial priority with the disclosure test for novelty. So if you are entitled to partial priority, it exactly counters the novelty attack. And if you are not entitled to partial priority, the novelty attack will fail too.

  21. @Anon of 5 Feb, 7:24 GMT:
    If subject-matter is novelty destroying for a claim, then by necessity the claim encompasses that subject-matter.

    @Anon of 4 Feb, 19:20 GMT:
    Since G 1/15 goes against several aspects of G 2/98, I would say it creates new law. I guess one could also say it clarifies G 2/98.

    I think this post is the origin of "Tufty's law" (see also here) and it certainly does propose the partial priority solution. As Tufty admitted already in that post, he was helped by comments to earlier posts. At least some of the earlier suggestions were based on the idea of applying a G 1/03-type disclaimer (after all, the poisonous part is 54(3) prior art). Unfortunately G 2/10 seems to rule that out since one would typically have to disclaim disclosed subject-matter.

    As far as I know, "priority" for the partial priority solution goes to the board deciding T 1222/11. But earlier disclosures of the idea may exist in literature or blogs...

  22. Anon of 4 Feb at 07:21 - What is ground that you don't think the FICPI Memo considered inventive step?

    Previously I suggested that if chlorine was publicly disclosed between the first and second priority dates, then maybe that would render bromine, iodine or fluorine obvious, depending on the facts of the case. This assumes that the FICPI Memo was only considering novelty.

    Paragraph 23 of the FICPT Memo says that if multiple priorities are allowed for one and the same claim, then the halogen claim would be valid, even if there was public use of the chlorine embodiment between the two priority dates. If that was considering inventive step as well as novelty, then it would contradict my suggestion.

    However, in that case, the FICPI Memo would also contradict Decision G3/93. Bromine, iodine and fluorine cannot validly claim the first priority date. In accordance with G3/93, therefore, the disclosure of chlorine is part of the state of the art under Article 54(2) EPC and can be cited for inventive step, to the extent that the claim covers bromine, iodine or fluorine.

    Actually, I think it would be quite nice if the FICPI Memo meant that G3/93 was wrong. It would be safe again for inventors to publish the content of their initial priority document in the priority interval. Decision T301/87 found that was OK, but G3/93 overruled it.

    However, this would take another Enlarged Board decision. I see no prospect of that happening. They would need to overrule G3/93. Or possibly they could say that G3/93 is limited to the AND-claims which it actually considered, and doesn't apply to OR-claims.

  23. I don't believe the decision has gone against G2/98, and the law applied is basically the Paris Convention from a long long time ago, so no new law created by the Enlarged Board.

    At least we agree that it is not Tufty's Law.

  24. In response to e.g. Tim Jackson, it seems that
    e.g. a process done at a temperature 50 ... 100 is
    not encompassed by a process done at a temperature 70 ...120.
    The latter does not encompass the part 50 ... 70.
    Moreover the former is novelty destroying prior art.

  25. 50 - 70 is not encompassed, but neither is that part novelty destroying. The part which would destroy novelty is 50 - 100, but that enjoys partial priority.

    The first step of G1/15's process is to determine which subject matter is relevant.

  26. Excuse me, the relevant part which would destroy novelty (but enjoys partial priority) is 70 - 100.

  27. I beg to differ. 70 -100 is an "and" claim in view of the original 50 - 100.
    According to the Memorandum, which G1/15 seems to agree on, "and" claims cannot
    enjoy partial priority. To say the least, this matter seems to still be
    open for discussion.

  28. Tim Jackson of 5 Feb at 15:31 – Thank you for your pleasant comment for me.

    In G3/93, the Enlarged Board stated as follows:

    1. A document published during the priority interval, the technical contents of which correspond to that of the priority document, constitutes prior art citable under Article 54(2) EPC against a European patent application claiming that priority, to the extent such priority is not validly claimed.

    2. This also applies if a claim to priority is invalid due to the fact that the priority document, and the subsequent European application, do not concern the same invention because the European application claims subject-matter not disclosed in the priority document.

    This provision is definitely applicable to AND-claim, because AND-claim has no priority. However, there is some ambiguousness whether the provision be applicable to OR-claim because OR-cliam has at least partial priority.

    I guess that Beloit v Valmet made the ambiguousness clear, namely the judgement introduced the extended the above-mentioned provision of G3/93 to OR-claim. Black Book also explains well, I guess.

    Now I am requesting these documents to my acquaintance in Europe.
    Anyway I am studying legal framework in the UK.

  29. I don't see the temperature range 70 - 100 as an AND claim.

    An AND claim would include (feature A) AND (feature B). The reason it's a narrower claim than one for feature A alone is because it would only cover embodiments which included both features A and B together.

    In the case of these temperature ranges, that's not the reason why 70 - 100 is narrower than 50 - 100. An embodiment might use 80 degrees and fall into both ranges, or it might use 60 degrees and fall into one but not the other. But no embodiment uses both 60 and 80 together.

  30. Note that 50...100 is novelty destroying for 70...120 because the range 50...100 discloses the endpoint 100 and it is that endpoint that destroys novelty of (and is encompassed by) 70...120.

    If the priority document discloses 50...100, then a claim to 70...120 will have partial priority for 100. I don't see why it would have partial priority for 70...100, because 70...100 is not subject-matter that is disclosed by 50...100.

  31. Either 70...100 is subject-matter that is disclosed by 50...100, so it is encompassed by 70...120 and gives partial priority.

    Or 70...100 is not subject-matter that is disclosed by 50...100, in which case neither would it be novelty destroying for 70...120. (E.g. perhaps there is a selection invention.)

    G1/15 unifies the disclosure tests for partial priority and novelty. So in a potential 'poisonous priority' situation, they cancel out.

  32. I would like to emphasize, that according to G1/15 the right
    to priority operates to exclude self collision
    of _identical_ subject matter (see r. 4.4).
    As I see it, this should be read as identical subject matter _only_.

    In the present example, I can not see any identical subject matters.
    However, it would be possible to conceptually divide the
    later invention (70-120) to three parts, whereby the single point 100
    would be identical subject matter, as indicated by the previous anonymous.

    Moreover, the Board was well aware that all novelty-destroying prior
    art will not necessary enjoy (partial) priority. As indicated in
    reasons 4.3.4, novelty objection can succeed when priority can
    not be recognized.

    To me it seems that if you put e.g. these two _different_ inventions
    to a same application, you may arrive at poisonous divisional,
    provided that only one of them has the right to priority.

  33. Tim, you have to consider the endpoints of the ranges. A range discloses its endpoints. An endpoint (or any other individually disclosed value) is novelty destroying for any range that it belongs to / is encompassed by. So 50...100 is novelty destroying for 70...120 simply because 100 is.

  34. Anon on 8 Feb at 08:47 -
    "So 50...100 is novelty destroying for 70...120 simply because 100 is"

    Anon on 7 Feb at 19:45 -
    "If the priority document discloses 50...100, then a claim to 70...120 will have partial priority for 100"

    Me on 7 Feb at 23:16 -
    "G1/15 unifies the disclosure tests for partial priority and novelty. So in a potential 'poisonous priority' situation, they cancel out."

    Perhaps we should stop agreeing so vehemently?

  35. The anon who stated the following got it spot on. The end point of 100 is the only subject matter disclosed in the priority document that is novelty destroying of the latter range, but as this end point has partial priority where priority claimed, there is no problem.


    Anonymous said...
    Note that 50...100 is novelty destroying for 70...120 because the range 50...100 discloses the endpoint 100 and it is that endpoint that destroys novelty of (and is encompassed by) 70...120.

    If the priority document discloses 50...100, then a claim to 70...120 will have partial priority for 100. I don't see why it would have partial priority for 70...100, because 70...100 is not subject-matter that is disclosed by 50...100.


  36. Anon at 06:23 "In the present example, I can not see any identical subject matters."

    We're talking about the subject matter of the priority document, both for novelty and partial priority. Not only is the subject matter identical, it's the same document in both cases.

  37. It is also probably worth to note, that while 100 seems to have the
    partial priority, 100 to 120 (excluding 100) is an extension to
    the original invention and clearly novel, 70 to 100 (excluding 100) is a different
    invention than 50 to 100. Moreover it seems to be a selection invention
    from the original. And yet, it is as close as possible to a known example (i.e. 100).
    So it seems that it cannot be "sufficiently far from known
    examples" (see Case-Law for selection inventions).

  38. A selection invention would be an exception to the general rule that 70...100 is disclosed (made available) by a more generic disclosure of 50...100. If 70...100 is disclosed, then it gives rise to a partial priority. If it's not disclosed then it's not novelty destroying (selection invention). There is no problem.

    This just repeats my comment on 7 Feb at 23:16.

  39. Well, not to bother anyone with the selection part of the new invention, I wish also
    to point out that as indicated e.g. in the decisions T871/08 and T1186/05, the boards
    tend to use, when comparing values of prior art with values of the invention,
    the least amount of significant numbers. In the present case,
    100 is given to the accuracy of only one significant number. Thus, it might be possible
    that even the range 100-120 (excluding 100) would not be novel, since
    120 would be rounded to 100. To emphasize this point, the upper
    limit could be e.g. 104 or 100.4; neither of which is _identical_ to 100,
    but are both _novel_ in view of 100?

    I'm not saying that such claiming would be wise, but it is clearly something that
    can be disclosed.

  40. So, I have a question. Priority application claims process carried out at from 1 to 100 deg C. EP application is narrower - say 2 to 22 deg C. Ignore point disclosures within the range for the time being.

    Using "added subject matter" test, generic claim in the EP application not entitled to priority (had you amended "1 to 100" to "2 to 99" on an EP case it would add matter, so the amended claim should lack priority).

    If the disclosure of 1 to 100 becomes citeable under Art 54(3) - eg because priority application is an EP application which publishes, or via a divisional, does the generic claim in the EP application lack novelty. Applying the case law on selection inventions, range is not "narrow" compared to the prior art, so arguably lacks novelty. Could there be self collision under such circs?

  41. This wouldn't be a generic OR-claim, so G1/15 isn't directly relevant. It would be decided on the basis of previous case law.

    Unless there is a good selection invention, why would you want to limit the scope of your claim in this way, compared to the priority document? The only reason I can think of would be to avoid some newly-discovered prior art, falling within your priority disclosure but published before your priority date. But then you would need to be non-obvious and/or a good selection over that prior art, not just over your priority document.

  42. Interesting decision in that priority is now all about whether the priority document discloses any subject matter which can be used to overcome the intervening disclosure and it does not matter how the claim in the Convention case is phrased as long at it encompasses the relevant matter. So we might find ourselves looking into the experimental details in the Examples of the priority document and comparing them to experimental details in the intervening disclosure, i.e. the analysis might not depend on claim language and that will inevitably be difficult in some circumstance. Claims are clearly worded and have sharp boundaries, and I wonder if the EB realised the implications of not having analysis of the disclosure of the Convention claim as part of the test. Being able to generalise from the Examples of the priority document might now become a much more frequent part of arguing priority entitlement.

    Also for the specific situation where a sequence was corrected at Convention filing (without retaining the incorrect sequence in the Convention specification) this is welcome news. So disclosure of incorrect sequence+% homologues in the priority document will support priority for correct sequence+%homologues as long as the Convention claim encompasses the disclosure in the priority document. [This specific situation happens more often than it should!]


All comments must be moderated by a member of the IPKat team before they appear on the blog. Comments will not be allowed if the contravene the IPKat policy that readers' comments should not be obscene or defamatory; they should not consist of ad hominem attacks on members of the blog team or other comment-posters and they should make a constructive contribution to the discussion of the post on which they purport to comment.

It is also the IPKat policy that comments should not be made completely anonymously, and users should use a consistent name or pseudonym (which should not itself be defamatory or obscene, or that of another real person), either in the "identity" field, or at the beginning of the comment. Current practice is to, however, allow a limited number of comments that contravene this policy, provided that the comment has a high degree of relevance and the comment chain does not become too difficult to follow.

Learn more here: http://ipkitten.blogspot.com/p/want-to-complain.html

Powered by Blogger.