Thursday 18 July 2013

Partial priority - a referral to the Enlarged Board?

Following my earlier post here about the (in my view) flawed judgment of Arnold J in the Nestec v Dualit coffee machine case, there have been further interesting developments on the corresponding European patent EP2103236 (register: 09007962.5). The matter is currently waiting to be heard before the EPO Board of Appeal in case T 1674/12, after a decision by the opposition division upholding the patent in amended form was appealed by the opponents. The points to be considered are similar to those that were considered in the Patents Court, and there is certainly now a large amount of overlap between the cases. Oral proceedings in the European case are scheduled for 9 October 2013 which, if the information here is correct (confirmation here; thanks to Joe) will happen before the UK Court of Appeal gets to hear the arguments on the appeal from Arnold J's judgment.

Each side in the European case have now submitted their arguments, the latest being those from Nestec, provided on 15 July 2013 countering those of the opponents. In brief, as for the EP(UK) case, the arguments primarily concern whether the claims of the patent are entitled to priority, and whether they are valid over an alleged prior public use. The main point, at least as far as I am concerned, is that concerning priority, as regular readers of this blog will already know.

The main issue over priority is whether a broader definition of a feature in the patent claims is entitled to priority and, if not, whether this would make the claim anticipated by the published priority application. The feature relates to whether a housing for receiving a capsule is located in a first or second part of the device. The application claimed the housing without stating where it was located, disclosing that it could be in either or both, whereas the priority application only disclosed that the housing was in the second part. The question is then whether the claim is entitled to partial priority for embodiments where the housing is in the second part, or if the claim is not entitled to priority at all. Arguments have been provided by the opponents for the latter case (of course), whereas Nestec have argued that the claim would be entitled to partial priority. Interestingly, both arguments derive from different interpretations of G 2/98, leaving open the distinct possibility of a further referral to the Enlarged Board.

G 2/98, which has been raised many times in the various discussions relating to the Nestec case and the more general poisonous priority/divisionals issue, considered the question of whether a claim could be entitled to more than one priority date. The board referred to a memorandum that led to the introduction of Article 88 EPC, which explicitly allows a single claim to have more than one priority. The relevant parts of G 2/98, which are worth repeating in full to avoid any confusion, are points 6.5 to 6.7:
6.5 According to the memorandum, in evaluating whether there is any justification for claiming multiple priorities for one and the same claim of an application, a distinction has to be made between the following situations: (i) "AND"-claim (ii) "OR"-claim
6.6 As regards the "AND"-claim (point 6.5(i) supra), it is held in the memorandum that where a first priority document discloses a feature A, and a second priority document discloses a feature B for use together with feature A, "then a claim directed to A+B cannot enjoy a partial priority from the first priority date, because the invention A+B was disclosed only at the date of the second priority document". From this it clearly follows that, according to the legislator, multiple priorities cannot be claimed for an "AND"-claim. Hence, the application of the so-called "umbrella"-theory (according to which the feature A in the claim directed to A+B would enjoy a partial priority from the first priority date, with the result that the feature A could under no circumstances become state of the art in relation to the claimed invention A+B) is to be disregarded. Besides, the application of the "umbrella"-theory would manifestly be at variance with Article 88(4) EPC.
6.7 As regards the "OR"-claim (point 6.5(ii) supra), it is held in the memorandum that where 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 directed to A or B can enjoy the first priority for part A of the claim and the second priority for part B of the claim. It is further suggested that these two priorities may also be claimed for a claim directed to C, if the feature C, either in the form of a generic term or formula, or otherwise, encompasses feature A as well as feature B. The use of a generic term or formula in a claim for which multiple priorities are claimed in accordance with Article 88(2), second sentence, EPC is perfectly acceptable under Articles 87(1) and 88(3) EPC, provided that it gives rise to the claiming of a limited number of clearly defined alternative subject-matters.
The argument in this case is all about point 6.7, in particular whether a broader claim can be considered to be of the "OR" type if it encompasses a narrower priority disclosure. On the face of it, it seems obvious that it would, given that the examples in the memorandum include broadening ranges and applications. There have, however, been decisions by the EPO Boards of Appeal that have taken opposing views on what is meant by "a limited number of clearly defined alternative subject-matters".

T 1877/08, which was cited by the opponents, says that a broader claimed range cannot be given priority to a narrower range, even partially, because G 2/98 requires "separable alternative embodiments" (point 2.4).

T 1222/11, cited by Nestec, says that a broader generic formula can have partial priority because the clearly defined alternatives are the narrow formula and the rest of the scope of the claim (point 11.5.5).

Nestec have therefore argued that, based on T 1222/11, the claim is entitled to partial priority for the housing being in the second part, whereas the opponents argue, based on T 1877/08, that the claim is not entitled to priority at all.

In my view, the interpretation based on T 1222/11 is clearly the right one, not least because the memorandum referred to in G 2/98 includes examples of this type as ones where partial priority would apply. It would also avoid the (also in my view) absurd outcome that the patent would be found to lack novelty over its own priority document. The EPO board in this case will clearly have to decide one way or the other and they are led in two opposing directions by previous T decisions. I therefore see a referral to the Enlarged Board being a distinct possibility, unless the Board consider T 1877/08 (or T 1222/11 of course, as pointed out by the first commenter below) to be a 'rogue' decision that should be ignored. The question to be referred might be of the following form (thanks to the commenter "Joe", who seems to have already thought of it):
Can a generic term in a claim, encompassing species having multiple priority dates, be regarded as defining clearly defined alternative subject-matters (as defined in G2/98) so as to be compliant with Art. 88(2), second sentence?
Would this resolve the issue? I hope so.

48 comments:

  1. Given that T 1222/11 already acknowledges that it deviates not only from T 1877/08, but also from T 476/09, T 1443/05 and T 1127/00, a priori it seems more likely that T 1222/11 will be considered the 'rogue' decision.

    I wonder if Tufty agrees with one of the main conclusions of T 1222/11: if an earlier application discloses a specific embodiment that is novelty destroying for a more generic disclosure in a priority document with a later filing date (by the same applicant), then this priority document is not the "first" application for the generic disclosure, so that a later filed application claiming the generic disclosure cannot validly claim priority from the priority document.

    Although this conclusion has been held valid before, e.g. in 2007 by the Examination Board for Paper C (a quite notorious paper), it is really not in line with G 2/98.

    G 2/98 in fact starts from the premise that "same subject-matter" in Art. 87(4) is to be interpreted narrowly (point 2). The main question in G 2/98 was whether "same invention" in Art. 87(1) is just as narrow. The answer was affirmative (point 5).

    G 2/98 is also explicit regarding the test to be applied:
    "The requirement for claiming priority of "the same invention", referred to in Article 87(1) EPC, means that priority of a previous application in respect of a claim in a European patent application in accordance with Article 88 EPC is to be acknowledged only if the skilled person can derive the subject-matter of the claim directly and unambiguously, using common general knowledge, from the previous application as a whole."

    This is the same test as for Art. 123(2). This means that if an earlier application discloses "copper" and a later filed priority application by the same applicant discloses "iron", then this is not the "same subject-matter" in the sense of Art. 87(4), so that priority can validly be claimed for "iron". But T 1222/11 denies this.

    There are many other ways of pointing out the flaws of T 1222/11, but I think it is better to look at one at a time.

    Maybe it helps to point out that T 1222/11 is not the only possible road to a solution for the problem of poisonous divisionals.

    ReplyDelete
    Replies
    1. Ok, points taken. It may well be that T 1222/11 is the 'rogue' decision, especially if we are counting votes in terms of decisions for and against, but it still seems correct to me and I think it is worth getting an Enlarged Board to look at the issue.

      The conclusion you mention about T 1222/11 does not seem right to me, or at least not clearly expressed.

      Delete
    2. The conclusion is not correct, but T 1222/11 does draw it.

      The situation in T 1222/11 is as follows. The content of the priority document is identical to that of the application. D4b is another application filed by the same applicant with a publication date between the priority date and the filing date of the application. The priority date of D4b is earlier than the priority date of the application. D4b is therefore at least prior art under Art. 54(3).

      Since D4b discloses 6 specific compositions falling under the scope of claim 1 of the application, those compositions have to be disclaimed. This is possible, as long as D4b is only prior art under Art. 54(3).

      But the Board reasons as follows. D4b discloses 6 embodiments falling under the scope of claim 1. The priority document, which discloses the subject-matter of claim 1, is therefore not the "first application" for this subject-matter. Priority is therefore invalid, so that D4b is in fact prior art under Art. 54(2). The disclaimers are therefore not allowable.

      The application and the priority document claim "metal". D4b claims "copper". According to T 1222/11, priority is invalid and D4b is novelty destroying 54(2)-prior art.

      This was one of the two big problems of paper C in 2007. This comment sums it up quite thoroughly.

      See also the discussion here.

      Delete
    3. Before I invite Tufty (or anyone else) to comment on a more direct rebuttal of the "multiple priorities" theory of T 1222/11, it might be interesting to note that the Board for T 1674/12 has been extended to 5 persons, the reasons being "des questions juridiques compliquées (priorité, vices substantielles de procedure)".

      T 1222/11 relies on example (c) of the Memorandum and in particular the following part:
      If multiple priorities for one and the same claim are allowed, it will suffice in the European patent application to draw up a claim to a method for coating the inner wall of hollow bodies. If multiple priorities for one and the same claim are not allowed, the applicant will have to draw up two parallel claims, one directed to a method of coating the inner wall of pipes, which claim will enjoy the first priority, and a second claim directed to a method of coating the inner wall of hollow bodies not being pipes, which claim would enjoy the second priority.

      It is important to note that P1 disclosed "coating pipes" and P2 disclosed "coating hollow bodies".

      Suppose the applicant had in fact included in the European patent application two parallel claims, a first claim directed to "coating pipes" and a second claim directed to "coating hollow bodies not being pipes". Obviously the first claim would have validly claimed priority from P1.

      My question is: would the second claim ("hollow bodies not being pipes") have validly claimed priority from P2 ("hollow bodies")?

      The drafter of example (c) of the Memorandum stated it would. But would it?

      Delete
    4. Myshkin,

      I think you might be going too far with the conclusions you draw from T 1222/11.

      I would see it as follows:

      D4b discloses "copper".
      Priority document discloses "metal"
      Application claims "metal".

      Thus, under the Board's analysis in T 1222/11, only subject matter relating to Copper was not entitled to priority. The rest of the claim was (all metals but copper), hence the Board's reference to multiple priorities.

      In T 1222/11, the fact that copper wasn't entitled to priority meant that, for that subject matter, document D4b was 54(2) EPC. Accordingly, no disclaimers allowed.

      However, I still believe that, had there been an intervening document disclosing "iron", the Board would have found the claim entitled to priority (since all metals but copper are entitled to priority) and therefore would not have found it relevant to the assessment of patentability.


      According to T 1222/11, priority is invalid and D4b is novelty destroying 54(2)-prior art.

      Delete
    5. Would a second claim to "hollow bodies not being pipes" validly claim priority from P2 in the FICPI Memo?

      You would have to have had the prescience to draft P2 so that it provided support for hollow bodies not being pipes. That's an obvious pitfall.

      But that merely adds to the point FICPI was making. If you don't allow multiple priorities for one and the same claim, you need to carefully draft separate claims, and there are problems. If you do allow multiple priorities, these problems fall away.

      The Munich Diplomatic Conference accepted FICPI's arguments, and allowed multiple priorities for one and the same claim.

      Delete
    6. @Anonymous:
      Thus, under the Board's analysis in T 1222/11, only subject matter relating to Copper was not entitled to priority. The rest of the claim was (all metals but copper), hence the Board's reference to multiple priorities.

      Ok, T 1222/11 indeed phrases it as follows:
      "In view of the above, the subject-matter defined in amended claim 1 in positive terms (i.e. without any disclaimer) is not entitled to priority as far as far it concerns the subject-matter already disclosed in D4b."

      But how could the other part of claim 1 ("all metals but copper") be entitled to priority? The priority document does not disclose "all metals but copper".

      Or is it sufficient that one can also "intellectually envisage" the subject-matter "all metals but copper" within the scope of the disclosure of "all metals" in the priority document? Does it work both ways?

      Delete
    7. @Tim Jackson:
      You would have to have had the prescience to draft P2 so that it provided support for hollow bodies not being pipes. That's an obvious pitfall.

      I agree. According to G 2/98 a claim to "hollow bodies not being pipes" is only entitled to priority from P2 if P2 discloses "hollow bodies not being pipes".

      Note however that the Memorandum clearly states that "hollow bodies not being pipes" in the application would be entitled to priority from P2:
      "If multiple priorities for one and the same claim are not allowed, the applicant will have to draw up two parallel claims, one directed to a method of coating the inner wall of pipes, which claim will enjoy the first priority, and a second claim directed to a method of coating the inner wall of hollow bodies not being pipes, which claim would enjoy the second priority."

      According to the Memorandum, there is no pitfall if one allows multiple priorities, but there is also no pitfall if one does not allow multiple priorities. In my view, the only point that the Memorandum makes is that multiple priorities are a convenience in that it reduces the number of claims one has to draft.

      Let's look again at example (c), now allowing multiple priorities and the application having one claim to "hollow bodies". Apparently, this claim can be notionally split into "pipes" and "hollow bodies not being pipes", where "bodies" is entitled to priority from P1 and "hollow bodies not being pipes" is entitled to priority from P2.

      I agree "pipes" would be entitled to priority from P1. But how do you explain that the notional part "hollow bodies not being pipes" would be entitled to priority from P2?

      The point I am trying to make should be obvious: with or without multiple priorities, the Memorandum clearly assumes that one can claim priority from P2 for "hollow bodies not being pipes". But this is contrary to the outcome of G 2/98 (unless P2 actually discloses "hollow bodies not being pipes").

      Delete
    8. P2 discloses hollow bodies generically (and presumably with enough examples to be an enabling disclosure). Under G2/98, the whole of the claim is entitled to the priority of P2.

      Part of the claim (pipes) is also entitled to an earlier priority, from P1.

      So for pipes, you get the earlier P1 date. If you are considering something other than pipes, you don't. But the claim as a whole still has the P2 date, as if P1 had never been filed.

      So if there is an intervening publication of, say, coating the internal surfaces of a bottle, or a teacup, then that is full prior art if published before P2, but not prior art at all if published after P2.

      Your problem with the Memorandum stems from an additional issue which it doesn't consider. However, the whole of the claim is clearly entitled to the P2 date in any case.

      So that additional issue would only arise if multiple priorities for the same claim were not allowed. That's the only time you would have to file a separate claim for "hollow bodies not being pipes". And that's the only time you would need to think about whether P2 provides support for such a claim.

      Had FICPI thought of it, this support problem would have been another reason to urge the Munich Diplomatic Conference to permit multiple priorities for a single claim. But the Diplomatic Conference accepted FIPCI's proposal anyway. So we don't need to worry about it.

      Delete
    9. Myshkin: "But how could the other part of claim 1 ('all metals but copper') be entitled to priority? The priority document does not disclose "all metals but copper". Or is it sufficient that one can also "intellectually envisage" the subject-matter 'all metals but copper' within the scope of the disclosure of 'all metals' in the priority document? Does it work both ways?"

      No, you don't have to intellectually envisage that. The question of whether 'metals not being copper' is entitled to priority simply doesn't arise.

      In the absence of the D4b problem and the disclaimer, the generic claim to metals (claimed in positive terms) is entitled to the priority, and you don't need to worry about 'metals not being copper'.

      D4b means that the claim does not have priority in so far as it covers copper. At this point there is no need to ask about the priority of 'metals not being copper', because the claim does not say that.

      The applicant tries to disclaim copper. Since it's a negative 'undisclosed disclaimer', not a positive claim term, that doesn't actually disclose 'metals not being copper'. So there is still no issue about whether 'metals not being copper' is entitled to priority. Note that an allowable disclaimer does not change the identity of the invention for priority purposes (G1/03 point 4 of the Reasons; compare point 6 of the Reasons of T1222/11)

      The only time it would be necessary to ask whether there is priority for 'metals not being copper' would be if the applicant tried to amend the claim in positive terms, instead of as a disclaimer. But of course, he can't make such a positive amendment since he doesn't have support for it.

      [As an aside, I'll note that T1222/11 is not about metals and copper. That example has come from somewhere else.]

      Delete
    10. I said: "the whole of the claim is clearly entitled to the P2 date in any case."

      I think you are getting bogged down by assuming that if "pipes" in entitled to the P1 date, then it is only "hollow bodies not being pipes" that has the P2 date. And that therefore P2 must support "hollow bodies not being pipes".

      But the reality is that the whole generic claim has the P2 date, as if P1 had never been filed.

      However, this only comes into consideration if you are considering intervening prior art disclosing a hollow body which is not a pipe. If the prior art is a pipe, you have the P1 date.

      Note that example (c) of the Memorandum does not actually say that "hollow bodies not being pipes" would have the P2 date, if multiple priorities are allowed. It does say in example (a) that bromine, iodine and fluorine would have the P2 date, but in that example P2 explicitly discloses bromine, iodine and fluorine.

      Delete
    11. @Tim Jackson (20 July 2013 12:30):
      I agree that "hollow bodies" is entitled to priority from P2, and that "pipes" has priority from P1.

      If you have an intermediate document disclosing a pipe, this is not prior art for the subject-matter "pipes", but it is full prior art for the subject-matter "hollow bodies". So the claim as it stands falls for lack of novelty. Of course it is possible to fall back to "pipes", provided the application discloses "pipes".

      But we still have no answer to my question. The Memorandum is clearly assuming that the subject-matter "hollow bodies not being pipes" is entitled to priority from P2. How can that be?

      I can give some solutions:
      1. The Memorandum silently assumes that P2 discloses "hollow bodies not being pipes". Now the example is completely understandable. It is in fact completely in line with the conventional view that multiple priorities are only possible for a claim defining a limited number of alternative individually disclosed subject-matters.

      2. The Memorandum silently assumes that if a document discloses "hollow bodies" and "pipes", then this implicitly discloses the complement "hollow bodies not being pipes". Now the situation is not different from the previous paragraph: example (c) is completely in line with the conventional understanding of multiple priorities. (G 2/10 decided that this assumption, at least in general, is false, but the drafters of the Memorandum clearly did not know G 2/10.)

      I'll explain again why I think this is so important.

      As I said, there is the conventional understanding of multiple priorities: multiple priorities are only possible for a claim defining a limited number of alternative individually disclosed subject-matters. There is a whole line of case law for this.

      According to T 1222/11, this conventional understanding is not correct. The decision comes up with a quite different view of multiple priorities. T 1222/11 defends this new approach by looking at example (c) of the Memorandum. According to T 1222/11, the Memorandum cannot be explained by the conventional understanding of multiple priorities.

      My point is that if you study example (c) very carefully, you will find an implicit assumption in the example, namely that P2 discloses "hollow bodies not being pipes". This assumption is necessary independent of whether you apply the conventional understanding of multiple priorities, or T 1222/11's understanding.

      Once we make the assumption, the example makes complete sense. It makes complete sense for the conventional understanding, and it makes complete sense for T 1222/11's understanding.

      The important thing is that T 1222/11's core argument is that the example would NOT make sense when applying the conventional understanding. This core argument of T 1222/11 is incorrect.

      Delete
    12. @Tim Jackson (20 July 2013 13:21):
      No, you don't have to intellectually envisage that. The question of whether 'metals not being copper' is entitled to priority simply doesn't arise.

      Note that the Anonymous made the point that "all metals but copper" was entitled to priority. Hence my question.

      Delete
    13. @Tim Jackson;
      I think you are getting bogged down by assuming that if "pipes" in entitled to the P1 date, then it is only "hollow bodies not being pipes" that has the P2 date. And that therefore P2 must support "hollow bodies not being pipes".

      But the reality is that the whole generic claim has the P2 date, as if P1 had never been filed.


      Of course the whole generic claim has the P2 date. But look at what Example (c) states:
      a second claim directed to a method of coating the inner wall of hollow bodies not being pipes, which claim would enjoy the second priority

      The problem considered by the Memorandum is the "problem" of having to draft two claims. The Memorandum's point is that instead of drafting two claims "pipes" (priority P1) and "hollow bodies not being pipes" (priority P2), it should suffice to draft a single claim "hollow bodies" (with priority P1 for "pipes" and priority P2 for "hollow bodies not being pipes).

      If you say "hollow bodies" has priority P2 for the whole and priority P1 for its disclosed part "pipes", then I am fine with that. It just means that if the applicant limits to "pipes", he can get the date of P1. This is how it has always been.

      If you then have a 54(3) document disclosing "pipes" with effective date between P1 and P2, it will be novelty destroying for "hollow bodies" (having priority from P2) and it will not be prior art for "pipes", so you can safely limit to "pipes". You can not keep "hollow bodies". (T 1222/11 and Tufty think you could keep "hollow bodies" because you could read it as "pipes" OR "hollow bodies not being pipes", the second part not being novelty destroyed by "pipes".) I am conveniently ignoring G 1/03 and it's interpretation by G 2/10 now.

      Delete
    14. However, this only comes into consideration if you are considering intervening prior art disclosing a hollow body which is not a pipe. If the prior art is a pipe, you have the P1 date.

      No, the priority date of a subject-matter is NOT dependent on what the prior art discloses.

      If the prior art is a pipe and it is prior art for "hollow bodies" (i.e. it has a date before the P2 date), then it is novelty destroying for "hollow bodies". The only way to cancel it is to argue that you could somehow split "hollow bodies" in "pipes" and "hollow bodies not being pipes" AND that the latter somehow would enjoy priority from P2 (which is problematic even if you DO follow T 1222/11).

      Delete
    15. I don't know if some of the other comments I posted (I think before the one of 27 July 2013 00:53) got lost?

      My main point is this. There is the conventional understanding of "multiple priorities" that a claim only can have multiple priorities if it defines a limited number of alternative individually disclosed subject-matters. With this understanding, any claim with multiple priorities can always be amended within the bounds of Art. 123(2) into a limited number of claims, each claiming subject-matter with a single priority.

      There is a long line of case law supporting this conventional understanding. If a decision like T 1222/11 wants to deviate from it, it better have a good reason.

      T 1222/11 finds this reason in example (c) of the Memorandum in combination with G 2/98's condoning of the Memorandum. I believe T 1222/11 is already reading too much into G 2/98's reference to the Memorandum (this part of G 2/98 really just says "look, the multiple priorities thing doesn't bite our very strict approach" and not "we'll now loosen our very strict line to allow multiple priorities"). But anyway, the reasoning of T 1222/11 is that example (c) cannot be explained by the conventional understanding of multiple priorities. Therefore some other theory of multiple priorities must be found.

      However, if you very carefully study example (c), you will find that it makes a silent assumption: somehow the subject-matter "hollow bodies not being pipes" is entitled to priority from P2.

      Independent of whether you use the conventional understanding of T 1222/11's approach to multiple priorities, the subject-matter "hollow bodies not being pipes" can only be entitled to priority from P2 if P2 itself discloses "hollow bodies not being pipes".

      Now if we assume that P2 discloses "hollow bodies not being pipes", then the application will normally also disclose "hollow bodies not being pipes". The application will also disclose "pipes". So in the application of example (c) we can find the two alternative individually disclosed subject-matters "pipes" and "hollow bodies not being pipes". Therefore we can put them in one OR-claim having two priorities and formulate that claim generically as "hollow bodies". This is completely in line with the conventional understanding of multiple priorities.

      Conclusion: if you study example (c) well, it does not bite the conventional understanding any more than it bites T 1222/11's approach. Example (c) gives no reason whatsoever for dismissing the conventional well-established understanding of multiple priorities.

      Delete
    16. They didn't get lost, it's just that I've been on holiday and nobody took up my offer to moderate.

      Delete
    17. Myshkin

      I would agree that example (c) in the memorandum sets out a situation where:

      (i) P1 discloses "pipes" (i.e. subject matter "A"); and
      (ii) P2 discloses "hollow bodies not being pipes" (i.e. subject matter "B"). I gather this from the statement “a second priority document discloses the use of the same method for coating the inner wall of bottles or ANY OTHER hollow bodies”.

      The claim in the application discloses generic matter “Z” which encompasses “A” and “B”, i.e. limited alternatives.

      I think we’re both on the same page that under the classical G2/98 understanding of priority, an intervening disclosure D1 between P1 and P2 of “A” would not prejudice the novelty of the claim because it can be artificially divided into two claims: (i) “A” and (ii) “B”. D1 is not prior art for “A”, and does not fall within the scope, and therefore cannot prejudice the novelty, of “B”.

      Another way to think about this is to say that:
      (a) you work out whether the disclosure of “A” in D1 falls within the scope of the claim.
      (b) assuming it does, you then look to see what date that subject matter is entitled to.
      (c) in this case, “A” is disclosed within the priority document and so D1 cannot be prior art.

      However, this is clearly defined alternatives. We start to disagree when we move onto generic claims that encompass (for all intents and purposes) unlimited embodiments. Under the classical understanding G2/98, these claims CANNOT ever claim priority (unless the corresponding generic claim is also disclosed in its priority document(s)). This leads to the problem of poisonous divisional applications. As you know, T 1222/11 takes a different approach.

      (As an aside, I would note that T 1222/11 doesn’t use only example (c) to support its case. It also refers to examples (a) and (b), the latter of which relates to numerical ranges, i.e. an instance which the Boards have previously argued does not count as a limited number of clearly defined alternative subject matter (see T 1877/08 and T 476/09)).

      So what about this scenario?

      P1 discloses a specific polypeptide, (i.e. subject matter “A”). The application discloses the specific polypeptide, but only claims the polypeptide by its function (i.e. a claim to a class of polypeptides that encompasses “A”). Since this claim defines an unlimited number of alternatives, the classical understanding of G2/98 would rule that this claim is not entitled to priority from P1. Accordingly, an intervening disclosure of “A” would prejudice the novelty of the claim.

      However, T 1222/11 states that the claim is entitled to priority regardless of how many embodiments it defines. It suggests that that the test is whether you can:

      “conceptually identify a limited number of clearly defined alternative subject-matters by a comparison of the subject-matter of that “OR”-claim with the disclosure of the multiple priority documents, so as to determine which parts of the “OR”-claim or alternative subject-matters are covered by the rights of priority claimed”.

      That is, you compare the subject matter of the claim with the priority document, and if the disclosure of the priority document falls within the scope of the claim, then that subject matter is entitled to priority. In this example, “A” would be entitled to priority and therefore the intervening disclosure is not prior art. This doesn’t strike me as being any different from the example we discussed above. In essence:

      (a) you work out whether the disclosure of “A” in D1 falls within the scope of the claim.
      (b) assuming it does, you then look to see what date that subject matter is entitled to.
      (c) in this case, “A” is disclosed within the priority document and so D1 cannot be prior art.

      Do you have a problem with that example?


      Delete
    18. Myshkin asked “But how could the other part of claim 1 ("all metals but copper") be entitled to priority? The priority document does not disclose "all metals but copper"?”.

      I understand the problem, and it seems to stem from “Tufty’s law”. However, I think when T 1222/11 is applied in the correct manner, then there isn’t actually a problem.

      The analysis goes like this:

      (1) P1 discloses copper
      (2) P2 discloses metal
      (3) Application claims metals.

      Myshkin’s problem arises from an incomplete analysis. By artificially dividing the claim up, you get:
      (i) copper
      (ii) all metals but copper.

      Now once you decide that subject matter (i) (copper) gets the priority date of P1, then you have to question what you do about subject matter (ii) (all metals but copper). Myshkin feels uncomfortable giving subject matter (ii) the priority date of P2 because P2 does not disclose “all metals but copper”. In fact, since nothing discloses “all metal but copper”, there could be an added matter objection too (although G1/03 and G2/10 seem to contain this fear).

      Thus, if we assume that there is an intervening disclosure:
      (ii) D1, which published between P2 and the application, and which discloses iron.

      “Iron” falls within subject matter (ii), and since subject matter (ii) is not entitled to priority, D1 prejudices the novelty of the claim.

      However, I believe this analysis is wrong, and it shows why the thought of artificially dividing up the claim can get you into trouble. A much better approach is the one set out in T 1222/11, which, as I noted above, involves a comparison between the subject matter of the claim and the disclosure in the priority document.

      So the first point to note is that although the subject matter overlaps, they (“copper” and “metal”) are actually distinct disclosures and therefore are each entitled to different priority dates.

      So in this case, “copper” is entitled to the priority date of P1, and “metal” is entitled to the priority date of P2.

      The question of D1 is now easy to solve. Iron falls within “metal”. Since “metal” is entitled to P2, D1 is not prior art.

      I have tried all morning to think of an example where T 1222/11 causes difficulties. The simply way to think of it is that the only difference between T 1222/11 and the classical understanding is that the claim does not have to define a limited number of alternatives.

      Delete
    19. @Anonymous:
      (a) you work out whether the disclosure of “A” in D1 falls within the scope of the claim.
      (b) assuming it does, you then look to see what date that subject matter is entitled to.
      (c) in this case, “A” is disclosed within the priority document and so D1 cannot be prior art.


      This is not at all how priority works (and for this it does not matter whether one follows T 1222/11 or not).

      This is how priority works in case of single priority. You determine the subject-matter of a claim. You determine which earliest priority document discloses exactly that subject-matter (as in directly and unambiguously derivable). The date of this priority document, or else the filing date of the application, is the effective date for this subject-matter. Now the relevant prior art consists of all documents published before this date (and for novelty you add 54(3) documents). Based on this prior art you determine whether the claim is new and inventive.

      In case of multiple priorities, the idea is that the claim defines a limited number of clearly defined alternative subject-matters (G 2/98). For each one of these clearly defined alternative subject-matters you perform the exercise as just described for a claim with single priority. If one of the subject-matters lacks novelty, the claim is not new. If one of the subject-matters lacks inventive step, the claim is not inventive.

      Your approach has no basis in the EPC and is explicitly dismissed in G 3/93. In G 3/93, intermediate document D1 disclosed subject-matter "A" which was also disclosed by the priority document. But the claim was not entitled to priority (since the priority document did not disclose the subject-matter of the claim), so that the effective date for the claim was the filing date and D1 was prior art for the claim.

      I note that "Tufty's law" does not depend on your approach. Tufty's law does depend on being able to equate the subject-matter of a claim to two "clearly defined alternative subject-matters", namely the part that is in the priority document (and is encompassed by the claim) and the complement of this part. T 1222/11 does give support for this. But the way T 1222/11 derives this from the Memorandum is logically flawed, as I have already explained.

      Delete
    20. @Anonymous (29 July 2013 12:11):
      So in this case, “copper” is entitled to the priority date of P1, and “metal” is entitled to the priority date of P2.

      Here I agree.

      The question of D1 is now easy to solve. Iron falls within “metal”. Since “metal” is entitled to P2, D1 is not prior art.

      Here I completely disagree, because this is not at all how priority works under the EPC. And I am pretty sure Tufty agrees with me on this.

      Again: whether D1 is prior art for a particular claimed subject-matter depends only on the date of D1: is the date of D1 before or after the effective date of that claimed subject-matter. The content of D1 is irrelevant for the question whether D1 is prior art.

      See G 3/93, points 8 and 9 of the reasons.

      Delete
    21. Myshkin

      I think I was a little unclear above and a little lazy with my terminology. Let me start off by saying that I’m aware of G 3/93 and am not advocating the approach that was dismissed by the Enlarged Board of Appeal. That case is talking about “AND” claims. Here, I’m talking about “OR” claims.

      I failed to properly explain that I meant a claim that used the generic term “Z” (e.g. halogen) to encompass a number of embodiments “A”, “B”, “C”, “D” (e.g. chlorine, iodine). According to G 2/98, these different embodiments can have different priority dates, provided that the generic term “gives rise to the claiming of a limited number of clearly defined alternative subject-matters”.

      As you state, in order to determine the date of each embodiment within the claim:

      “You determine which earliest priority document discloses exactly that subject-matter (as in directly and unambiguously derivable). The date of this priority document, or else the filing date of the application, is the effective date for this subject-matter.”

      This was effectively my steps (b) and (c) of the test above. Thus, you determine the embodiments “A”, “B”, “C”, “D” that are encompassed by the generic term “Z”, and then determine whether they are disclosed in the priority document. If e.g. the embodiment “A” is disclosed within the priority document, then it is entitled to priority and it has an effective filing date of the priority document. If this filing date is before the date of D1, then D1 cannot be prior art for that subject matter. On the other hand, if the embodiment “B” is not disclosed within the priority document, then it is not entitled to priority and it has an effective filing date of the application as filed. If this filing date is after the date of D1, then D1 is prior art for that subject matter.

      Delete
    22. Where we differ (and where my attempted step (a) comes in) is when the claim is a generic term “Z” that does not embrace a limited number of clearly defined alternatives. An example of such a claim, as I mentioned above, is the functionally defined polypeptide. According to the classical G 2/98 approach, this claim is not entitled to multiple priorities because it does not define a limited number of clearly defined alternative subject matters. A good example of this approach is set out in T 1127/00, paragraph 6:

      “Claim 1 relates to a generic formula which covers a great number of alternative compounds….However, the alternative compounds are not, as such, spelled out in the claim. The fact that they might be intellectually envisaged to fall within the scope of the claim does not make up for a clear and unambiguous presence of these alternatives, individualized as such, in the claim. Claim 1 does not embrace a limited number of clearly defined alternative subject-matters in the form of an "OR"-claim which could be split up into groups of different priorities.”

      I presume that you agree with this outcome. To me, however, this is unfair and unreasonable. The priority document directly and unambiguously discloses a specific embodiment “A” that falls within the generic term “Z” defined within the claim. To me, this embodiment should be entitled to claim priority.

      T 1222/11 attempts to circumvent this problem by coming up with a new test for “clearly defined alternatives”. It says that it doesn’t matter if the claim doesn’t specify “a limited number of clearly defined alternative subject-matters” – the question is whether you can:
      “conceptually identify a limited number of clearly defined alternative subject-matters by a comparison of the subject-matter of that “OR”-claim with the disclosure of the multiple priority documents, so as to determine which parts of the “OR”-claim or alternative subject-matters are covered by the rights of priority claimed”.

      This test makes sense when the claim doesn’t define a limited number of defined alternatives. How can you list out all the embodiments encompassed by the claim and determine their effective filing date if the claim encompasses (to all intents and purposes) an unlimited number of alternatives? You can’t! So instead you look to see whether there are any embodiments disclosed within the priority document that fall within the scope of the claim. These embodiments are entitled to priority. So for example if the priority document discloses a specific polypeptide “A” that is encompassed by the generic polypeptide “Z”, this specific polypeptide “A” is entitled to claim priority. Now if there is a disclosure of “A” after the priority date, it is not relevant prior art for that part of the claim. (I note that it could be relevant for obviousness for the remaining subject matter if it is full prior art).

      The reason I introduced step (a) was because even comparing the embodiments disclosed within the priority document and the claim could be a tedious task, particularly if the priority claim discloses a large number of different embodiments. Thus, for efficiency’s sake, I thought that you should only carry out the analysis for the embodiments disclosed within the prior art. You can scrap it if you like (I don’t want it to draw away from the main point of the discussion, i.e. whether a generic term is entitled to claim priority).

      Delete
    23. I should also add that I think the approach of T 1222/11 and “Tufty’s law” differ. As I mentioned above, Tufty’s law requires the claim to be split up into parts that effectively equal the whole. This leads to the problem that you have identified, which is that it leaves you with embodiments that are not disclosed within the application or priority document (e.g. all metals but copper).

      T 1222/11 doesn’t require the claim to be broken up into equal parts. It just indicates that you effectively compare the disclosure of the priority document (and the embodiments disclosed therein) to the subject matter of the claim. All embodiments that are (directly and unambiguously) disclosed within the priority document and that fall within the scope of the claim are entitled to priority.

      The reason I prefer this approach is because claims are not really suited to being broken up in the manner that Tufty suggests. This is because they can encompass embodiments that overlap (e.g. copper and metal). Claims in essence contain a lot of venn diagrams, which are difficult to separate out in an equal manner.

      An example of how this can lead to problems is the example I set out above where you have a priority document disclosing the specific embodiment (copper) and a priority document disclosing a more generic embodiment (metal). Under Tufty’s law you split out into copper and not copper, but what do you do about metal? Repeat the exercise? This runs into all sorts of difficulties.

      The exercise clearly works better when you follow the approach set out in T 1222/11 (see above).

      On this note, I am a little surprised at your statement:

      “The question of D1 is now easy to solve. Iron falls within “metal”. Since “metal” is entitled to P2, D1 is not prior art. Here I completely disagree, because this is not at all how priority works under the EPC. And I am pretty sure Tufty agrees with me on this.”

      Think about the example if it were two separate claims. Claim 1 directed to copper and Claim 2 directed to metal. Would you consider “Iron” to prejudice the novelty of Claim 2?

      Delete
    24. In reply to the last anonymouse, I think you misunderstand me. I explicitly do not require the claim to be broken up, but suggest that a broad claim is theoretically considered to be in two parts, one of which is entitled to an earlier priority. This seems to me to be exactly the same reasoning as in T 1222/11. If you think otherwise, please re-read my original post on the subject and let me know where any misunderstanding arises.

      Delete
    25. Tufty,

      I understand what you said:

      "The concept that a broad claim can theoretically be divided into two parts, one of which is entitled to priority and the other which is inevitably novel over the priority disclosure, is itself sufficient to overcome the problem even if the exact form of wording is not easily derived from the application. This is because the collective scope of the resulting two imaginary claims would be exactly the same as the original claim, and Article 88(2) EPC indicates that there is no need to actually rewrite the claim to be two separate claims. There is therefore no need to find the right form of wording to satisfy added matter requirements because no amendment is needed."

      You may not be rewriting the claim on paper, but you are at least mentally breaking up the claim:

      "The claim ...can at least theoretically be considered to be two claims: i) a widget made of copper; and ii) a widget made of a metal other than copper.."

      The problem I have with that is when you have the metal and copper disclosures in different priority documents (see my example that I set out on 29 July at 12.11), you can't easily divide the claim into its consituent parts because some of those parts have overlapping scope and different effective filing dates. I don't know what the mentally reimagined claim looks like. Perhaps you can help?

      On a related note, have you read this month's CIPA journal?

      Delete
    26. The point I was trying to make is that it is not necessary to separate a broad claim into into its constituent parts, because however this is done you will inevitably end up with a part of the claim that is entitled to priority, and therefore not anticipated, and the rest that is not entitled but is inevitably novel over the priority document and therefore also not anticipated. This is the 'imaginary claim' concept in my mind; I call it imaginary because it is not necessary to consider what the wording of the separate claims might be. The copper/metal example is only a very simple one used to illustrate this. As I have mentioned several times before, this only applies in situations where A54(3) in relation to the priority disclosure is the relevant issue (i.e. the poisonous priority/divisionals problem). Once any intervening prior art comes into play the general rule cannot be applied as it depends much more on what particular part of the claim is entitled to priority.

      I haven't read this month's journal yet, but having just fished it out of my bag I see there is another article by Malcolm Lawrence, which I guess is what you have in mind. Too many words as usual, and it's only part 1! I would prefer to wait and see what the EPO say about the issue rather than having to wade through another load of pompous obfuscatory verbosity, but I might take a quick look anyway.

      Delete
    27. It's actually quite an easy read this time, Tufty! Footnote 7 is interesting. I wonder if the author himself had read it when writing the original Poisonous Divisionals® paper. There is only a brief discussion of G2/98 and no mention of Memo C in the article in CIPA Feb 2012. Also, who is this mysterious Peercat? Are they a Kat?!

      I think in reality the claim broadening situation has been dealt with, and has never really been a problem - it's only recently that people have actually started thinking about this and discovered that it had already been covered off in Munich in 1973.

      That is no reason I don't think, however, to be complacent about the threat of true poisonous divisionals: I suspect Art 88 and Memo C do not always come to the rescue.

      One thing that I don't think has been resolved is a situation similar to that described by this Anonymous on your now (worryingly old) original post.

      This could be a familiar story for some. File priority application P disclosing A. File PCT application disclosing A (and claiming priority from P) and some new kind of unrelated matter B.

      Enter European phase with an EP1 claiming A. File a divisonal EP2 claiming B. B has a later priority date than A, so under Art 54(3), the disclosure of B in EP1 (which remember has an earlier date of filing than EP2 for the purposes of Art 54(3) due to its valid claim to the priority date of P) anticipates the claim to B in EP2.

      So, my question is, can the contents of a patent application have different dates of effect for Art 54(3)?

      Delete
    28. Joe

      Subject matter B is not entitled to claim priority because it is not disclosed in the priority document.

      It therefore cannot anticipate the claim to B in EP2.

      Delete
    29. @Anonymous:
      Thus, you determine the embodiments “A”, “B”, “C”, “D” that are encompassed by the generic term “Z”, and then determine whether they are disclosed in the priority document. (...)

      Sure, but if you have one of the "limited number of clearly defined alternative subject-matters" that is not new, then the whole claim is not new (and you'll have to amend or cancel it).

      Unless you split "metal" into "copper" and "all metals but copper", you'll keep having "metal" as one of the "clearly defined alternative subject-matters".

      Regarding T 1127/00:
      The priority document directly and unambiguously discloses a specific embodiment “A” that falls within the generic term “Z” defined within the claim. To me, this embodiment should be entitled to claim priority.

      And of course a claim directed to specific embodiment "A" will be entitled to priority. I see no problem here. Once you claim the same invention as in the priority document, you are entitled to priority and the effective date becomes the priority date.

      Btw, I don't see any problem in an OR-claim that defines as one part the specific embodiment and as other part all the compounds defined by the formula minus that specific embodiment, as long as the application actually discloses that other part, i.e. as long as you could draft a claim to that other part without infringing Art. 123(2). And provided that in principle one can form the OR-claim (while complying with Art. 123(2)) and such an OR-claim would survive all prior art attacks, I see no reason to not allow the claim to the generic formula (if only for reasons of clarity, plus G 2/98 seems to approve it).

      I should also add that I think the approach of T 1222/11 and “Tufty’s law” differ. As I mentioned above, Tufty’s law requires the claim to be split up into parts that effectively equal the whole.
      But you are also splitting it up. You are talking about "the remaining part". If by "remaining part" you mean the generic formula, then that covers "A". You seem to want to excluse "A" from the generic formula, which means you are notionally splitting the generic formula into "A" and the complement of "A".

      Under Tufty’s law you split out into copper and not copper, but what do you do about metal?
      Huh? Tufty says: (notionally) split into "copper" and it's complement "all metals but copper".

      Think about the example if it were two separate claims. Claim 1 directed to copper and Claim 2 directed to metal. Would you consider “Iron” to prejudice the novelty of Claim 2?
      Maybe the problem is not that you don't how priority works, but that you don't know how novelty works.

      Yes, iron destroys the novelty of metal.

      Delete
    30. @Tufty:
      Once any intervening prior art comes into play the general rule cannot be applied as it depends much more on what particular part of the claim is entitled to priority.

      How should I understand this? Does the priority situation of a claim depend on whether there exists an intermediate document with particular content?

      Delete
    31. All I mean is that the general rule about imaginary claim splitting can only be applied to the specific situation of the poisonous priority/divisional problem. If there is any intervening prior art, the question is no longer just about A54(3). The priority situation of the later claim is the same, but the question of its validity is then complicated by the prior art. In the metal/copper example, if there is an intervening publication of iron then the later broader claim is no longer valid. The question then becomes whether amendment can be made to the claim to overcome the prior art. Imaginary claim splitting is then no longer relevant and A123(2) comes into play.

      Delete
    32. @ Anonymous 31 July 2013 17:42

      Thanks for your comment, I was sure this is the case but I can't seem to square that with what Art 54(3) actually says.

      The Guidelines GIV 5.1.1 say "the priority date replaces the filing date (Art 89) for that subject-matter in the application which corresponds to the priority application" - fair enough, I just can't find any "legal basis" for it.

      Delete
    33. @Myshkin:

      Given that you have commented extensively on this issue here and elsewhere, would you like to put your thoughts together in a guest post for this blog? It is becoming quite hard now to wade through all the arguments and counter-arguments, and it would be useful if at least the counter-arguments could be summarised. It looks like you would be the best person for it.

      Delete
    34. Myshkin

      "Maybe the problem is not that you don't how priority works, but that you don't know how novelty works."

      I think the confusion lies in the fact pattern. I will repeat for clarity:

      (1) P1 discloses copper
      (2) P2 discloses metal
      (3) Application claims metals.

      D1, which published between P2 and the application (i.e. after P2), discloses iron.

      In the more specific example, Claim 2 is directed to metal. Do you still think D1 destroys the novelty of Claim 2?

      As for your comments, I think we have a fundamental disagreement. I don't see any need for the application to actually disclose the complement that is not claiming priority, i.e. any need to be able to draft a claim to that other part without infringing Art. 123(2).

      To me, any perceived disadvantages of not being able to comply with your requirement are outweighed by the advantages offered by the solution presented in T 1222/11, particular with respect to poisonous divisional applications.

      In fact, I would be interested to hear why you believe that one should have to comply with your requirement. And I don't mean a legal explanation (I think I understand that part), rather I mean why you think it would be morally right/wrong, or inequitable for third parties if the applicant could claim priority in a situation where the application does not actually disclose the complement that is not claiming priority an equity basis.

      The reason I ask for your answer in these terms is because patent law has a historical trend of breaking legal precedents in order to ensure fairness(think about Swiss-type claims or even the recent CJEU decision on SPCs).


      Delete
    35. @Anonymous:
      I think the confusion lies in the fact pattern.
      Yes it did. Sorry, I did not realise iron was in a document published after P2.

      (1) P1 discloses copper
      (2) P2 discloses metal
      (3) Application claims metals.

      D1, which published between P2 and the application (i.e. after P2), discloses iron.

      In the more specific example, Claim 2 is directed to metal. Do you still think D1 destroys the novelty of Claim 2?


      If we follow the "classical approach", then claim 2 has priority from P2, so the effective date of claim 2 is the date of P2, so D1 is not prior art (and it does not matter what it discloses). Everything as expected and as it should be.

      If we follow the approach of T 1222/11, then my very first thought would still be that claim 2 has priority from P2. However, you are probably thinking of the condition that you can only claim priority from the "first application" disclosing the invention? This is a good point.

      According to T 1222/11, P2 is not the "first application" for metal, because P1 discloses "copper" which is encompassed by "metal". The application cannot claim priority for metal "in so far as" it claims copper.

      The implication is interesting. T 1222/11 suggests that we can assign different priorities to different parts of "metal", but let's first see for what subject-matter we could ever claim priority from P2.

      According to G 2/98:
      "The requirement for claiming priority of "the same invention", referred to in Article 87(1) EPC, means that priority of a previous application in respect of a claim in a European patent application in accordance with Article 88 EPC is to be acknowledged only if the skilled person can derive the subject-matter of the claim directly and unambiguously, using common general knowledge, from the previous application as a whole."

      So subject-matter in a claim can only be entitled to priority from P2 if that subject-matter can directly and unambiguously be derived from P2. In other words, subject-matter in a claim is only entitled to priority from P2 if, based on the whole content of P2, one could draft a claim to precisely that subject-matter without infringing Art. 123(2). (Note that I write "subject-matter in a claim" on purpose, because T 1222/11 suggests that one can assign priorities to "intellectually envisage" subject-matters within the scope of a claim. Still for any such intellectually envisaged subject-matter you can only claim priority from P2 if G 2/98 allows you to.)

      If P2 just discloses "metal", what claims could we draft without violating Art. 123(2)? The only claim that I can think of is "metal".

      Or suppose that P2 discloses "copper" and "metal". The only claims I can think of that do not violate Art. 123(2) are "copper", "metal" (and "copper or "metal"). It follows from G 2/10 that a claim to "all metals but copper" would in general violate Art. 123(2) (it is different if P2 discloses "copper or any other metal", i.e. "copper" OR "all metals but copper").

      We already saw that (following T 1222/11) we cannot claim priority from P2 for exactly "metal". We also cannot claim priority from P2 for exactly "copper", because P1 is the first application for "copper". So there is no subject-matter that could ever claim priority from P2.

      So whatever subject-matter in claim 2 you want to single out (using T 1222/11), it will not be entitled to priority from P2. At least part of the subject-matter will have as effective date the filing date of the application, and "iron" will be in such a part. D1 is prior art for that part, and therefore takes away novelty of claim 2.

      Yes, this is unfair, but this is what follows from applying T 1222/11. The decision is flawed and would have serious consequences if confirmed by the EBA. T 1222/11 is a very dangerous antidote to poisonous divisionals.

      Delete
    36. @Tufty:
      That seems a good idea. I don't want to make a hard promise now, but I will try to produce something comprehensible. It will take a bit of time.

      Delete
    37. @Joe (here):

      I have read the article in this months' journal (with some difficulty). The writing style has not improved much, and it is still far too wordy. I suspect the author wants to make the issue look more complicated than it actually is.
      I also suspect that the author had not seen the memorandum referred to in G 2/98 until recently, as there is no mention of it in the various Avidity articles from 2012 and 2011. I am also a bit annoyed that he persists in mentioning "blog commentary" without providing any references. If the article has been prompted by the recent discussions here or on the IPKat (which I suspect it might have been), it would have been nice if these could be referenced.

      Delete
    38. @Joe:
      The Guidelines GIV 5.1.1 say "the priority date replaces the filing date (Art 89) for that subject-matter in the application which corresponds to the priority application" - fair enough, I just can't find any "legal basis" for it.

      See Art. 88(3): "the right of priority shall cover only those elements of the European patent application which are included in the application or applications whose priority is claimed".

      Together with G 2/98 this means that a subject-matter in the application is only entitled to priority from a priority document P if that subject-matter is directly and unambiguously derivable from P (and if P is the "first application" for that subject-matter).

      It would of course be contrary to reason if subject-matter in the application that has little to do with what is in the priority document would still be entitled to the priority date.

      Slightly more tricky is the interaction between Art. 89 and Art. 54(3). It is clear that Art. 89 implies that the date in "which were published on or after that date" is, where applicable, the priority date, because that is the whole purpose of claiming priority. It is less self-evident that Art. 89 also refers to "the dates of filing" of other European patent applications (and their corresponding priority claims). Still it makes sense, because the purpose of Art. 54(3) is to avoid double patenting, so if two European patent applications claim the same subject-matter, only one should be allowed to proceed to grant (for that subject-matter). That goal would not be achieved if for Art. 54(3) prior art we only looked at the filing date and not at the effective date after taking into account priority claims. And the language of the EPC does support this interpretation.

      Delete
    39. Myshkin you write that the purpose of 54(3) is to stop double patenting. I doubt that.

      For me, the purpose of 54(3) is to adjudicate between rival filers who are both novel and inventive over everything that had already been made available to the public at the patentability-testing date of their respective claims. As the world gets better and better connected, the Chance of that Happening just keeps on going up.

      One could decree (as in the USA) that the earlier filing is available for obviousness attacks, or one could invoke Heinz Bardehle's "enhanced novelty" concept, one could have 54(3), or one could have no 54(3) and let them both through to issue. but you don't want any of those solutions except 54(3) do you?

      Now precisely because rival filers seldom file for exactly the same subject matter, it is not ideal to say that 54(3) is to stop "double patenting". There is no explicit prohibition in the EPC on double patenting by the same Applicant. This is another reason why the "poisonous divisional" line of caselaw is so iniquitous. We want 54(3) for adjudicating between rival filers. To try to write out of the ambit of 54(3) the other filings of the same Applicant/Inventor leads to US-style complexity and uncertainty about the legal outcome. So leave 54(3) as it is, short and sweet. That done though, there is no need to penalise filers, to the extent of having their parents kill their children and the children their own parents. Where's the sense, justice, fairness in that?

      Delete
    40. I meant double patenting in the same wider sense as the term is used in G 1/03:
      "These explain that the fictitious extension of the prior art to cover earlier applications in Article 54(3) EPC serves the purpose of avoiding simultaneous protection." (r. 2.1.1)

      "This approach restricts the effects of Article 54(3) EPC to resolving the problem of double patenting." (r. 2.1.3)

      And it's not restricted to rival filers: "since the legislator did not want to make a distinction between the cases of third- party collision and self-collision (van Empel, supra, note 105 f)." (r. 2.1.1)

      Art. 54(3) eliminates double patenting, except for the case where subject-matter is claimed in different applications with the same effective date.

      but you don't want any of those solutions except 54(3) do you?
      I don't know what you mean by this. The solution to conflicting applications provided by Art. 54(3) is law and I have no say about that.

      To try to write out of the ambit of 54(3) the other filings of the same Applicant/Inventor leads to US-style complexity and uncertainty about the legal outcome. So leave 54(3) as it is, short and sweet.

      I can't follow you, because just before this, and after, you complain that poisonous divisionals are unfair.

      Application of Art. 54(3) "as it is" leads to the "poisonous divisional" effect. Getting around it needs a judicial trick as was pulled in G 1/03. One that does not completely leave Art 54(3) "as it is". I am not necessarily against such a trick. But the "solution" that people see in the approach of T 1222/11 leads to far more iniquitous results. In fact, T 1222/11 itself is already unfair, although the outcome might be correct (the ED came to a refusal for lack of inventive step without unfairly denying priority).

      Delete
    41. Myshkin, many thanks indeed for that very clear and helpful reply. Anybody who thinks that the "poisonous divisional" situation is unfair and unreasonable would hope (I dare say) that the EBA can come up with a "trick" that will leave Art 54(3) jurisprudence exactly as it is, in all other respects except the special situation where an Applicant files a divisional and allows it to go to A publication and, in so doing, wipes out any legal protection he might otherwise have been able to secure, if only he had NOT allowed that wretched divisional to be A-published.

      I remain optimistic, that the EBA can execute this trick. I do not want to change Art 54(3) to the way the rest of the world handles earlier-filed later-published applications, because I think that Art 54(3) is superior to their laws, in all respects except for this unreasonable "poisonous divisional" situation. I want there to be no aspect of EPC law which the world outside Europe can legitimately criticise, because I'm proud of the EPC and want to see its reputation rise ever-higher.

      Delete
  2. I will be away on holiday for the next week, so won't be able to comment further or moderate any comments. If anyone else would like to step in to moderate, please let me know by email and I may be able to give you access (provided I can trust you of course).

    ReplyDelete
    Replies
    1. By the way, please email me at tuftythecat@gmail.com, not my work address.

      Delete
  3. Oh, my dear Tufty, you have become intoxicated with this toxic priority stuff. It has become obsessive. What a good idea to take some time off. Have a nice holiday.

    ReplyDelete
    Replies
    1. Thank you, George. I shall try not to think about this nonsense for the next week.

      Delete
  4. Myshkin, T1222/11 could perhaps have given a better explanation, but point 10 of the Reasons does say:

    ....the Appellants cannot benefit from the priority claimed for the combination of features defined in positive terms in amended claim 1, in so far as it covers subject-matter disclosed in D4b, in particular in so far as it covers the six compositions disclosed in D4b which the Appellants are seeking to disclaim.

    So it doesn't say that the priority is lost for the whole generic scope of claim 1, just in so far as claim 1 covers D4b.

    It was the subject matter that they were trying to disclaim which lost priority. For that subject matter, D4b was relevant under Article 54(2) rather than 54(3). So the disclaimer could not be made.

    Without the benefit of the disclaimer, D4b was then novelty destroying for the undisclaimed subject matter.

    The idea that the priority is only partially lost is consistent with the subsequent discussion of the FICPI Memo C.

    It is also consistent with G03/93: an intervening publication constitutes prior art which is citable under Art 54(2) to the extent that the priority is not validly claimed. Though G03/93 considered a different reason why the priority was not validly claimed.

    I think the above explanation also holds for the 2007 EQE exam paper C. The priority is only lost partially, to the extent that the claim overlaps the earlier application filed outside the priority year. But that partial loss is coextensive with the disclosure of the earlier application, which is then a novelty citation.

    ReplyDelete
  5. Just in case the question I drafted actually does get referred (!!!) I'll post this comment with my full name visible!

    The details of the appeal can be found here http://casetracker.justice.gov.uk/listing_calendar/getDetail.do?case_id=20131655

    ReplyDelete