On poisonous priority: taking the debate further

IPKat team blogger Darren's post on Nestec v Dualit ("Poisonous priority – how many ways can a patent be toxic?", here) attracted a large number of comments from European patent experts and enthusiasts, including several from Dutch and European patent attorney Walter Hart (EP&C). It was therefore a welcome pleasure to this Kat that Walter has chosen to enlarge on his comments via a separate, stand-alone guest post which he has written together with his colleague Reinier Wijnstra.
"In a recent decision in Great Britain in Nestle v Dualit, a European patent was held invalid over its own priority document. In our opinion, this decision goes against basic principles of the patent system.

The patent system is created to stimulate innovation. It does so on a quid pro quo principle. The inventor gives information on an invention to society, and society gives back an exclusive right for a limited period, a patent. If the invention is big, the patent is big. If the invention is small, the patent is small.

The essence of innovation is improvements. Improvements do not stop after a priority application is filed. In the priority year, another embodiment based on the exact same innovative concept may be found. The field of the invention may be broadened. Further, an inventor may find that a feature he initially considered essential and explained as such to his patent attorney, is non-essential after all. The inventor may develop a better understanding of the working of the invention without actually changing the embodiment. It is also possible that a different light is shed on the invention by the prior art mentioned in the search report of the priority application, resulting in better understanding of the invention.

Further, many innovators in particular SMEs, invent something, file an application and build a prototype. During the priority year the prototype may be further improved. These things and others may happen in the priority year.

When drafting the priority application, it is the job of the patent attorney to get the broadest protection and draft claim 1 accordingly. But as Niels Bohr already explained, prediction is very difficult, especially about the future. A patent attorney cannot predict future improvements of the invention. As a result of the improvements discussed above, the view on the invention may change somewhat. As a result, in order to get the full scope of protection, it can be necessary to amend claim 1. The required amendment is not always straightforward. Innovation is not always simple, and neither is the resulting required amendment of claim 1.

However, according to thie ruling in Nestec v Dualit, such an improvement and the corresponding amendment of claim 1 is very dangerous to the inventor, because it may result in a complete loss of the subsequent application which is filed at the end of the priority year.

It was already known that this could happen. That is why many patent attorneys often or always withdraw priority applications before publication, or file national priority applications instead of European priority applications. In other words, users withdraw rights or file lesser rights because of improvements or the mere possibility of improvements in the priority year. Our patent system punishes these improvements. You either withdraw the first right, file it in a lesser way, or loose the subsequent right.

This ruling is based on a very strict interpretation of the term ‘same invention’ and the corresponding EPO case law, in particular G2/98 and a strict application of Art 54(3) EPC. However, the only thing the inventor did was to improve the invention in the priority year. The subsequent application presents an improved invention to society. According to the quid pro quo principle via which the patent system stimulates innovation, an improvement should lead to more or perhaps better defined rights, not less rights.

Some may say: so what? Broadening of claim 1 is possible, you only need to withdraw the priority application. But improvements in the priority year occur quite often. Each time the patent attorney needs to make an assessment whether to amend claim 1, what the amendment would do with the priority and whether the priority application needs to be withdrawn. All this must be explained to the inventor, who mostly does not understand it. For the inventor, the situation is straightforward. He simply improved his invention. However, the legal consequences are very difficult. Moreover, withdrawing does not solve the problem of the loss of priority. So any disclosure of the invention will still kill the subsequent application.

We should ask the question: what is the damage to society if the priority application is published? The subsequent application is filed before publication, and both are usually published at the same time, so the public can see the correspondence and the differences between both. We cannot see any damage. Why punish this publication with capital punishment?

Further, foreigners who file European priority applications for the quality search of the EPO may also loose rights if not properly advised. In an ever converging world, what is the use of a “European Priority Punishment” ? Will non-Europeans understand it? It is a patent attorney’s paradise, but was that the intention of priority?

Some have argued that for any further improvement in the priority year, another independent patent family should be created. That solution may also be great for patent attorneys, but doubles the costs for SMEs. The patent system is not here only for Apple, Nestle and the likes, but also for a local inventor who invents a new mousetrap.

Moreover, it is not always possible to be novel relative to the priority application without losing scope of protection or without taking at least some risk that the priority application is considered novelty destroying for the subsequent application after all. This occurs in particular if the inventor’s understanding of his invention has progressed, i.e. if the improvement, new insight or redefinition of the invention is based on the same figures. Things can become very messy if you try to formulate a claim 1 that is novel over the figures of the priority application, but needs to be supported by the very same figures.

Further, many European countries have registration systems which result in a quick patent. Inventors use these national routes to get quick protection on home turf. A subsequent European patent is often filed which claims priority, and upon grant is validated in the same country. Some of these European patents will have priority problems. According to this decision, these patents will be invalid on home turf. Worse, European countries have provisions that prevent double patenting. As a result, the priority patent becomes unenforceable once the European patent is validated. The priority patent destroys the European patent and the European patent destroys the priority patent. Mutual assured destruction.

How would this situation play out in the future unitary system? It appears the unitary patent would be invalid as a whole over a published national priority application.

Was all this the intention when the priority system was created in 1883? We believe the intention was to create a priority system that helps the inventor, not hurts him. The subject of priority is complicated and we do not have all the answers. But this ruling is unjust. We need to rethink priority, and in particular the meaning of the words ‘same invention’. We need a priority system that does not hurt an inventor who improves his invention in the priority year, because our patent system is intended to stimulate these improvements".
Do you agree with Walter and Reinier? Do let us know!
On poisonous priority: taking the debate further On poisonous priority: taking the debate further Reviewed by Jeremy on Wednesday, June 12, 2013 Rating: 5


  1. I agree entirely with Walter and Reinier, but personally I don't think arguments based on public policy are enough. The poisonous priority problem is in essence the same as the poisonous divisionals one, which has been discussed at length on my blog, the latest post on which is here. I have yet to see how my solution is wrong.

  2. The very purpose of a priority claim is to plant the applicant's flag in his territory of interest. If you must improve your invention following filing of a priority application and still wish to have the newly minted embodiments (improvements) addded to the specification then you have three options: (i) file a further priority application if you are still well within the priority year followed by a PCT / EP claiming priority from the two applications, (ii) withdraw the first priority before its anniversary of filing, and refile it together with the new embodiments and claim priotity only from the second priority, or (iii) directly file a PCT / EP etc application containing the specs of the priority app(s) + the improved embodiments - under this option you need to accept that the newly added embodiments will not be entitled to the priority date rather they would have the date of filing of the PCT / EP as their date of filing. 3rd parties need certainity as to when new embodiments are added - we need to balance the rights of applicants with that of 3rd parties.

  3. The original point of the priority system was to give you time to file the SAME application in other countries after the first filing in your home country, to avoid having to coordinate filings in many countries on the same day. The priority year was not originally intended to be a time for catching improvements to the invention and adding them in to applications filed at the end of the year. If using it this way causes problems then it is not the fault of the system, as it was never designed for an update-adding purpose. Some seem to consider the priority year as a kind of grace period, but that's not what it is or what it was supposed to be.

  4. It sounds like the last comment came from the EPO, which often goes on about certainty for 3rd parties without really explaining what this means or the unintended consequences of its actions (amended Rule 36 anyone?). In the case of priority-claiming applications, the certainty is there in the form of the wording of the priority application, which forms part of the public record once the application is published. There is no need for the overly restrictive (and in my view wrong) interpretation of A54(3) to try to establish some more certainty for 3rd parties when it is there already, and certainly not at the expense of now making a lot of existing patents and applications a lot less certain.

  5. I agree with Anonymous. The priority year allows you to defer the costs of an international filing portfolio, not further refine your invention. If you don't know what your invention is yet, then you didn't reduce it to practice properly; if you file on a speculative basis then you should take the risk that your priority filing is incomplete or inaccurate, not the public.

    I do not see why the priority year should entitle you to add subject-matter and still claim the original priority date. If your further developments are made before your application publishes, and assuming they are novel, you can file a new application that claims these developments.

    If your priority case was too narrow, sorry, but that means you got it wrong. The deal with the Paris Convention is that you can keep your earlier filing date for what you invented at the priority date, not what you invented afterwards. If you invent something later that lacks novelty over what you filed on earlier, that is tough luck, but is equally likely to happen after the priority year as during ther priority year. That's the first-to-file system for you...

  6. I am not sure why the new debate - when I was training 15 years ago, my partner hammered into me that if the invention was still being developed, perhaps the worst thing to do is file a patent application for the invention in an unfinished state. The potential downside was too great for the possible benefit of being able to beat the filing date of some unknown prior art.

  7. In response to Jeremy (non Kat)'s point, if the original point was to only allow the same application to be filed claiming priority, then can he explain what the purpose was of Article 4F of the Paris Convention?

  8. Tufty, Article 4F is a sensible provision allowing for the possibility of new subject matter being added. It is designed to prevent an application being refused simply because of having new matter, which would be rather draconian. It also allows the new matter to give rise to its own priority date. That doesn't take away from the fact that the main point of the Paris Convention was not about giving an effective grace period for new improvements to an invention to be added. Anyone wanting to rely on Article 4F to add improvements is allowed to do so if they wish, but they should be aware of the risks of this, and should be aware that Article 4F does not provide a grace period.

  9. The anonymous commenter above at 10:58 is totally wrong. New subject matter clearly does not have a valid claim to priority. This is not the argument. The flipside, however, is that any new subject matter is new, i.e. novel, and is therefore not anticipated by publication of the priority application. This is the whole point of my argument (see link to my post above).

  10. Problems are not only caused by claiming too narrowly in your priority application and broadening your invention during the priority year that. There are other scenarios (which arguably can be due to bad patent drafting0 that can lead to the same results:
    imagine that your invention has to do with a process using a metal, and you have the examples on iron (Fe) and gold (Au) in your priority application. Now some prior art has been found that discloses the same process with copper (Cu). Our 'smart' patent attorney limits his invention now to noble metals, thereby rescuing at least part of the invention. However, this would lack priority, which means that your own priority application with Au as example would destroy the novelty of your amended claim.
    Erroneous drafting? Hmmm...there seems to be something that would justify obtaining a patent on the more narrow invention.

  11. Of course a national prior right in just a single participating MS would destroy the unitary patent in all MS (this is a big problem which is not present in the European patent).

    On the other hand, art. 4(2) of the unitary patent regulation does not exclude the double patenting via a national patent and a unitary patent, does it?

  12. There's nothing new here. The only thing that surprises me is that there are people who are surprised to learn that priority applications are just... applications.

    Australia just abolished any distinction in status between provisional and complete applications. And a good thing, too, because all it did was give applicants a false sense of security which was worth nothing when subsequently filing overseas.

    The solution to all of this, in the vast majority of cases, is a limited grace period, protecting the inventor against her own disclosures. If the US can live with 'first to file', surely Europe can get with the program and introduce a 12-month grace period?

  13. What is the alternative? The applicant would then essentially be free to choose any (of his own) applications as priority document, and such a document could never be held against him.

    Often enough, a priority document has very little to do with the claimed invention. Why would such a document, if published, be "protected" by the applicant's claim that it is a priority document, so that it cannot be used under Article 54(3)?

  14. I agree with Tufty here. I wrote an article on this point last year in the March 2012 CIPA Journal.

    In this article the proposed solution involves splitting the claim so that it has two effective priority dates. The first part of the claim is entitled to priority and therefore cannot be anticipated by the earlier application. The second part of the claim is not entitled to priority but it is not anticipated by the later application.

    I have not yet heard a valid criticism of this technique...

  15. non-Kat Jeremy and some Anonymouses don't seem to appreciate that the problem many have with Justice Arnold's interpretation of the EPC and the Paris Convention in Dualit vs. Nestec is that, because a simple difference in claim wording, a first filing can be held not to provide a valid priority claim for a subsequent filing, and yet be completely novelty-destroying for the same subsequent filing. This, in my very humble opinion, is completely aberrant and goes against the whole spirit of the Paris Convention. It's the "intermediate generalisation" nonsense on steroids.

  16. Often enough, a priority document has very little to do with the claimed invention. Why would such a document, if published, be "protected" by the applicant's claim that it is a priority document, so that it cannot be used under Article 54(3)?

    Hmm, if the priority document "has very little to do with the claimed invention", how could you consider it as "novelty-destroying" under Art. 54(3) EPC?!

  17. "Broadening of claim 1 is possible, you only need to withdraw the priority application." I don't see a problem. G2/98 was given in 2001 and rather clear, at least clearly rather strict.

    We should ask the question: what is the damage to society if the priority application is published? - that is about 54(3), why take the whole contents of the published application and not only the granted claims?

  18. The amendments to the EPC made a few years ago mean that an EP application no longer needs to contain any claims, something that has long been UK practice. The situation of such an application, which in principle could simply consist of a detailed description of one or more embodiments, being used as a priority document does not appear to have come up for consideration yet.

  19. I've discussed this at length a long time ago and have nothing new to add. The previous discussion is no longer able to give rise to a valid priority claim so any repetition would would lack novelty.

  20. I was only a humble patent-leather-shoe-shine-clerk when the convention was signed, but the partners (who only beat us when we deserved it and looking back I'm a better attorney for it) in our office (with a fine view of the hanging station outside Newgate) did explain the purpose and intent of the new priority system to the worthless souls of the office. If only I could remember what they said, I'm sure it would be of interest.

  21. Thank you Peter. I'm glad somebody agrees with me, or at least understands what the issue actually is. The rest seem to be missing the point entirely.

  22. Peter - I think whether there's a problem or not depends on whether the new bit, with the old bit taken out, can really be considered an invention in its own right. If, for instance, the original description describes X chloride, and you then discover that any X halide will work, that could be considered another invention (or perhaps five new inventions). But if the theme linking the new stuff is more vague, and the old example is a typical example of it, you may have problems arguising that "new stuff minus old stuff" is a plausible invention in its own right, which makes your "this bit isn't anticipated" argument fall down.

  23. Peter, does law support the proposed "splitting of a claim" into two such that its part 1 is entitled to a valid priority claim and its part 2 does not.

  24. My three pennyworth:

    I'm with those who see Paris as a scheme for protecting the "same" invention. Improvements are not the "same". Paris imposes great demands on drafting skills, sure. What do we want though: something only those who have qualified by EPO examination can do, or something any old non-specialised lawyer can do? Which of these serves the public better?

    To Change Art 54(3) to grace other earlier filings of the Applicant is a cure that is worse than the disease.

    So long as a div counts as a different application from its parent though, the mere act of running a div through to A publication can be a sentence of death to both the parent and the div. That seems wrong.

    I'm a bit sceptical where Tufty's claim splitting notions might creep, over time. What is to stop the EBA interpreting Art 54(3) as including all earlier filings of Applicant except those linked to the case in suit in a patent/div relationship?

  25. For your consideration:

    The prohibition of added matter post-filing allows 3rd parties to make business decisions based on the A publication. The same considerations do not apply to priority claims.

    If a claim lacks novelty over its priority document it should be entitled to priority.


  26. Dear Tufty
    As a fellow feline it is to be expected, and I think you know it already, but I agree with you. And with Peter Arrowsmith. And with Walter & Reiner. I think my views are already known so I won't try to squeeze them into this comment.

  27. Thanks Darren. I'm not surprised that a fellow EPA who has thought about the issue would agree. It's a shame it has become an issue at all, as it shouldn't be.

  28. I don't have much more to add to the discussion, but just in case others have not read it, I thought I might leave a link to Memorandum C of the Munich Diplomatic Conference 1973.

    This is the paper referred to in T1222/11 (which I think has been discussed previously on this and other blogs) in their analysis of multiple priorities, and confirms the solutions of Peter and Tufty.

    Actually, something has just come to mind which I would like to add to the discussion. With the greatest respect to Mr Justice Arnold, I believe that the source of all the column inches (or should that be blog inches?!) of discussion on multiple/partial priorities is a result of possibly an erroneous analysis of the nature of the housing at para. 96 of Nestec v Dualit.

    It is asserted that the claimed housing may be formed in one of three places, derived with reference to para. 0011 of the patent. The positions are (i) in the moveable part, (ii) in the fixed part or (iii) divided between the moveable part of the fixed part. The reason given as to why priority was not afforded to any option is that option (iii) is not a clearly defined alternative to option (i) or (ii).

    However, for the purposes of assessing priority in the claim, I believe that only two options exist: the housing is either (i) only formed within the moveable part or (ii) not formed only within the moveable part. Alternative (i) is entitled to priority. Alternative (ii) is not entitled to priority, but is not disclosed in the priority document. Thus the notion of "poisonous priority" in respect of this set of facts disappears.

    Please feel free to correct me if I have misunderstood the Judge's analysis!

  29. I actually quite like the imaginary claim splitting solution as discussed by Tufty, and think it has more weight given that the old Memorandum seems to propose it as well. Peter's article is also good, though the non-imaginary version of actually amending the claim might have added matter risks.

    My objection is to the argument that it is "unjust" for the priority system not to behave like a grace period. I don't think there's a moral argument there as that's not what the priority system was set up to be. Legal arguments are fine, so if the right way to interpret the Paris Convention happens to result in a situation where claims in a later application are safe over claims of the priority application, then that's good news for applicants. This does seem possible, with the imaginary claim splitting interpretation.

  30. Let's imagine that the priority application discloses "1 to 5" and the subsequent application claims "2 to 4".

    In this case I understand that the subsequent application cannot enjoy priority, is not new and the claim cannot be split: am I correct?

  31. Joe: The problem there is the "doctrine of equivalents" (or equivalent). If the housing is almost entirely within the movable part, but slightly goes outside of it (but not enough to make a difference), then that would still come under the priority document (option i), despite also being part of option (ii) in your reasoning (and option (iii) in the judge's). Since there could be a product that is in both alternatives, they are not clearly defined.

  32. Claim to a value of x = 4 is entitled to priority. Later claim to range x = 1 t 10 is not entitled to priority. Split the later claim to avoid anticipation, ie x = 1 to 10 except 4.

    How do we measure the scope of each claim? Which claim covers x = 3.99999?

    Such claims must lack clarity in many circumstances.

  33. I just returned from peddling a shipment of moonshine at the teetotaller conference, and hope I'm now sufficiently trained to take on this crowd. ;-) You can start cooking tar and plucking feathers...

    The commentators seem to feel or imply that this decision constitutes a hideous miscarriage of justice resulting from a fanciful and original misinterpretation of the law.

    I must point out that Arnold J is not the creator of this line of reasoning. My evidence is a good 54(2) document, a doctoral dissertation published more than 20 years ago:

    Edgar Lins, "Das Prioritätsrecht für inhaltlich geänderte Nachmeldungen", Verlag Peter Lang GmbH, Frankfurt am Main, 1992, ISBN 3631450176

    Pages 14 to 18 ("Spezielle Problemfälle") address the exact issue at stake in the present case. On page 16:

    Nimmt der Anmelder für eine Nachanmeldung die Priorität der Erstanmeldung in Anspruch und offenbart in der Nachanmeldung das in der Erstanmeldung lediglich in der Zeichnung dargestellte Merkmal nunmehr als zur Erfindung gehörig und richtet hierauf allein oder in Kombination mit anderen offenbarten Merkmalen einen Anspruch, kann aus denselben Gründen für diesen Anspruch nicht die Priorität [...] anerkannt werden [...].

    This is apparently the situation here (cf. §102 of the decision).

    Page 17 goes on to explain the possible consequences of lost priority:

    Wird in diesem Fall der Anmelder [...] gezwungen, auf den Prioritätsanspruch gänzlich zu verzichten, stellt bei einer deutschen Nachanmeldung die Erstanmeldung, wenn sie ebenfalls Deutschland betrifft, eine ältere Anmeldung dar, deren gesamter Inhalt für die Prüfung der Neuheit zum Stand der Technik gehört (§4 PatG und Art. 54(3) EPÜ). Kann der Fachmann das betreffende Merkmal [aus] der Zeichnung der Erstanmeldung entnehmen - wenn auch nicht als zur Erfindung gehörig -, wird die eigene ältere Anmeldung, für die der Anmelder an sich die Priorität in Anspruch genommen hatte, der späteren Anmeldung neuheitsschädlich entgegenstehen, wenn der Prioritätsanspruch als unzulässig angesehen wird. Dies ergibt sich daraus, daß für die Offenbarung eines Merkmals als "zur Erfindung gehörig" ein Unterschied zur neuheitsschädlichen Offenbarung gemäß §2 PatG 1981 bzw. Art. 54 EPÜ besteht.

    I find it unfortunate that §111 of the decision does not delve into the substance of the analysis of novelty w.r.t. the prio doc. I would be surprised that the arguments of the parties significantly diverged from those in the corresponding opposition proceedings before the EPO. (The wrapper is quite interesting, together with those of the related opposition cases. I like that bit!) The division had decided one year before Arnold J that for at least some requests the prio doc could not support the priority claim, and that it was also novelty destroying. If the guy with a wig got it wrong, he wasn't alone... The procedurally messy case in now before the BoA. (The patent was maintained in amended form. The Opponents squeal about Art. 113 violations, and it appears to me that they have a point).

    Footnote 16 on p. 17 lists references to various GRUR essays on the relevant German and EP case law, including a text by Singer on the then-current EPO practice. The treatises by Benkard or Schulte on German patent law are also cited. I don't currently have access to GRUR, but I note that the references span the 1970s and 1980s. I see nothing hinting that either Lins or the authorities he cites felt this "self-poisoning" doctrine to be abhorrent. I don't feel any revulsion either.

  34. Part 2/4:

    The authors of the blog post first underline that improvements to inventions often occur after the initial filing. Fair enough, this is widely known. The improvements will only be entitled the filing date of the second filing. I think this is common ground.

    The argument then subtly slips by equating "broadening" with "improvement", and somehow implies that any "improvement" ought to be automatically accorded the priority date. I have trouble following them on that path.

    They write: Further, an inventor may find that a feature he initially considered essential and explained as such to his patent attorney, is non-essential after all.

    The recognition that a specific feature (eg: "copper") can be generalised (eg: "metal") -- or deleted altogether -- very much represents a new technical teaching by itself. It follows that specific embodiment of the first filing ("copper") cannot support priority for the entire range later claimed ("metals": bismuth, lithium, copper, iron, etc.), YET it can at the same time be held against it should the condition (namely: publication) of Art. 54(3) be met. Furthermore, the other possibilities (Bi, Li, Fe, etc., save Cu) are not in my opinion "improvements" but rather "different inventions" or "alternative embodiments".

    The post laments upon the difficulty of getting protection for an improvement, new insight or redefinition of the invention [...] based on the same figures. Isn't it the representative's job to UNDERSTAND the invention and determine the insight of the inventor? (I am maddened by attorneys that merely recast the inventor's disclosure into a generic document in which every single feature is given as being optional. What is the technical problem? What are the essential elements for its solution? The image of nailing Jell-O to the wall comes to mind).

    This rhetorical device is then put forward: Was all this the intention when the priority system was created in 1883?

    Certainly not, and neither did the framers of the Magna Carta and of the Declaration of Independence... But it wasn't their goal either.

    To understand what happened 130 years ago, I recommend this hard to find book which I once borrowed through interlibrary loan: Yves Plasseraud, François Savignon: "Paris 1883: génèse du droit unioniste des brevets", Litec 1983, ISBN 2711104559. It consists mostly of a compilation of contemporary documents, but is still readable. A more accessible ressource is the Commentaire et critique de la convention internationale du 20 mars 1883 at Gallica (the answer of the Bibliothèque Nationale de France to Google Books).

    A lot of new ground was broken in those difficult negotiations, but an exact definition of what constituted a "same invention" was not a subject of the discussions. You can see that for yourself by downloading the document and scanning for "même invention". This was apparently left for national law to decide, or maybe it was just too early to even think about this question. Other documents are available commenting on the follow-up conferences (Brussels, The Hague, etc.), which don't go either into that question.

    Are the commentators perhaps pleading for a return to "novelty test" for deciding what constitutes the "same invention", which was already deprecated long before G2/98?

    The universal "directly and unambiguously derivable" approach is already rather applicant friendly. (Yes, really!) The Lins reference indicates that the former Reichspatentgericht was way more demanding for acknowledging "identity of invention".

  35. Part 3/4:

    I have a lot of difficulty feeling sorry for the proprietor, and don't regard the law relating to priority and interfering applications as it is drafted and interpreted to be fundamentally unjust. The rules of the game are known and identical for everyone.

    I would liken this case to the situation where an applicant files a second application more than one year after the first filing. He gets a nice letter from the FO informing him that the prio claim is invalid, and then a nice communication from the examiner telling him that the second application is nuked by its own "prio". But I hear you cry: "what [was] the damage to society if the priority application [was] published?".

    There are plenty of other reasons to get worked up about this case.

    For instance, that the claim that was knocked out by its own "prio" was actually cobbled up afresh six years after the initial filing and foisted upon the public in the form of a "divisional", with the benefit of knowledge of the prior art put forward by the EPO and the the opponents in parallel cases.

    That even though the invention concerns the design coffee brewer, the claim "innocently" includes a definition of the capsule, which was part of the prior art at the filing date, and are now off-patent (§20 of the decision: "Nespresso capsules were formerly protected by Nestec's [EP512148], which expired in May 2011.").

    That this claim is then asserted against Migros, Casino, etc., who manifestly have no interest in the alleged invention per se. These distributors package and sell coffee, not brewers.

    I will contrast the preceding with your own statement: The patent system is created to stimulate innovation. It does so on a quid pro quo principle. The inventor gives information on an invention to society, and society gives back an exclusive right for a limited period, a patent. If the invention is big, the patent is big. If the invention is small, the patent is small.

    The incremental improvement to the capsule holder is hardly of the civilisation changing type. Nestec is manifestly attempting to perpetuate her monopoly for supplies, in the best tradition of razor blades or printer cartridges.

    You could also question whether you as an end user would really be tempted or able to take apart a demonstrator coffee machine lent to you for a field test (I say yes), but would never inquire or deduce details about a control algorithm during a guided tour of a rolling mill in China (I say no).

    You mention the danger of your national patent filings coming back to haunt you under the proposed community patent. How about petty patents spawning like viruses claiming the priority of utility patents, like this German one here? There are a few of these that thicken the thicket even more.

  36. Part 4/4:

    On a final note, notwithstanding the preceding considerations, I think that Nestlé should get to keep her patents. My perverse logic is that these coffee makers are a really terrible invention, and that a monopoly would reduce the overall number of capsules consumed.

    Last time I saw those capsules from up close they appeared to be made entirely of aluminium, even if the patents allow other materials.

    Many, if not most, spent capsules end up in landfills.

    For those capsules that are left in those blue bags or yellow or green containers by the kerb, recycling facilities have trouble separating the small objects from the stream.

    I don't know whether the current capsules contain liquid or ground coffee, but the latter type certainly add contamination that must be separated at some point.

    Melting and reprocessing spent aluminium still consumes a lot of energy.

    For everyone of those new capsules, more than their weight in tailings in the form of toxic red mud is abandoned behind a dam somewhere for all eternity -- or maybe not...

    And a little bit of fluoride finds its way into the environment, if not in your drinking water...

    Then there is the CO2 released from the manufacture and consumption of anodes.

    And let's not forget the shady deals involved in procuring cheap cheap cheap and dirty electric power generation, for example in Australia. Other cases abound.

    And of course, the plight of the coffee growers themselves, who are exposed to all descriptions of chemicals.

    The coffee suddenly doesn't taste so good anymore, even when I look at George Clooney.

    A solution to all that misery is to forego coffee. BTW, how about a bit of 90° proof that will blind no one but the ol' revenoo officer? I still have a bit left over. Actually, the entire consignment...

  37. What is rubbish is the idea that an applicant's own unpublished application can be treated as 'prior art'. That is simply crazy. The EPO is trying to insert into law the concept of 'double patenting' without any statutory basis,on the ground that 'the applicant has no legitimate interest in obtaining two patents for the same invention'. This should be balanced here by recognising a legitimate interest in obtaining one such patent.

  38. I'm with Tufty, and the original blog post. It's ridiculous that your own disclosure, from which you are entitled to claim priority, should invalidate you.

  39. Except you are not so entitled, Tim.


  40. This discussion reminds me of my discussion on Catnic with my late father, a mechanical engineer by profession and with a wealth of experience with a major oil company. According to him a difference of few degrees at a bottom angle would result in a difference of several (read - a lot) degrees at the top. He boldly rejected the coveted purposive approach - "Your case is wrongly decided". Now I routinely see, even in the US applications, examiners routinely object to inclusion of word "substantially" and ask applicants to remove it from their claims. Now I am wondering what are we asking here for? The law on priority claims is so settled, why and what motivated the applicant into taking risk on such a settled issue? Don't we patent practioners have a tendency to stick to the law on such issues? Creative solutions can be risky and debilating, and may not work all the time.

    OR, should I say, (with a chuckle) patent litigators are turning smart because now they are looking into the issue of priority which is so often than not taken as granted at the point of subsequent filings, and not much cared for - a string of recent patents losses solely on the issue of an invalid priority is the testimony to that neglect.

  41. I want to pick up on what Tim Roberts wrote at 19:26 yesterday June 13.

    There is nothing to stop me declaring as my priority, my "earlier application", one with a disclosure that has nothing at all in common with the claims in view. So, I can declare the priority of every single one of my own disclosures made within the preceding 12 months. It is not "ridiculous" that, if the so declared earlier filing is not for "same" subject matter as is claimed, then my claim does not get the priority.

    Neither is it "ridiculous" that one or other of my own multitudinous earlier filings can prejudice the novelty of the said claim.

    Since 1973, patent attorneys outside Europe dislike Art 54(3) EPC because their own earlier unpublished filings can destroy their novelty. Jurisdictions outside Europe are stalwart, that only the filings of others can have this destructive effect. But as far as I know, there is within Europe, still no outcry, that this aspect of Art 54(3) EPC is iniquitous.

    But now there is an outcry. Why? Because of the A publication of something termed a "divisional". What makes a "divisional" so special? As I see it, the debate boils down to whether a divisional is to be treated for Art 54(3) as the "same" as the parent from which it is divided, or just another "different" one of Applicant's earlier unpublished filings.

  42. Seems to me that half the people who have posted here have not bothered to look at T1222/11 and its reference to Memorandum C of the Munich Diplomatic Conference 1973. (See Joe's post above).

    This says it all and confirms that Tufty's solution (shall we call it "Tufty's law") is what was intended by those who drafted the EPC. Please take the time to read and understand it before posting.

  43. Roufousse is spot on (except in part 4/4).

    I would really like to understand why others feel that Nestle was unfairly treated?

    Their second application claimed a different invention than was described in the first, why on earth should they be entitled to priority for a different invention?

    And if they are not entitled to a priority, why should their own earlier application not be considered A 54(3) prior art.

    Which aspect is unfair?
    -priority not allowed or A 54(3) tout court?

  44. MaxDrei

    You said "Jurisdictions outside Europe are stalwart, that only the filings of others can have this destructive effect."


    I do not think so. Prior art includes a priority application should priority claimed from the application becomes invalid. It joins the rank of other prior art.


    Additionally, the content of European patent applications as filed, the dates of filing of which are prior to the date referred to in paragraph 2 and which were published on or after that date, shall be considered as comprised in the state of the art.

    where does it say only the applicant's own filings will incurr wrath?

  45. To the anon poster at 11:42.

    Sorry that with my "only" I did not make myself clear. I will try again.

    Under the EPC, the earlier unpublished patent applications of the Applicant herself are good material for novelty attacks.

    Outside Europe, in countries like the USA and Japan, "only" the earlier unpublished applications of others can be used as a basis for novelty attacks. The earlier unpublished applications of the Applicant herself are excluded, as novelty attacks. Accordingly, the world outside Europe is going to struggle to understand what all the "poisonous divisional" drama is all about.

    Why the difference between the EPC and the rest of the world? See the Paper from the Tegernsee Group, on the EPO website. It appears that it is all too difficult, for patent drafters in Japan, to draft with novelty over their client's other filings. In the USA, 54(3) docs can be used in obviousness attacks. If all your own earlier filings can render you obvious, that will thin down your patent portfolio rather seriously, I should have thought.

    Should Europe harmonise with the rest of the world? I think not. I think that Europe, in 1973, actualy took the best bits from American and Asian patent law, and carefully shut out the bad bits. I think the EPC is all the harmonisation the world needs, already done and dusted, back in 1973.

    Come to think of it, most of Asia has a grace period. Should Europe introduce one too, in the interests of harmonisation? I hope not.

  46. Anonymous @ 23:40. Certainly I am entitled to claim priority for what I disclose. But in the way the law is being interpreted, the opponent is entitled to claim priority (for the invalidating disclosure) and I am denied it.

  47. This discussion makes the best argument for a grace period that I have come across.

  48. Absolutely no need for a grace period. Attorneys need to learn the law and learn to draft and then there will be no such problems. But then, aren't attorneys supposed to be so qualified in the first case?

    So many attorneys harp on about the EPC and the EPO just because their cases fall down, which in the majority of cases is probably down to their own failings. Recent referrals to the Enlarged Board only serve to support my view.

  49. MaxDeri

    Singapore Section 14(3), which I think is equivalent to EPC Art 54(3) reads

    (3) The state of the art in the case of an invention to which an application for a patent or a patent relates shall be taken also to comprise matter contained in an application for another patent which was published on or after the priority date of that invention, if the following conditions are satisfied:
    that matter was contained in the application for that other patent both as filed and as published; and
    the priority date of that matter is earlier than that of the invention.

    The word "patent” means a patent under this Act and includes a patent in force by virtue of section 117(3);

    So I would think, that not only the applicant's own earlier filed, unpublished application but also those filed by others in Singapore would be citable as prior art.

    Am I missing something? Please enlighten.

    Also, I find it hard to accept that
    "It appears that it is all too difficult, for patent drafters in Japan, to draft with novelty over their client's other filings."

  50. Here is a piece I just posted to Tufty:

    Having followed the more recent thread on the IPKat, and read again the Tufty solution above, and thought about it a bit, I want to add my support to it.

    What comes to mind is the provision in the UK Patents Act 1949 that graced the disclosure content of the priority document. It went like this:

    If you had told the Patent Office (in your prio doc) then told the world during the Paris year, any claim of yours entitled only to the date of filing of your "complete specification" at the end of the Paris year could not be attacked on the basis of your public disclosure, during the Paris year, of that which you had already told the Patent Office about, in your prio doc.

    The EPC had no such provision. Many in the UK warned "No good will come of it. It ain't fair to Inventors". But nobody took any notice.

    Instead, we got the "Claim can have multiple priorities" sentence.

    What I like about Tufty's "imaginary" (square root of minus one) claim splitting is that you do not have actually to find the words to do it. As we all know, at the EPO it very often happens that everybody can see there is still patentable matter there, after the EPO's excellent search activity reveals closer art, but one can't protect it simply because the words you need are not in there, verbatim, in your appln as filed.

    So, given the strict line the EPO has on verbatim text, I really like it that, with the Tufty solution, a wave of the hand is instead enough.

    The EPO should like it too, should be grateful for the help to inventors that it provides, and for the relief from the mounting tension in the wretched area of "disclaimers".

  51. Thanks, anon, for that clarification on Singapore. Singapore perhaps takes more note of what the UK does, than do other jurisdictions in Asia. My impression of Japan and Soeul is that they try to agree first and foremost with the USA. Beijing might be more influenced by the EPO but why should China want to make difficulties for its domestic applicants, by adopting the full rigour of the EPC's Article 54(3).

    Upthread, there is reference to a Grace Period. Tufty's "notional claim splitting" idea can deliver a sort of Grace Period, I should have thought. File at the PTO, then disclose to the world, during the Paris year. If you disclose no more than the content of your priority document, you can have your PCT claims valid, even if they are not entitled to the date of your prio doc. Bingo.

  52. I can't see the EPO relying on imaginary splitting of claims. An invention as claimed must be unambiguously derivable from the application. Surely, an invention whose priority is claimed must also be unambiguously derivable from the application?

    In the case of an imaginary splitting of the claim to support the priority of two inventions, this would aid in avoiding the self-conflict issue, but what about a subsequently identified intervening publication destroying the novelty of the later invention? In practice, the claim must amendable to limit to the invention entitled to that with the entitlement to the earlier claimed priority.

    There are other examples in the comments above that the Tufty position fails to address, and for which Tufty has not commented. The analogy to the mathematical symbol "i" is, in my view, unhelpful. The problem with such analogies is that it causes people who understand the analogous term to believe the new theory make sense. The theory needs to stand on its own.

    The old blog discussion referred to by myshkin is doing a better job at assessing the legal position that this current blog which unfortunately is full of "life ain't fair" comments.

  53. Well thanks anon, and I do appreciate that a claim has to be clear, but here I do not advocate using the "i" idea to overcome just any Art 54(3) issue. For every other 54(3) doc except Applicant's own prio doc, old Rules apply. ONLY for Applicant's own priority document is the "i" idea invoked.

    What offends so many reasonable thinkers like Tim Roberts is the idea that your own prio doc is no help to the inventor but is a killer in the hands of everybody else. That is assymetric and unfair.

    As I understand the "i" theory, it requires no real re-writing of the Claim. No Claim amendment is needed. It is simply a thought Experiment for neutralising your own prio doc as a novelty killer.

    Look, that principle (that your own priority doc should not be the Instrument of your own execution)was unassailable, in the UK prior to 1978. The Tufty "i" theory is not a loony tune, simply a way to restore sanity.

    I too can't imagine many within the EPO swallowing it, true. It does take a bit of imagination, which many there lack. EPO people will be well aware that the Rest of the World simply does not "get" the EPO's Problem and Solution Approach to the obviousness enquiry. That too requires a bit of imagination and familiarity. But still the EPO perseveres with it, for the good reason that it manages the obviousness enquiry fairly and cleanly.

  54. Max, I'll just get the problem solution approach out of the way first. Those that do not understand it need to clear their minds and work with it. I can work in both imperial and metric and am equally able to apply the problem-solution approach as the pozzoli test (though I see nothing wrong with the original windsurfing decision, so let's stick with the label).

    I'm still not sure I understand Tim's problem. The priority doc is of help to the inventor for the same invention. It is of no help to 3rd parties in respect of applications direct to said same-invention.

    I'm not opposed to the idea of an inventor's own filing not being held against an 'improved' application, but I don't want to see attempted fudges, such as the 'i-claim'. I'd prefer to see the issue dealt with head-on, even if it involves a change of law (yes, fat chance) or a useful re-interpretation of the current statute. Maybe an EU-regulation?

    For the 'i-claim' interpretation to stand-up, the real-world scenario of an allowable claim amendment must be dealt with. This is also, in my view, a required interpretation of the T-decision, which posters to this blog have been accused of not reading.

  55. Thanks anon. We differ in the area of your penultimate sentence, where you use the word "must". I don't see that imperative yet and I would like to think that, one day, the EBA will also not see it as a "must".

    OK there is already a T-Decision. But one of the benefits of the DG3 set-up is that a single T-Decision is not binding on anybody. Another benefit is that the EBA can take an issue, like "poisonous" , only after it has been thoroughly debated, and then make a wise opinion that helps all the TBA's to get on to a consistent line on that issue, for ever more. I'm thinking, for example, of how the EPO now judges the patentability of computer-implemented inventions. I look forward to the day when the EBA hands down a definitive opinion on the poison issue, that takes full account of how the world thinks about inventor disclosures during the Paris year killing the claims they file at the end of the Paris year, directed to subject matter after they have seen the results of an EPO search on the claims of their priority document, no better than the subject matter of their prio doc, unimproved but nevertheless in the eyes of the EPO not "the same" as in their prio doc.

    I do engineering rather than chemistry cases. I don't have punctuate lists, one, two or more. For me "AND" claims and "OR" claims are artificial simplifications. Instead, I have conceptual terms in my claims like, say "ratchet" and I have a specific form of ratchet in my illustrated embodiment.

    During the Paris year, I find out that "ratchet" per se was not novel on my prio date. A 54(3) doc deprives my conceptual claim of novelty but does not disclose the specific sort of ratchet in my drawings. So, for the first time, I need an intermediate generalisation to restore novelty, one that I can only visualise after having been given sight of the search report doc.

    I file a divisional at the EPO. My revised claims, in both my EPO cases under substantive examination, are not entitled to my prio date but they are new and inventive over everybody else. My difficulty is that none of them is novel over my own priority document. So both of my two cases go down on novelty. Each destroys the other. I say, that's not right. An "i" claim legal fiction would save both my cases though.

    For me, the inventor's own prio doc is different from all other 54(3) docs because its illustrated embodiment is identical with the one in the later application that is the support for the claims under examination. Every other 54(3) doc can be finessed away by a disclaimer, but not my own priority filing. So I need a fix, just for the special case of the prio doc.

    Please, where am I going wrong?

  56. Max you go wrong in two occasions. You state "you need an intermediate generalisation". In other words you need to add matter to avoid a prior art document. Unfortunately this isn't allowed.

    Secondly, the other problem is that with this theoretical "i-claim" solution to the problem is that the claim is not to a single inventive concept but at least two. This is also not allowed.

  57. Thanks again, anon. Encouraging and depressing at the same time. Not every intermediate generalisation under the sun adds matter per se. And your "plurality of invention" point, I don't see that being an automatic consequence of notionally splitting a claim in two to give it multiple priorities and so avoid a self-collision purely novelty issue.

    Encouraging because, if that's all you've got, I don't see it as beyond the imagination of the EBA to come up with a fair solution.

    Depressing, in the way you give the impression of gleefully banging your gavel twice. Thus 1) "Intermediate generalisation is it then, what you've got there? OK. FAIL." and 2) "Claim divisible? OK then. Two inventions. Another FAIL. Sorry."

    Rightly or wrongly, I have gained the impression that, for the bean-counters of the EPO the "Rules are Rules", "More than my job's worth" and "Don't Rock the Boat" are what count, rather than fair outcomes for genuine inventors.

  58. 11:27 is a new anon.

    The problem applies also to chemical inventions, so is not subject-matter specific.

    If 'ratchet' lacks novelty I see that as a separate issue. You are still entitled to file a claim directed to your preferred embodiment. If there is no basis for a valuable claim based on the illustrated embodiment, then is that not a drafting issue?

    You are still able to add and claim other embodiments in later applications without the priority disclosure being novelty destroying.

    If, however, the priority discloses only a specific ratchet and you later expand you application to 'ratchets', then you would have a novelty issue. But again, why is that not a drafting issue? Even if the issue was only identified when drafting the later application, at worst you would only have a unity problem as you attempted to cover multiple embodiments. But even here, 'ratchet' would be an allowable unifying feature, as multiple priorities are irrelevant.


  59. Again, I think that with the ranges you have a clear example of the problem:

    You disclose "1 to 10" in the first application, then you disclose to the public "1 to 10" and in the prior art search you find that "2" is already disclosed, so you file a second application claiming "5 to 10" and also claiming the priority of the first application.

    At the end you find out that your priority claim is invalid and that the second application is not novel over the priority application.

    I do not think that this is "fair".

  60. In response to this anonymous comment, I have yet to see an example that would not work, given the very limited circumstances that the poisonous priority/divisional problem applies to. The examples others have come up with involving various ranges and intervening publications miss the point entirely, as they do not apply.

    You may consider the mathematical analogy to be unhelpful, but that doesn't really matter. The solution stands up quite well by itself. I just thought that the comparison with imaginary numbers was quite nice. I wouldn't expect everybody to understand it though, particularly those who don't have the required maths.

    The main problem with the solution is, as has been pointed out, that it requires a bit of imagination that might be lacking at the EPO. I still think it is worth a try though, and look forward to seeing somebody have a go. They at least now have plenty of ammunition to work with.

  61. On thing is clear from all this is Tufty's inability to accept there are questions over the theory that need addressing. Comments/queries are simply dismissed as wrong/missing the point/not relevant. All without explanation. Zero points in an exam for not including any reasoning.

    The main problem is not that the EPO have no imagination. Just how can such a comment have basis?

  62. I'm not convinced this comment was correctly published when I clicked the button so I'll have another go...

    @ I'm a robot:
    I don't know if it's still considered a good decision (I note no other decisions cite it), but in T680/08 such a scenario was solved by claiming "from 5 to 10 but not 5" or ]5, 10]. Commented upon by K's Law here and here.

    @ Tufty
    "I have yet to see an example that would not work." What are your thoughts on its applicability to Nestec v Dualit, c.f. my comment last week?

    In view of Anon's comment of 13 June 2013 16:24, I still think that the analysis of multiple priorities takes place almost in a "priority vacuum," separately from other claim construction issues. For instance, in the case in question, when assessing the right to priority, the subject-matter disclosed on the priority date was that the housing was in the moveable part, which was extended to being also not in the moveable part in the patent.

    This is a comparable situation (I am open to correction here) to the third example in Memorandum C, concerning coating the inner walls of hollow bodies, priority being afforded following claim splitting for those bodies being pipes, and not for those bodies not being pipes.

    I think para. 11.5.8 of T1222/11 sums this up nicely, for those that have not read it:

    Thus the Memorandum, which was considered by the Enlarged Board to express the legislative intent underlying the possibility of claiming multiple priorities for any one claim as defined in Article 88(2), second sentence, EPC, confirms through the situations exemplified therein the significance to be attributed to the condition referred to in point 6.7 of the Reasons for opinion G 2/98 ... namely the ability to 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.

  63. I may have missed the point but then this discussion does not comply with Arts. 84 and 123(2) EPC

  64. In response to this anonymous comment, I am not unable to accept there are questions over the idea, and readily admit I may be wrong. I just don't have the time/inclination to respond to queries where the example facts don't even apply to the problem in question.

  65. In response to Joe's comment above, my thoughts are that you might be right. If you are, the judgment in Nestec is flawed in more than one way. However, it doesn't really matter as the main point is about whether a narrower priority application can in principle anticipate a broader later application (see paragraph 111. It looks to me like the point was not even argued properly.

  66. At I'm a robot-17 June 2013 17:16:00
    Realistically the situation is rather more likely that the inventor found that 4 and 7 worked, and then considered that 2 to 8 should work, and then added a bit extra and claimed the nice round numbers of 1 to 10;
    Let’s look at what he then really disclosed: He disclosed 4 and 7, and he disclosed the opinion of the expert that 1-10 might work. That’s his technical contribution to the world, nothing more, nothing less.
    So now it turns out that 2 is already disclosed. And all of a sudden his expert opinion of what might work changes from 5-10. Yeah, sure….
    He’s entitled to change his mind of course, but the (perhaps undefined) technical reason why the expert believes that 5 to 10 should work must by definition be different from the technical reason that made him believe that 1-10 should work.
    Ergo, the 5 to 10 must be based on a new technical disclosure, so it is logical that there is no priority for this.
    Should his priority document be held against him? That’s a different matter. I see this as a risk he willingly took, and which then backfired.
    The inventor would argue that he has placed a claim for 1-10, and he is fully entitled to narrow that.
    But…his true contribution to the art was only 4 and 7. He is entitled to get protection for that. He is entitled to some generalisation as well, based on his expert understanding of the subject matter, but must accept that generalisation carries risks. In short, since he only technically disclosed 4 and 7, anything more is an, arguably undeserved, bonus.
    A patent application is also a technical document with a technical teaching. Lawyers tend to forget that sometimes,
    With NESTEC it is even worse. At priority date they considered a feature to be technically important; and later found out it wasn’t. Why then should they get legal priority for the later technical finding that a feature wasn’t important?
    I would really like to see some reasoning there.
    And they could have gotten close to what they wanted if the’d continued the prosecution of the priority filing. Instead they chose to play the system, and they lost. They overbet their hand, simple.
    I can understand that people want milk the system to its max for the benefit of their clients, but realistically, deep in your heart, do you really think it would have been fair if NESTECs claim to priority had been allowed?

  67. News just in: Tufty admits to not being infallible, but is still too important to be bothered with queries raised by unimportant folk.

  68. In response to the further anonymous comment, if you care to raise a point that is relevant to the problem in question I will see whether it is worth answering. If it is, I am sure either I or somebody else will have a go. Alternatively, of course, you could just be trolling.

  69. Quick reaction to B Brandsma above is that we must be careful to keep the discussion free from obviousness issues. 54(3) is not about obviousness, more about adjudicating between rival filers who are all inventive over what has been made available to the public. So, we address novelty tests, photographic, directly and unambiguously derivable, and forget "contribution to the art" and other litmus tests of patentability, for all we have to settle is who gets what scope, amongst rival filers who are all inventive.

    54(3) is often said to be the antidote to double patenting. For that job, it has to adjudicate between rival filers. What function does it perform, as between parent and divisional of the self-same Applicant? To my mind, that is the issue here.

  70. @ MaxDrei

    Art. 54(3) may well be intended to be solely concerned with stopping double-patenting, but I often think of it in the following way. I do not know if this is consistent with the real intention of the provision, though.

    When filing a patent application, that there is evidence that an invention was known at the filing date. The details of this invention may not have been made available to the public, but someone somewhere knew of it.

    Thus, if someone files a later application for the same invention, it is not new, because it was known at an earlier point in time. For a slightly different invention (i.e. novel), and in terms of the second hurdle (inventive step), it would be unfair though to say to an inventor that the notional skilled person has access to unpublished patent documentation when the inventor did not. I find 35USC 103 quite unfair as it allows this.

    Now, with regard to your question of what function does Art. 54(3) perform as between applications of the same applicant? Well, quite typically, the response is that oft-used phrase "fairness to third parties." Why should an inventor get a patent for something that was known at an earlier date (by virtue of their own filing of a patent application) when third parties would not?

    That deals with priority. For the Poisonous Divisional® scenario, I think it is (only just!) fair when you look at the above principle. Consider filing two applications on the same day, both declaring a priority date established by the filing of an earlier application. One claims an invention in broader terms than the earlier application, and the other claims it more narrowly. Say the broader application did not validly claim priority, then the other, narrower application would become citable because it is entitled to the priority date.

    This is of course what, in effect, happens when you file the divisional. An unfortunate consequence of the whole contents approach.

    At least you have a choice with the Poisonous Divisional® problem. There was a comment on an earlier post in which someone stated that this is all well and good, but the thing you have to worry about "in the wild" is inventors' own disclosures.

    As an aside, I have noticed an subtle difference in wording between Art. 54(3) EPC and s. 2(3) PA77. The former simply refers to the content of European applications (this must mean all applications. The latter refers to matter contained in applications for other patents. I think either in this thread or another a question was raised as to whether a European application, which, at its date of filing, validly declares a priority date for matter disclosed therein, but whose claims are not entitled to priority, can anticipate itself. I have not (i) seen any comments saying it can't and (ii) I can't find any authority that definitively states otherwise. Help?!

    I can see now that this discussion has hit around 70 comments. I think this is not a case of beating a dead horse - I think it's both a great reflection of the enthusiasm of all commenters for what (for IP!) is an interesting, foundational and somewhat philosophical topic and emphasises that forums for discussion such as the IPKat are invaluable. Try getting colleagues talking about this sort of thing at the bar after work!

  71. Thanks Joe. Art 54(3) is confined to earlier unpublished applications in the same jurisdiction. If somebody outside the jurisdiction filed earlier than you, you are not the first to invent, OK, but so what? We don't do "First to Invent" in Europe but your vision of 54(3) is, to me, tantamount to a "First to Invent" philosophy.

    No, I think 54(3) is to adjudicate between rival filers at the same Patent Office only. They can't all have patents with full scope covering the full area. So we need those who filed first to get first bite of the cake, and those who came later to get the crumbs.

    As to self-collision, you invoke notions of fairness, to justify poisonous prio docs and poisonous divs. Me, I struggle to see it, but I do think I can see UN-fairness, if 54(3) attacks based on your own prio doc are legitimate.

  72. From the toxic divisional front T1439/09 issued today. The opponent was inspired by T1496/11:

    1.10 The appellant submitted that the priority document of the case under appeal did not disclose this subject matter. Thus, the relevant date for the assessment of novelty was the filing date of the patent application.

    The appellant further submitted that the divisional European patent application, EP No. 06011535, arising from the patent under appeal, disclosed antibodies falling within the scope of claim 20. In addition, the antibodies according to the divisional application were disclosed in the common priority document. Under these circumstances, claim 20 lacked novelty according to Article 54(3) EPC over its own divisional application.

    1.11 The appellant submitted, that the legal issues arising from this objection were not complicated and that admitting the novelty objection at this stage of the proceedings would not lead to a significant delay of the procedures.

    1.12 The board is of the opinion that a sound assessment of the question whether a divisional patent application could anticipate subject matter of its own parental patent application requires a substantial amount of legal analysis which would unavoidably lead to an adjournment of the proceedings.

    In view of the late filing, i.e. less than two months before oral proceedings were held, in accordance with Article 13(3) RPBA, the board decides not to admit the late filed objection under Article 54(3) EPC into the proceedings.

  73. Part 1/5

    I mulled over this thing over the last days, and would like to add a few further observations that no one will read.

    Was all this the intention when the priority system was created in 1883?

    On a formal, nearly pedantic level: the Paris Convention, which is designed to facilitate the filing of second applications in foreign countries, is not relevant here, at least for the reason that the EPO is not a party thereto. The present case concerns an internal priority (EP prio for EP 2nd filing), which is nevertheless regulated by the same Art. 87 and 88 EPC that governs foreign priorities. Those EPC provisions are close to, but not identical with, the Paris Art. 4.

    So, again, what was the intention of the drafters of the original Paris Union?

    I stated in my initial comment that the original drafters did not give much thought about what constituted the "same invention". This suspicion got confirmed when I discovered the February 1891 issue of La propriété industrielle, the OJ of the Bureau International (the forerunner organisation of WIPO).

    The first three pages contain an article with the title "Portée et application de l'article 4 de la convention, en ce qui concerne les brevets".

    It was initially envisaged that an applicant could perform a simultaneous foreign filing at the other nation's consular representations. That would have brought about an improvement, as inventors practically had to sail overseas to secure foreign protection, and accomplish the administrative steps in the right sequence. But the accessibility of consulates was not good enough to make this a practical solution.

    At that point the delegates stumbled upon the idea of mutual recognition of foreign filings. Eight years after the entry into force of the Paris Union it was obvious that there was a large number of practical details that needed to be hammered out. Is a priority declaration sufficient? Essential? Or must a copy of the foreign application be supplied? Should the examination and duration of the national and foreign be linked? etc. etc.

    You will clearly see in the paper that the question of what constituted the "same invention" just began to be addressed, by mentioning inter alia the possible differences between British provisional and complete applications. The priority delay was shorter (6 months, or 7 months if you lived overseas), which didn't give you that much time for coming up with improvements (the 1 year delay was introduced in the Brussels text to get Germany on board). My conclusion is that in 1883 the drafters implicitly thought that the first and second filing would always be identical in content and thus represent in all cases the "same invention", whatever that means.

    Articles 4F(1) and 4F(2) were respectively introduced in the 1934 (London) and (1958) Lisbon texts.

    From the discussions on this topic I could glean in the issues of the B.I. OJ from the 1920s and early 1930s, it seems to me that it was mainly, or only, the case of "additive" improvements was considered, and not the one of a "lateral shift" or "broadened" scope. Adaptations between first and second filing could however also be brought about by conflicting national requirements, e.g.: your first filing is in a country that allows product claims, but the foreign filing is destined to a place where only process claims are permitted. The question of multiple priorities was entangled for a long time with the one of unity of invention and divisionals. I also see that the consensus was building to favour a "whole content" approach instead of the "prior claims" one.

    For 4F(2) I don't have (yet) material covering that period.

  74. Part 2/5

    We should ask the question: what is the damage to society if the priority application is published?

    The "damage" does not come from the publication itself, since the public would be apprised of the content of the prior application whose priority is claimed at the exact same time anyway, through Rule 53(1) and the opening of the second application under Art. 128). It comes from the birth of provisional rights conferred by Art. 67(1) EPC.

    The publication informs the world that by maintaining the prior application you intend to obtain the rights defined claims published, either in their original form, or through amendments filed after the receipt of the search report and before publication under Rule 137(2). These rights can potentially conflict with other rights you may later effectively seek, and Art. 54(3) (under the condition of Art. 153(5)) is the provision that takes care of this.

    Some have argued that for any further improvement in the priority year, another independent patent family should be created.

    Who are those "some"? What about Art. 88? If you create a separate family without any priority connections you run an even greater risk that one application will come to haunt the other one.

    Moreover, it is not always possible to be novel relative to the priority application without losing scope of protection or without taking at least some risk that the priority application is considered novelty destroying for the subsequent application after all.

    Come on. If you are the one who drafts the first and second application, all you need to do is to add a tiny little bit of subject-matter to render the second application novel w.r.t. the prior one. That's not too difficult, considering how narrowly novelty is assessed. You could for instance include an additional statement somewhere in the second disclosure to the effect that the gizmo comprises a material selected from the transition metals, Mg, C, H, Cl, O, N, Pb, Sn, or Al. (You cover just about any industrially useful material, metal, wood, or polymer). Should you run into problems you could always pull that out of your hat as a claim amendment and solve your problem, without having surrendered any real scope. Try to picture a competitor attempting to build the gizmo out of alternatives left, such as metallic sodium or argon... I had a case once where the claim differed from the 54(3) document (by the same applicant) only by the presence of a casing around the contraption.

  75. Part 3/5

    Will non-Europeans understand it?

    It is certainly easier to grasp and accept than, say, the (former) Hilmer doctrine, or the use of late documents for 103 rejections...

    As mentioned in my first comment, this risk isn't new. But how often did such a case occur in 35 years of EPO operation? Not too often, I think. As I argued earlier,you should also consider the other elements of the case, before you concentrate on the fairness of the priority collision.

    The problem I have with grace periods of any kind is that they create an inquiry into the identity of the party who made the disclosure, and revives the question of the respective rights of the inventor and of the employer. The First-to-File system would begin to look a bit like a First-to-Invent one. E.g.: prof. X publishes his great idea in a paper, and gets hired shortly thereafter by BigCorp, who thinks it would be nifty to get a patent on X's work, and files an application. Who made the original disclosure? Prof. X or Bigcorp? Who gets the benefit of the grace period? Prof. X or Bigcorp? Can BigCorp retroactively obtain rights on X's idea? Or Prof. X's university?

    The issue isn't only academic.

    Should you be shielded of your own prior application whose priority you do NOT claim? What if the first, colliding, application was filed by BigCorp's US operation (drafted in US style and with US inventors), and the second, collided, one by its European branch? Same applicant? I've met that one. Did the left hand really did not know what the right one was doing?

    I have been mentally building a number of cases with Alice filing EP applications A1 and A3, and Bob filing A2 in between. Depending on what is disclosed and claimed and whether A1 is published or not I get outcomes that don't quite feel fair to Bob if there were to be some sort of prohibition of self-collision or grace period.

    US-style CIPs and Jerome Lemelson also come to mind.

    It seems to me that the introduction of a grace period in EP law not very likely, considering the discussions made by the Tegernsee group:

    [The] Tegernsee heads should consider whether to mandate a closer examination of the feasibility of creating a common pool of international secret prior art under the PCT, by determining that a published PCT application will become prior art as of its priority or filing date, as well as the policy issues which such a change would raise.

    My translation and extrapolation: PCT applications wouldn't even need to be translated into an official language or enter the regional phase to be considered as interfering rights. One benefit: examiners wouldn't have to wait 31+ months before deciding whether that cute WO document could be used to attack novelty.

    I read this as an extension of the prior that falls under Art. 54(3), probably through an amendment to Art. 153(5), which technically could be promulgated by the Council alone under Art. 33(1)(b), should the work of this group eventually find its way into an international Substantive Patent Law Treaty or even the PCT itself.

    An amendment narrowing the scope of Art. 54(3) for avoiding self collision would, on the other hand, or introduce a form of free-for-all into Art. 88, probably require a full diplomatic conference, and would go against the general trend implied by a broadening of the 54(3) base envisaged above.

  76. Part 4/5

    The thread diverged on a slightly aggressive exchange about how the self-collision could be avoided or argued away in case of a broadened 2nd filing. I don't have access to the CIPA journal. Here's my take on Tufty's approach.


    SOTP: A slice of the pie
    WOTP: The whole of the pie
    ROTP: The rest of the pie [without the slice]


    Application A1 discloses and claims SOTP, with no mention of WOTP.

    Application A2 discloses and claims WOTP, with no mention of SOTP. Priority of A1 is claimed.

    A1 is published, and belongs to the prior art for A2 under Art. 54(3).

    Examiner deems priority of A1 invalid, and dares cite A1 against A2 alleging a lack of novelty.

    Representative Tufty argues as follows:

    The claimed pie contains a virtual dotted line, as visible as the Emperor's new clothes, that separates WOTP into ROTP and SOTP.

    The ROTP part of the claim is cut away from WOTP by the virtual dotted line and is patentable through the logic of undisclosed disclaimers, GL H-v-4.1

    Limiting the scope of a claim by using a "disclaimer" to exclude a technical feature not disclosed in the application as filed may be allowable under Art. 123(2) in the following cases (see G 1/03 and G 2/03, and F-IV, 4.20):

    (i) restoring novelty over a disclosure under Art. 54(3)


    The virtual dotted line makes the ROTP part of the claim entirely novel against A1, as it virtually disclaims SOTP.

    Then let's look at the SOTP part. The slice within the virtual dotted line coincides exactly with the disclosure of A1, and is therefore entitled to its priority. Therefore A1 cannot be cited against the SOTP section.

    Since the claim's WOTP is the exact union of SOTP and ROTP, which are both allowable, then the claim as a whole is allowable. If the examiner insists on being a jerk, I can file a claim for "SOTP+ROTP" instead of "WOTP", just to make the seam connecting visible, but I'd rather not.

  77. Part 5/5

    The examiner looks at Tufty's reply, racks his brains, and wonders:

    - Must the virtual dotted line lie at the boundary between SOTP and ROTP, or could it be elsewhere?

    - Should the dotted line be apparent in the claim?

    - Was the scope really narrowed? Or am I being bamboozled again?

    - Wait! Aha! I have an objection! What about double patenting (GL G-iv-5.4)?

    The EPC does not deal explicitly with the case of co-pending European applications of the same effective date filed by the same applicant. However, it is an accepted principle in most patent systems that two patents cannot be granted to the same applicant for one invention. [...] It is permissible to allow an applicant to proceed with two applications having the same description where the claims are quite distinct in scope and directed to different inventions. The applicant may, for example, be interested in obtaining a first quicker protection for a preferred embodiment and pursue the general teaching in a divisional application (see G 2/10). However, in the rare case in which there are two or more European applications from the same applicant [...] and the claims of those applications have the same filing or priority date and relate to the same invention, the applicant should be told that he must either amend one or more of the applications in such a manner that the subject-matter of the claims of the applications is not identical, or choose which one of those applications he wishes to proceed to grant.

    The claims of A1 and A2 cover overlapping subject-matter, SOTP, presumably addressing the same problem, and the representative took pains to explain that the A2 contains a dotted line neatly delianating SOTP, at least virtually.

    However, if A1 was withdrawn, like in the Nestec case, then I have nothing left to argue. No risk of overlapping coverage.

    If A1 was still pending or was even granted, then wouldn't be much either I could say, as the EPC is sorely lacking effective provisions prohibiting double patenting. I could try to sound menacing and bark like a nasty little yapper, the exact opposite of Teddy Roosevelt's "talk soft and carry a big stick". Case T1423/07 comes to mind.

    Not very good for job satisfaction. Coming to think about this, I did see cases analogous to this one.

  78. Well Fly I've read your stuff, and I'm very sure I am not the only one. Many thanks for those thoughts: I hope they will encourage Tufty and his friends, to pursue the issue further. I see that as another opportunity to plot for European patent law a course forward that can see off all legitimate criticism from the rest of the world.

  79. In response to the latest comment from Mr Firefly, I think one part of the example is different to how I would usually see it. I would expect application A2 to claim WOTP and also disclose everything that was in A1, i.e. including SOTP, as well as some support for generalising to WOTP. This makes the definition of what has a right to priority (SOTP) and everything else (ROTP) quite clear, without the need to start talking about the Emperor's new clothes (see T 1222/11 for example). Without the disclosure of SOTP in A2, the example becomes one that is open to being made fun of, as Mr Firefly appears to be doing. An applicant who does not include the whole content of an application to which priority is claimed is, in my opinion, not doing things right and risks all sorts of problems with later amendments (which of course have to be based on the application as filed, not the priority application).

  80. In the Memo relating to priority issues considered in the Munich Diplomatic Conference referenced on the K’s law blog, an example was given relating to multiple priorities.

    A first filing described a composition including chlorine and a second priority filing "claiming the use of bromine, iodine or fluorine" so that the total disclosure of both priority filings is the collective "halogen" and the memo considers that an EP application directed to "halogen" has a valid priority claim since each possible member was disclosed in one of the claimed priority filings.

    My argument would be, however, that since the collective term “halogen” is not disclosed in any of the priority filings, there is no valid claim to priority for this broader term. Only claims directed to either “chlorine” or “bromine, iodine or fluorine” would have a valid priority, which goes against what the drafters of the EPC were intending to achieve with the concept of claiming multiple priorities.

  81. Tufty, I was surprised to read that you suspect the Fly of teasing you. The way I read him, he is coming to terms with your "divide the claim, but only in your imagination" approach, and is actually having a hard time finding a flaw in it. Fly?

  82. Perhaps my whiskers are too finely tuned at the moment. I think any flaw comes from the possible misinterpretation based on the pie analogy.

  83. Indeed. The "pie" analogy is good visually, and amusingly drawn, but like all analogies, not quite fitting and therefore a bit dangerous.

  84. Following the news that there will be a follow-up post summarising what's been said, I'd just like to add that it would be nice to clearly distinguish two possibilities, namely entitlement to priority versus avoidance of your own priority (and divisional/parent) applications. But I may well be stating the obvious, if so sorry!

    Suppose you have a first published application A disclosing X+Y together, and a second application B claiming priority from A, claiming X. The two questions are:

    1) Should X in B get the priority date of A because X+Y is in A?

    2) Should X in B lack novelty because X+Y is published in A?

    1 is what Berend Brandsma for example is saying would be unfair (I think), and I'd agree because it allows you to avoid third party's applications as well as your own.

    However, for myself at least a solution that satisfied 2 but not 1 would seem fair to the applicant of A and B, as you'd not knock yourself out merely due to arbitrary circumstances of the prosecution, but also wouldn't get you an after-the-fact way to get past other people's applications (i.e. X gets an earlier priority date even though X alone was never disclosed in the earlier application). I think 2 is the effect that Tufty and others are trying to achieve.

    (Finally, to give an example of an "arbitrary circumstance of prosecution", imagine you have a never-published application disclosing X+A and X+B. You file an application claiming X.

    First scenario, the application is granted, but post-grant X is discovered in independent prior art. You amend to sub-inventions X+A and X+B, all fine as lack of unity is ok after grant, and the first application disclosing X+A and X+B was never published.

    Second scenario, X is discovered in independent prior art before grant. To get both sub-inventions X+A and X+B you need to file divisionals, but then they are both knock each other out, so you don't get either. The difference is purely because of the stage at which the prior art was found, well after all the disclosures and filings (except the divisional of course).

    Another way to prevent this arbitrariness would be the suggestion above of having the claims of an application able to lack novelty over its description, of course ...)

  85. MaxDrei: and is actually having a hard time finding a flaw in it.

    Yup, at least in my pie interpretation of the argument, but please don't quote me on this. It still doesn't feel right.

    I don't see why the exchanges are somewhat sensitive, it's not as if anything trivial like money or careers were at stake. A cross-cultural communications issue?

    The EPO division's decision on the second auxiliary request was on the same claim which was discussed before Arnold J. Point 6.7 of G2/98 is cited to deny priority, and the claim is then found to read upon D1's drawings. (Page navigation should be repaired by July 2nd according to the EPO.) The pattern seems to fit the one outlined in that German book.

    I should withhold eventual further comments for the follow-up promised by the original posters.

  86. Thanks Fly, and for pointing out that Arnold J. and the Division considered the same claim under the same matrix of fact and the same law.

    Yes indeed, let us now wait, with 'bated breath, for the authors' promised posting. I hope we won't have to hold our breath for too long though.

  87. Looking forward to the fresh posting from the original authors, specially as this is one of the Kat's most-read Posts.


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

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

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

Powered by Blogger.