It is now several years since the Court of Appeal gave its controversial decision in Unilin v Berry. It found that once an award of damages was made by the English court, and once the avenues of appeal in the UK have been exhausted, the obligation to pay damages is effectively res judicata - a matter already judged- even if the patent is subsequently found to be invalid.

The consequence, which causes heads to shake, is that a party may be obliged to pay very substantial damages to the "owner" of a patent which, under established principles of law, is deemed never to have existed once it has been found invalid, or only to have ever existed in an amended form is subsequently amended by the European Patent Office (EPO).

Later this year the matter may find its way to the Supreme Court in Virgin v Premium and it remains controversial, attracting regular comments on IP blogs.

The fact that this issue gives rise to such strong feelings and heated exchanges is clear evidence that it needs to be resolved once and for all by the Supreme Court.

In the meantime, there are a few misconceptions which need to be addressed, and a few points of law which are not as clear as some people think...

First, let us get one thing quite clear at the outset. There is no correlation between the position of a licensee who has paid royalties and an alleged infringer who has had to pay damages.

The licensee enters into a deal. He chooses not to challenge the patent, but rather to pay for the ease of using the technology "unthreatened" by possible action by the patentee. The alleged infringer chooses the other course - to challenge the patent, saying "come and sue me" to the patentee, on the basis that he does not think the patent applies to his product or process, either because of non-infringement or invalidity or both. If he loses in court, the obligation to pay damages is imposed on him.

If the patent is subsequently found to be invalid, the "infringer" is proved right, and should not, in any just system, be obliged to pay for committing an act which was not, at any time, unlawful. The licensee, who has chosen to nail his flag to the patentee's mast, will have had the benefit of his royalties, and can now also have the benefit of the "infringer's" action.

Secondly, patents are a complication for the law of res judicata, even before any European factor is taken into account. In almost all cases, the principle of res judicata makes perfect sense. Once a matter has been litigated at court, the final judgment should be unassailable unless compelling evidence which was not available beforehand comes to light.

However, in patent law, if a patent is found to be invalid or amended, then the effect of the invalidation or amendment is retrospective. The patent is deemed never to have existed, or always to have existed in its amended form. The problem this creates is obvious. The courts have approached it in different ways.

In Poulton, Coflexip and Unilin a strict compliance with basic principles of res judicata has been applied. However, in Dudgeon v Thompson, Kenrick & Jefferson and PLG v Ardon a totally different approach was adopted. In PLG Aldous J said clearly:

"The practice of this court is that, after amendment, a patentee must start a fresh action rather than seek to enforce any relief he has obtained in a previous action [emphasis added]. The practice is based upon the fact that there has been no judgment on infringement and validity in relation to the amended patent."

This passage illustrates precisely the point made above; that patent actions challenge the usual basis of the principle of res judicata. Those who support Unilin say simply of this passage from Aldous J that "he was wrong". That is hardly an answer, and demonstrates precisely why this is a matter which needs to be resolved by the Supreme Court. It is worth noting that Dudgeon v Thompson is a House of Lords decision.

All of this becomes even more confused when the European angle is taken into account. In 1977 the United Kingdom adopted a new patent system which accepted the involvement of the European Patent Office and the European Courts. One effect of this is that the final arbiter on validity is the European Patent Office. At least as far as findings of invalidity are concerned, the EPO Boards of Appeal "trump" the UK court in an irreversible manner.

Remember that, a couple of years ago, the Court of Appeal accepted a new exception to the rule in Young v The Bristol Aeroplane Company to acknowledge that it could overrule itself if a new line of authority could be shown to have been established in the EPO. This was a clear acknowledgement of the primacy of the EPO and its complete integration into our legal system.

Bearing in mind that the principle of res judicata is based on finality - in other words that all avenues in the UK courts have been exhausted - how can a decision be truly res judicata in the UK courts if there is still recourse to the EPO?

For these purposes the EPO must be seen as being within the UK system. That is how the system works. When a patent case reaches its conclusion in the UK, it will be known whether there are pending opposition proceedings in Munich - that is the way the timetable works. Where it is clear that such proceedings are underway, then no decision of the UK court can be truly final. Accordingly, the principle of res judicata is not in play.

It is no argument to say that infringement and validity are separate issues. That is not how our system works. We have not opted for a bifurcated approach as adopted in Germany.

What then is the rationale for Unilin? A simple, uncomplicated reading of the decision, and the same court's decision to refuse a stay in Virgin v Premium, reveals that the motive of the Court of Appeal in both cases was not upholding the ancient principle of res judicata (which demonstrably does not apply), but rather "commercial convenience" and the avoidance of the "emasculation" of the UK courts by having the effect of their decisions delayed pending an EPO decision.

There may, just, have been some merit in that in the days when EPO proceedings were desperately slow, with no hope of expedition, though it was nonetheless a clear example of "bending the law out of shape" to achieve a commercial result. That is no longer the case. As Lord Justice Jacob himself has repeatedly asserted, if the courts and the parties engage with the EPO, it is possible to accelerate proceedings to avoid unhelpful longueurs.

In fact, in the Virgin case the Court of Appeal made its decision not to stay a damages enquiry at a time when the final decision of the EPO was barely three months away.

Put simply, the "evil" which Unilin sought to address is now no more than a memory, and it is a poor excuse to bend both the law, and perceived justice out of shape. I emphasise "justice" because no lay reader of this exchange would understand a position where a company or individual was obliged to pay very substantial "damages" in relation to an allegation of infringement of a "right" which either never existed at all, or only existed in a form which was not infringed by the alleged "act of infringement". On the contrary, he may be somewhat surprised at the apparent reward for a patentee who pursues, and is willing to accept reward from a patent which is invalid. Surely that is contrary to public policy?

To test this one step further, look at other European Patent Convention jurisdictions. They also accept the primacy of the EPO, but find other practical means to avoid this situation arising. The Unilin type of situation simply cannot arise in other major EPC jurisdictions.

One line adopted by proponents of Unilin is to say "you have to define a time after which it is too late to get your damages back". That actually poses the wrong question, but can simply be answered in this way. In cases where there are pending EPO opposition proceedings, that time is the final expiry of those proceedings. That is when there is finality under the UK legal system, and consequently when the principle of res judicata applies.

The situation where, in subsequent proceedings in the UK courts the patent is revoked as a result of new prior art, can be answered by saying that the party who paid damages had their chance to find that prior art and failed.