Unusually, I’m lost for words. Not words in general, as you can see, but specific words to describe the phenomenon of self-collision between parent and divisional applications identified in European practice by Malcolm Lawrence and Marc Wilkinson of Avidity IP as the poisonous… I’d better stop there. In the light of UK Trade Mark Registration No. 2612561, I think we need to find a new generic term. “Putrid Priority” is often the real problem but as a term, this doesn’t sound like an improvement. It would be good to make some reference to the overarching phenomenon of self-collision. How about “divisional collision”? Why not – let’s see by the end of the article whether it trips off the tongue well enough as we consider its application in in Nestec S.A. et al v. Dualit Limited et al.
The theory of the poi… sorry, of divisional collision, runs as follows. A patent application claims one or more priorities, and a divisional application is subsequently filed, also claiming one or more priorities. If claims of either application are not entitled to the first priority date, that application may be in trouble. According to the theory, the divisional application may be citable in Europe as prior art for novelty only under Article 54(3) EPC against its parent – or vice versa – in respect of material entitled to that earlier priority date. Under well-established case law, the whole contents of prior art citable under Article 54(3) EPC should be considered in determining whether there is novelty. As the parent (or divisional) may be relying on essentially the same subject matter used to destroy it, the patentee may have a serious problem.
This potential risk has attracted broad attention over the last couple of years, despite a lack of case law. The general issue of self-collision is the subject of active debate. Since the repeal of the Hilmer doctrine in the US, the trilateral patent offices are more closely aligned than before in their treatment of self-collision, but there are still substantial differences between them, as one of the four Tegernsee Experts Group reports published last year indicates. The report notes that while all major patent offices now use the whole contents approach, “there is no harmonization with regard to the effect of conflicting applications, or the mechanism of anti-self-collision”. This divergence is likely to mean that any traps in self-collision practice are likely to find many victims.
Clear support for divisional collision has now been given in one Technical Board of Appeal decision, T1496/11. In this case, claim 1 of the parent application was found not to be entitled to the priority date because this claim contained a generalisation not supported by the priority document, and a divisional application was found to be novelty destroying prior art for this claim. The patent was saved in this case by an auxiliary request limiting the generalisation in claim 1 to the feature disclosed in the priority document.
This phenomenon has now spread from the EPO to the UK national courts. The typically rigorous and careful judgement of Arnold J. in Nestec S.A. et al v. Dualit Limited et al provides an important analysis of infringement under s. 60(2), but almost in passing, the legal theory behind divisional collision finds its first victim in UK litigation. The patent in suit, EP 2103236, claimed priority from another European application, itself published as EP 1495702. Arnold J. found that four claims in the patent in suit were not entitled to the priority date by virtue of a similar generalisation, and that as a result, these claims lacked novelty over EP 1495702 under s. 2(3) of the Patents Act 1977, which has similar effect to Article 54(3) EPC.
So, is this the first case of divisional collision in UK jurisprudence? Strictly speaking, it qualifies: EP 2103236 is divided out from EP 1757212, which claimed priority from EP 1495702, which was abandoned during prosecution (but crucially, only after publication). The collision did not, however, result because of this divisional status – if the claims of EP 2103236 had been in EP 1757212 (not a divisional), the same situation would have existed. Perhaps we do need a better term than “divisional collision”. Personally, I like “mitotic catastrophe”, though I suspect it may not have general appeal – perhaps I’ve spent too much time studying the division of cancer cells recently.
Richard Lawrence 15 May 2013