Tuesday 15 October 2013

Partial Priority - An opposing view

Regular readers will know that I have been concerned with the issue of partial priority under Article 88 EPC, and how far it can be applied, because this has serious implications regarding the extent to which a European patent or application can be held to lack novelty over its own priority application. I have consistently expressed the view that it should not be possible for a European patent to be held to lack novelty over its own priority application if the effect of Article 88 EPC in combination with Article 54(3) EPC is properly applied. My view, which has support from the EPC and the case law of the Boards of Appeal, is detailed in previous posts (the main one being here). There is, however, an opposing view that can also be said to have support from the EPC and case law, which has been expressed and discussed in comments on this blog and elsewhere. It is not yet clear which view will eventually win out, and I think it will need a referral to the Enlarged Board to settle it.

This opposing view has been detailed by one of the opponents in the Nestec case T 1674/12, which unfortunately will not be the case that settles the issue because a referral to the Enlarged Board has not been made. As noted last week, the Board decided in the end to revoke the patent, and will be issuing their written decision later. Whether their reasons will take into account the partial priority issue is not yet known. I suspect they will simply bypass the issue because the patent was invalid for other reasons anyway [Update 19/10/13: It appears my suspicions were correct; the Board did not even get to the issue of partial priority at the hearing, so it will definitely not feature in the reasons for the decision]. For the time being, it is interesting (to me at least) to see how the other side's view on the partial priority issue has been expressed.  The following is a rough translation of the relevant portion from the opponent's letter from 9 September 2013, which I have annotated with hyperlinks where appropriate. The original is available on the register here. My apologies for any errors or omissions resulting from my less than ideal language skills. If you can come up with a better translation, I would be happy to replace it.
2 . On the issue of partial priority

The patentee seeks, in its letter of 15 July 2013, to criticize the British decision of Nestec v Dualit especially regarding the issue of the invalidity of the priority of claim 1 (and by default novelty vis-à-vis document D1).

2.1. Position of the proprietor

The incumbent continues to assert that claim 1 has a partial priority that immunizes against the lack of novelty vis-à -vis document D1 . In support of this assertion , the owner cited and commented on a recent decision T 1222/11, which also refers to a part of the preparatory work of the EPC 1973, namely a memorandum of FICPI, mentioned in the reasons of the Enlarged Board of Appeal G 2/98.

G 2/98 accepted the principle of allocating separate priority dates for different subject matter covered by the same generic type of claim, provided that it leads to "a claim having a clearly defined limited number of alternatives".

According to the proprietor, the decision T 1222/11 as well as the memorandum of FICPI lends credence to the theory that the above condition posed by the Enlarged Board of Appeal is satisfied if it is possible to "identify conceptually" alternative objects by comparing the claim and the contents of the priority documents.

The approach proposed by the holder should not be followed for the following reasons.

2.2. Neither decision T 1222/11, nor the contents of the memorandum of FICPI, is authoritative

First, the decision T 1222/11 (by Board 3.3.7) is an isolated decision. In point 11.5 of the reasons, it is explicitly stated that the Board departs from the approach that has been followed in previous decisions T 1877/08 (by Board 3.3.10 ), T 476/09 (Board 3.3.9 ), T 1443/05 (Board 3.3.1 ) and T 1127/00 (Board 3.3.8) . We can add to this list the decision T 70/05 (Board 3.3.8) , which confirms the reasoning in T 1127/00.

Thus the decision T 1222/11 is opposed to the dominant jurisprudence of the Boards of Appeal. If only for that reason, it should not be followed by the present Board.

With regard to this Memorandum of FICPI, it is true that the opinion G 2/98 states in 6.4 that we can consider that it "express[es] the legislative intent underlying Article 88(2), second sentence, EPC". The purpose of the memorandum was indeed to suggest that the legislature can allocate multiple priorities in a claim. The proposal was accepted , and it is in this (narrow) sense that the memorandum reflects the ratio legis of Article 88.

Specifically, the Enlarged Board takes up the distinction made in the memorandum between the claims of "AND" type and claims of "OR" type. Only the second type can lead to the allocation of multiple priorities (paragraphs 6.5 to 6.7 of the reasons).

However, the Enlarged Board of Appeal has not expressed the opinion that the detailed content of the memorandum should be taken as is, without changing anything or adding an assessment of whether, for a given generic claim, partial priority can actually be recognized for a portion of the claim.

The Enlarged Board has actually implicitly taken a contrary position when said in the passage quoted in the opinion, at the end of Section 6.7 of the reasons that "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". Indeed, this condition of a "limited number of clearly defined alternative subject matters" is not listed in the Memorandum of FICPI. It is therefore a condition that was added to the memorandum by the Enlarged Board of Appeal , when analyzing the situation of type "OR" claims.

Consequently, it is not correct to take the examples cited in the memorandum of FICPI and indicate that the Enlarged Board approved the allocation of multiple priorities for these examples. The Enlarged Board has only taken over the general framework of the memorandum of FICPI (distinction between claims of type "OR" and "AND"), adding the condition of a "limited number of clearly defined alternative subject matters" regarding claims of "OR" type, without any comment on the illustrations of "OR" type claims presented in the memorandum.

Only the decision T 1222/11 has taken over the sample claims of "OR" type mentioned in the memorandum, but in doing so it follows an approach in contradiction to Boards 3.3.1 , 3.3.8 , 3.3.9 and 3.3.10 .

In addition, on the merits, the approach of the proprietor, which is based on the decision T 1222/11 and the Memorandum of FICPI taken in its entirety, is not consistent with the view G 2/98 as is shown below.

2.3 . The faulty reasoning of the decision T 1222/11 and part of the memorandum FICPI

In section 11.4 of the reasons for the decision T 1222/11, Board 3.3.7 is critical of the jurisprudence of Boards 3.3.1 , 3.3.8 , 3.3.9 and 3.3.10 of the fact that this jurisprudence has held that the requirement for the Enlarged Board on the "limited number of clearly defined alternative subject matters" characterizes the manner in which the subject of an "OR" claim must be defined. In point 11.5.2 of the reasons, Board 3.3.7 offers its own interpretation of the condition laid down by the Enlarged Board, that this condition simply refers to the ability to identify conceptually, by a comparison of the claimed subject matter with priority documents, a limited number of clearly defined alternative subject. This theory is also extended to the case of partial priority (paragraph 11.6 of the reasons).

But this interpretation cannot be correct because it leads to rendering meaningless the condition "limited number of clearly defined alternative subject matters" posed by the Enlarged Board. Indeed, it is always possible to divide conceptually (that is to say abstractly and thus artificially) a generic object into two objects of type A and non-A. Also, one wonders in what situations allocating multiple priorities (or partial priority) should not be accepted, following this approach. In other words, to follow the proposed T 1222/11 interpretation, the condition imposed by the Enlarged Board had simply no reason to be: it is not a requirement.

In addition, the Board 3.3.7 reviews examples in the memorandum of FICPI, and sees a confirmation of its theory (Section 11.5.8 of the reasons). But in fact, at least some of these examples are problematic.

The first example cited (listed in Section 11.5.5 of the reasons) seems consistent with G 2/98. In this example, a first priority document discloses subject matter with the element chlorine, a second priority document discloses subject matter with the elements bromine, iodine and fluorine, and a generic claim in the subsequent request is subject matter with a halogen. In this case, the generic claim has the first priority in terms of chlorine, and the second priority that is bromine, iodine and fluorine.

This first example appears to conform to the G 2/98, since the generic term "halogen" corresponds to a limited number of clearly defined alternative subject , namely the four elements chlorine, bromine, iodine and fluorine.

The situation is different when a generic claim covers a continuum of embodiments . This is particularly the case in the third example of the memorandum of FICPI in point 11.5.7 of the reasons of T 1222/11.

In this example, a first priority document discloses a method of coating the interior of a pipe, and a second priority document discloses a method of coating the walls of bottles or any other hollow body. The memorandum suggests that the provision of the EPC proposed on multiple priorities will permit a generic claim in a subsequent application of a method of coating the inner walls of a hollow body. The memorandum justifies this approach by the desire to avoid having to unnecessarily increase the number of claims of patent applications.

The memorandum states in this regard: "if multiple priorities are not allowed for a single claim, the applicant must write parallel claims, one on the method of coating inner walls of the pipe, which will benefit from the first priority, and the other on the method of coating inner walls of hollow body other than the pipes, which will benefit from the second priority".

But this position is completely incompatible with the position of G 2/98, that for the condition of the "same invention" to be met for a claim to be eligible for the priority of an earlier application means that: "the person skilled in the art [must ], using common general knowledge, directly and unambiguously deduce the purpose of the claim of the previous application as a whole".

According to the reasoning of the memorandum of FICPI, a putative claim "on the method of coating inner walls of bodies other than hollow pipes" is supposed to benefit from priority in the second priority document. The second document priority only describes a "method of coating walls of bottles or any other hollow body" and therefore does not describe a method of "coating internal walls of hollow bodies other than pipes". According to G 2/98 - and the application has been made consistently by the Boards of Appeal - this claim devised by the FICPI memorandum therefore cannot benefit from the second priority. It is necessary that the second priority document itself describes a "method of coating inner walls of bodies other than hollow pipes" for priority to be validly claimed.

Correspondingly, the generic claim proposed in the third example of the memorandum of FICPI cannot benefit from the multiple allocation of the first priority and the second priority. Not that the claim was a generic form - since this does not raise any difficulty in the case of first example of halogens, but the fact that the concept of cladding inner walls of hollow bodies covers a multitude of possibilities (while G 2/98 requires a limited number of subject matters), which are not clearly defined (although the G 2/98 requires that the objects are clearly defined). Thus, this claim cannot really benefit from the second priority, which is the only one to disclose the generic concept of cladding hollow bodies, a concept that is not divisible by a limited alternative objects clearly defined number.

In fact, the Board 3.3.7 seems to have identified the contradiction between the third example of the memorandum of FICPI and the position of G 2/98. But for the Board, the position of G 2/98 does not apply just to the situation of multiple or partial priorities .

It is true that the question that the Enlarged Board of Appeal responded to did not specifically concern the situation of multiple or partial priorities . Nevertheless, the principle laid down by the Enlarged Board has a very broad application and also covers claims of type "OR" with multiple or partial priorities. Thus, the allocation of a certain effective date of only part of the subject matter of a claim assumes that the subject matter can be directly and unambiguously derived by the skilled person from the relevant reference document .

To decide otherwise would be to establish an exception to the test for disclosure that was developed by the Enlarged Board of Appeal over the years including G 3/89, G 11/91, G 2/98, G 1/03 and G 2/10 - while nothing justifies such an exception.

This approach, termed the "gold standard" in G 2/10, is indeed precisely considering the coherent separate questions of the validity of the right of priority (Art. 87 EPC ), the extension of subject matter (Art. 123(2) and 76(1) EPC) and novelty (Art. 54 EPC).

And, in this consistent approach, the central criterion is direct and unambiguous disclosure for the skilled person. This is why the reference to the possibility of identifying conceptually alternative subject matters proposed by T 1222/11 and endorsed by the patentee has no place in this coherent approach. It is, in effect trying to implement a so-called rule of logic to conclude in a systematic way the possibility of benefiting from the allocation of multiple or partial priorities.

However, using a rule of this type of logic has been firmly rejected by the Board of Appeal, in a related context, that of the evaluation of disclaimers. The Enlarged Board [in G 2/10] concluded that it was not possible to make a case by case evaluation of the direct and unambiguous disclosure of relevant subject matters.

The memorandum of FICPI was written in 1973 , well before the establishment of the "gold standard" mentioned above. It is not surprising that it is not fully compliant with this standard. It is for this reason that the Enlarged Board of Appeal made clear in the G 2/98, the condition of a "limited number of clearly defined alternative subject matters", the only way to ensure consistency of the whole practice of the EPO .

2.4. Conclusion concerning claim 1

In view of the foregoing, the reasoning of the patent holder cannot be followed. Claim 1 of the main request does not cover a "limited number of clearly defined alternative subject matters". In fact the claim covers an infinity of subject matters that are not identifiable to each other - as has already been explained above. This claim cannot therefore benefit from the claimed priority, even partially. So there is a lack of novelty vis-à -vis D1.
This opposing view looks at least superficially persuasive, citing much relevant case law in support. The main flaw in the argument, however, is the assertion that the Enlarged Board in G 2/98 did not consider that all of the examples in the FICPI memorandum could justify a claim having a valid claim to partial priority under Article 88 EPC. If this was the case, it is very strange that the Board made no mention of this in their reasons. The Board clearly approved the memorandum and its justification for Article 88 EPC, so the argument that the criterion of "a limited number of clearly defined alternative subject matters" somehow makes some of the examples in the memorandum invalid is to my mind implausible. If instead we take all of the examples in the memorandum, which the Board in G 3/98 certainly at least considered, to be valid then the opponent's argument must fall away and a claim to partial priority of the kind in the Nestec case must be valid.

14 comments:

  1. If, unlike this opposing view, we accept that the "A or not A" formulation defines a limited number of alternative subject-matters, then can anyone put forward an example which does not define a limited number of alternatives? I can't think of any, and so begs the question, what does the G2/98 qualification actually mean?

    I think so long as there is clear support in the description for the conceptual split subject-matters then it should be okay.

    ReplyDelete
  2. The opponent spotted the same contradiction between the third example of the Memorandum and G 2/98 as I have spotted. The example is based on the premise that a claim "method of coating inner walls of hollow bodies other than pipes" can be based on the disclosure "method of coating walls of bottles or any other hollow body". This is in flat contradiction with the conclusion of G 2/98.

    The drafters seem to have assumed that adding a disclaimer to a claim was always allowable. At the time the Memorandum was drafted this may not have been an unreasonable assumption. But it has turned out to be an incorrect assumption.

    Ignoring such incorrect assumptions made by the drafters, what the Memorandum really proposes is that if it is possible to draft parallel alternative claims with various priorities, then the applicant should be allowed to OR them together into a single claim, possibly replacing the alternatives with a generic term covering exactly those alternatives, while letting each of the alternatives keep its priority. The if is essential.

    The opponent points out something that I had missed. In point 11.7 of the reasons of T 1222/11, the Board indeed appears to admit the contradiction between the conclusion of G 2/98 and the examples in the Memorandum. The Board brushes this contradiction away by arguing that the conclusion of G 2/98 is not valid for OR-claims. This is a remarkable argument, to say the least. According to the Board this follows "implicitly" from point 6.7 of G 2/98. The idea seems to be that if all the examples in the Memorandum fully comply with the law, then G 2/98 cannot apply to these examples because otherwise it would contradict with the examples! This is circular reasoning, and it does not explain why the conclusion does not make mention of this exception.

    ReplyDelete
    Replies
    1. I think your point here is the key issue on which we disagree. Your argument, as I understand it, is that for a valid claim to multiple priorities it must be possible to divide the claim up into individual claims without adding matter, i.e. there must be basis in the application for an actual division of claims. My argument is that this should not be necessary, because the scope of any actual divided claims when put together would be the same as the undivided claim. I think your argument comes from the overly formulaic approach often taken at the EPO regarding basis for amendments. My argument is, I think, more pragmatic as it avoids the need for the drafter (who is not necessarily familiar with European practice) to make sure there is explicit basis for the nominal "non-A" part as well as the "A" part that has a priority claim.

      Delete
    2. Your argument, as I understand it, is that for a valid claim to multiple priorities it must be possible to divide the claim up into individual claims without adding matter, i.e. there must be basis in the application for an actual division of claims.

      Indeed, and this is exactly what the drafters of the Memorandum assume to be the case in the examples that they have given. The Memorandum's argument for multiple priorities is that it leads to more concise drafting (one OR-claim instead of multiple parallel claims), not that it allows to obtain protection for subject-matter that could otherwise not be protected.

      Delete
    3. I can see why you might think that. However, the inevitable consequence of the ability to have multiple priorities in a single claim, at least for the range and pipe/hollow body examples in the memorandum, is that an application could have a more broadly defined claim with no literal basis for dividing it out into multiple claims that together have the same scope. Although the principle aim of the memorandum appears to be, as you state, to avoid the need for multiple claims, the consequence is that a claim without clear basis for being divided up would still be valid. This must therefore mean that there is no need to have basis for multiple claims for Article 88(2) to apply.

      Delete
    4. If it's true that "the concept of disclosure must be the same for the purposes of Articles 54, 87 and 123 EPC" - why do anti-T1222/11 commentators only compare A87 and A123?

      A disclosure can fall within the scope of the claim and destroy the novelty of that particular subject matter under A54, even if there is no basis within the application.

      The same could be true for A87. A disclosure in the priority document can fall within the scope of the claim and give rise to a priority entitlement for that particular subject matter under A87, even if there is no basis within the application.

      Delete
    5. The concept of disclosure is the same, but the fact that a specific disclosure takes away the novelty of a generic disclosure within the meaning of Art. 54 does not mean that a specific disclosure provides a sufficient basis for a generic disclosure (or the other way around) within the meaning of Art. 123(2) or Art. 87(1).

      Art. 123(2) and Art. 87(1) require that the subject-matters being compared are the "same". Art. 54 does not require the subject-matters being compared to be the "same", but only that a subject-matter disclosed in a prior art document is encompassed by the subject-matter of a claim.

      I'm sorry to say this, but this is rather elementary.

      Delete
    6. The Enlarged Board in G 2/98 said "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."

      It doesn't seem unreasonable to me to understand "the subject matter of the claim" to mean any subject matter that is encompassed by the claim. Thus, provided that an embodiment (encompassed by an OR-generic claim) is directly and unambiguously disclosed within the priority application, then that embodiment of the claim is entitled to priority.

      As T 1222/11 states, this involves a comparison of the subject-matter of the claim with the disclosure of the multiple priority documents, so as to determine which parts of the claim are covered by the rights of priority claimed.

      Delete
  3. I don't see why G 2/98's approval of the Memorandum's expression of the legislative intent underlying Article 88(2) EPC would make it improbable that G 2/98 did not also approve of the specifics of the examples.

    What if the EBA did not agree with the details of the examples. Should it have said so in G 2/98? I do not see a reason for that.

    The summary of the Memorandum given in G 2/98 abstracts away from these details and focusses on what, in the EBA's view, was really the point of the Memorandum: that if respective claims to A and B have different priorities, then a claim to A or B has multiple priorities. According to the EBA, this is how Art. 88(2) has to be understood.

    ReplyDelete
  4. Let me add that I am not necessarily against a "solution" to the perceived problem of poisonous divisionals and priorities. But I am convinced that the reasoning in T 1222/11 is incorrect and should not be followed. I think solutions can be found that fit much better with the case law of the Boards.

    My favourite solution is a simple exception to the letter of Art. 54(3): subject-matter X having effective date D in patent application A is not Art. 54(3) prior art for subject-matter X having effective date D in patent application B of the same applicant. (It might be better to drop "of the same applicant".) The EBA could easily formulate such an exception much like it formulated an exception to Art. 123(2) in G 1/03.

    I don't think there is any support for this interpretation of Art. 54(3) in the case law, but as far as I can see this interpretation would have no impact on cases not plagued by the poisonous divisional problem. This is very different for the approach of T 1222/11.

    ReplyDelete
    Replies
    1. "subject-matter X having effective date D in patent application A is not Art. 54(3) prior art for subject-matter X having effective date D in patent application B of the same applicant."

      This doesn't work.

      The whole problem with a poisonous divisional applications is that it (or the subject matter in it) has an earlier effective filing date than the subject matter of the claim in the parent application.

      Moreover, it is well-established that subject matter cannot be used as prior art if it has the same effective filing date so I'm not sure how this would change anything.

      Delete
    2. Sorry, my formulation is indeed incorrect. This is what I should have written: "subject-matter X having effective date D in patent application A is not Art. 54(3) prior art for patent application B (of the same applicant?) disclosing the same subject-matter X with the same effective date D".

      So the occurrence of the novelty destroying Art. 54(3) prior art in the application itself should neutralise that prior art.

      Without this rule, there is actually no good reason why an application could not be novelty destroying Art. 54(3) prior art for itself.

      So the EBA could reason as follows. The drafters of the EPC evidently could not have wanted Art. 54(3) to be interpreted as allowing self-poisoning. A priority-entitled specific embodiment in an application B therefore does not count as Art. 54(3) prior art for B (i.e. not even for claims of B that are not entitled to priority). Now we only have to accept that the same applies for the same subject-matter with the same effective date occurring in other EP applications and we have the rule.

      Delete
  5. A short comment on this discussion. Since I am representing the opponent (thanks for the translation of our brief), these comments are only my personal views.
    I am not convinced by the argument that the drafter of the original priority application should be considered - at all. This seems to be a matter of knowing whether or not the same invention had been made at the time the original priority application was filed. The whole purpose is security of the public, and the "need of the drafter" should in my opinion be a very secondary consideration.
    In addition, the same argument could also be used against G 2/98 as a whole. This decision drastically changed drafting practise - we all used to amend applications at the end of the priority year, which might not be such a good idea nowadays.
    I do not believe that construction of the law should rely on the user-friendliness for patent drafters. Encouraging innovation may still require a degree of legal certainty.

    ReplyDelete
    Replies
    1. Many thanks for your comment Grégoire. As you already know, I disagree with you on the partial priority issue and have made my own arguments quite clear (I hope) in previous posts. My argument does not, however, depend on the general idea of considering the drafter, which as you say is a secondary issue. Instead, my point is that if the EPC is properly applied then the poisonous priority problem goes away.

      Amending applications at the end of the priority year is always a risky business, and should obviously be done with care, not least because any broader definition of an invention should be fully supported by the application. This is an Article 83&84 issue though, and not an Article 88 issue. Defining an invention more broadly should not in itself result in the absurd outcome where the priority document becomes novelty defeating for the application.

      Delete