|A single market, perhaps,|
but the EU remains a patch-
work of jurisdictions
"As always I read with interest the IPKat, this time on my favourite subject: cross-border litigation. As I am acting in this case for Solvay, please allow me to give you a more precise factual background and some comments on the opinion of the Advocate General (AG).
Indeed, Solvay sued the Dutch and two Belgian companies for the infringement of their European patent (EP) in all countries in which the EP was valid, claiming that all three companies infringed the EP with the same product in all those countries.
Just after Primus/Roche (in which I acted for Primus) I wrote that a case such as this is different from Primus-Roche, in which each of the Roche entities only infringed in their own country.
However, it is important to realize, as to the facts, the following.
The EP patent in this case was valid in two different versions (A and B). After Honeywell lodged its defence, Solvay realized that in fact the defence (according to Solvay) could be read as an acknowledgement by Honeywell that they infringed version (A) of the EP. So Solvay asked the Court to give, for the countries in which the A version was already granted, a provisional measure for the duration of the main proceedings. This provisional measure did not address the Netherlands because, in that country, the other version of the EP was valid (version B).
The Advocate General comes to two unsurprising conclusions:
1. In the case at hand the cases are connected. It is difficult to see how you can come to a different decision because a different decision would mean that a case would not be connected because the defendants are different. That now is the essence of art. 6(1) of Regulation 44/2001 and such a ruling would therefore have made art. 6(1) meaningless.
2. Art 22(4) is inapplicable in cases where preliminary relief is asked. That is also unsurprising because, in such cases, the court only (in certain countries) expresses a preliminary view but does not establish invalidity inter partes or erga omnes.
The AG has apparently understood this at . However, now we come to the point where the AG has confused issues and has not followed Primus-Roche and GAT-LUK. The AG is describing four different scenarios.
a. Invalidity is invoked in both interim and main proceedings;
b. Invalidity is only invoked in the main proceedings;
c. 1) Invalidity is only invoked in the preliminary proceedings before the main action is started;
c. 2) Invalidity was not raised in the case on the merits but only during the preliminary injunction proceedings.
It is painful to see that the AG states that this seemed to be the situation in the case at hand. That is incorrect. (Of course) Honeywell had also raised the invalidity defence in the case on the merit (as in scenario (a)).
Before we now turn to what the AG says about these different scenarios, I think it is wise to remember that jurisdiction as to preliminary or provisional measure can be found on two grounds
1. Because the court that has been asked to grant the provisional measure has also jurisdiction in the case on the merit or
2. If the court does not have jurisdiction on the merit on the basis of art. 31 of Regulation 44/1002.
The AG stated that, in scenarios (a) and (b), the court must decline jurisdiction on the basis of art. 25 of Regulation 44/2001 and that this follows from GAT-LUK. However, this is clearly incorrect. From GAT-LUK and Primus-Roche it follows clearly that the court does not have jurisdiction with respect to the question of invalidity, but that the court does have jurisdiction with respect to the infringement. See e.g. Primus-Roche at :
“Finally, even assuming that the court seised by the defendant were able to accept jurisdiction on the basis of the criteria laid down by the national court, the consolidation of the patent infringement actions before that court could not prevent at least a partial fragmentation of the patent proceedings, since, as is frequently the case in practice and as is the case in the main proceedings, the validity of the patent would be raised indirectly. That issue, whether it is raised by way of an action or a plea in objection, is a matter of exclusive jurisdiction laid down in Article 16(4) of the Brussels Convention in favour of the courts of the Contracting State in which the deposit or registration has taken place or is deemed to have taken place (GAT, paragraph 31). That exclusive jurisdiction of the courts of the granting State had been confirmed, as regards European patents, by Article Vd of the Protocol annexed to the Brussels Convention.” (emphasis added)Thus speaks the CJEU: even if one would assume that art. 6(1) is applicable (as the Dutch court held), then such consolidation of all these different patent infringement actions before one court could not prevent at least a partial fragmentation. Partial fragmentation means that only a part is fragmented and, as follows clearly from the rest of , that part is the validity part: see: “That issue”.
The AG simply refers to art. 25 for stating that the court would also not have jurisdiction with respect to the infringement as soon as the defendant raises invalidity. However, it is clear from Roche that that is not the consequence of art. 25; moreover, GAT-LUK also gives no support for this view. In GAT-LUK the CJEU says the following about art. 19 (now art. 25 of Regulation 44/2001):
“Article 19 of the Convention, which, in certain language versions, refers to a claim being brought ‘principally’, does not provide further clarity. Apart from the fact that the degree of clarity of the wording of that provision varies according to the particular language version, that provision, as the Commission has observed, does not confer jurisdiction but merely requires the court seised to examine whether it has jurisdiction and in certain cases to declare of its own motion that it has none.”So it is clear art. 25 does not confer jurisdiction but only states that, if you have no jurisdiction, you should decline jurisdiction. However it is clear that with respect to infringement there is jurisdiction. See also very well reasoned Opinion of AG Geelhoed at :
“To begin with, the second proposition (Ho: the proposition that if invalidity is raised also the at the time of the writ of summons existing jurisdiction with respect to infringement falls away) must be discarded. As the Court pointed out in the Duijnstee judgment, Article 16(4) of the Brussels Convention is based on a distinction between disputes concerning the granting and registration of a patent, in which the validity of patents is generally at issue, and disputes concerning patent infringements. The second proposition may be attractive from the standpoint of legal certainty and coherence, but is inconsistent with the explicit choice of the drafters of the convention not to bring all disputes concerning patents and other industrial property rights within the scope of Article 16(4) of the Brussels Convention.”The Opinion of the AG in the present case goes on that point and therefore goes directly against the explicit choice of the drafters of the Convention (now Regulation) and cannot be reconciled with what the CJEU held in Primus-Roche.
The Supreme Court of the Netherlands followed these rules and held that an invalidity defence has no influence on the jurisdiction with respect to infringement but if, in the infringement case, the judge has to deal with validity because it is raised as a defence, the judge has to suspend the proceedings until the competent court has ruled about the invalidity.
This means for scenario (a) that a court is competent to hear the request for preliminary relief because it is competent to hear the main case. The court cannot however deal with the invalidity defence itself: this may mean that, if dealing with that defence is important for the decision in the main case, the court will have to suspend the action until the competent court has resolved the invalidity issue. However, if the court is competent to hear the infringement claims, on the basis of normal rules such as arts. 2 or 6(1) of Regulation 44/2001, the court is also competent to hear the application for interim measures. Thus, in such a case, a patentee will not need to rely on art.31 in order to establish jurisdiction.
Of course (contrary to what the AG seems to think at , in scenario (b) there is not even a request for a preliminary injunction. In that situation the infringement court can only deal with pure non-infringement defences and may be able to decide the case on these grounds (e.g. that the alleged infringing product does not fall under the scope of the claims) or, if it has to take into consideration the invalidity defence, it will have to suspend the proceedings until the court competent with respect to the validity has ruled on validity. Note: in this respect most continental courts only deal with defences as is necessary for deciding the case. They will not always deal with all defences, as is customary in the UK.
The court can (and Dutch courts frequently do) order preliminary measures (scenario a) for the time the case on the merits lasts.
Scenario (c 1)
What the AG states is correct, assuming the court has also jurisdiction on the main case. The fact that the court has jurisdiction in the main case gives the court jurisdiction for provisional measures. We immediately see here the strange consequence of what the AG has said about scenario (a):
If I start preliminary injunction proceedings before the case on the merit is pending then the court has jurisdiction over provisional measures and does not have to turn to art. 31. However, if the preliminary measures have been asked in a pending case where the defendant has invoked the invalidity, jurisdiction is suddenly lost and for provisional measures the court has to turn to art. 31.
This strange difference is absent if, in accordance with GAT-LUK and Primus-Roche, one would hold that the court does not lose jurisdiction with respect to the infringement question simply because a defendant raises the invalidity defence.
Scenario (c 2)
This is a type of scenario which only can be thought of by somebody who has no experience in practice. After GAT-LUK, every defendant in cross-border proceedings on the merit raises the invalidity defence; this is often used as a delaying tactic which can only be effectively countered in obvious cases by a provisional measure. The scenario of the AG where the invalidity defence is not raised in the main proceedings but only in the preliminary injunction case is therefore unrealistic. However, in such a case he considers the court competent to hear the preliminary injunction without the court having to turn to 31. In my view this is correct, but it shows again that the defendant who is clever and has raised the invalidity delaying defence in the main proceedings is better off because he forces the court to turn to art. 31. Again this is the consequence of the AG not applying GAT-LUK and Primus-Roche correctly by denying the court jurisdiction in infringement cases.
Finally it is good to see that the AG refers to the criticism which has come from all sides to the Primus-Roche decision.
It says a great deal if the highest court in the European Union deliberately -- as they did in that case -- cite only a part of art. 2(2) of the European Patent Convention (EPC), leaving out the part which did not fit their judgment and give a clearly wrong reading of articles 64(1) and 69 of the EPC. I say “deliberately” because this “omission” was already in the report for the oral hearing; during the hearing I pointed to that omission and its consequences.
However more serious is the fact that the CJEU in Primus-Roche de facto discriminated against the small economic weaker patentee in favour of the allegedly infringing multinational.
Primus-Roche is a telling story. Roche was allegedly infringing all over Europe and financially weaker Primus had found an economic way of acting in one court against it. After the decision of the CJEU, Primus was forced to give up all other countries outside the Netherlands as it could not afford to litigate in all these jurisdictions. Nothing prevents the CJEU from making it clear in its upcoming judgment that it has come to the conclusion that, in the fact scenario of Primus-Roche too, there is a connection in the sense of art. 6(1). A great court is only great if it is able to recognize its own errors".