The Fall and Rise of the Italian Torpedo in European Patent Litigation

Matthew R. Jones

©2014. Published in Landslide, Vol. 6, No. 6, July/August 2014, by the American Bar Association. Reproduced with permission. All rights reserved. This information or any portion thereof may not be copied or disseminated in any form or by any means or stored in an electronic database or retrieval system without the express written consent of the American Bar Association or the copyright holder.

As an opening strategy in patent litigation, the “Italian torpedo” is one of the most interesting and creative. It arose out of the peculiarities of the rules governing jurisdiction between European Union member states (and, of course, out of the minds of diligent and creative lawyers), and became one of the most controversial tactics used by alleged patent infringers in Europe. It was also something of a headache for patentees and the practitioners representing them. For many years, most practitioners thought that the Italian courts and the Court of Justice of the European Union (CJEU) had “killed” the Italian torpedo. However, in a 2013 judgment, the Italian Supreme Court revived the Italian torpedo. This article discusses what the Italian torpedo is, how it died, how it caused a little trouble even after its death, how it came back with a vengeance, and what patentees might do to defend against it.

What Is the Italian Torpedo?

The term “Italian torpedo” first appeared in an article by Professor Mario Franzosi in 1997, entitled World-Wide Patent Litigation and the Italian Torpedo.1 To understand the Italian torpedo, one must first know a bit about the rules relating to jurisdiction in the European Union set out in the so-called Brussels regime, specifically the Brussels Regulation.2 Under the Brussels Regulation, parties domiciled in the European Union can (and should) be sued in their country of domicile.3 However, in a tort case, a party can also be sued in the member state where the harm or damage occurred or may occur.4 The Regulation also contains provisions relating to lis alibi pendens (“dispute pending elsewhere”), which provide that in cases involving the same parties and same cause of action, once an action is filed in one court, any other court must stay a later-filed proceeding until the jurisdiction of the first court is established.5 Once the jurisdiction of the first court is established, any other courts must decline jurisdiction.6

One of the basic tools in a potential infringer’s kit is the ability to ask the court to grant declaratory relief that the party’s activities (typically set out in a product and process description) do not or will not infringe a given patent. The fun begins when a potential infringer asks a court in one jurisdiction to declare that the party will not infringe a patent in another jurisdiction. Couple this with the rules on lis alibi pendens, and the (crucial) fact that courts in some jurisdictions are much slower than others, and you might see where this is all heading.

The Italian torpedo thus works like this: a potential infringer starts an action in a jurisdiction with a very slow legal process (Italy, for example); the party asks the court to declare that the party does not infringe the Italian designation of the patent; it also asks the court to declare that the party does not infringe the English (German/Dutch/French/European) designation of the patent. The Italian case now involves the question of infringement of the patents in Italy and the other jurisdictions, and the Italian court may take years to decide whether it has jurisdiction to grant the relief sought. Meanwhile, the courts in the other jurisdictions have to stay any infringement actions relating to the same patents and parties. Congratulations, the potential infringer has blocked the patentee from suing it for infringement in England (Germany/Holland/etc.)!

It gets worse for the patentee: even if the Italian court eventually decides that it does not have jurisdiction to hear all of these declarations of noninfringement with respect to the foreign designations of the patent, the potential infringer can appeal that decision, and such an appeal will have suspensory effect. As such, the courts of other jurisdictions can be blocked from hearing infringement actions until the final stage of appeal has passed, which in Italy can take five or six years. Also, of course, the Italian court may eventually decide that it does have jurisdiction, in which case the other courts will be blocked forever.

Rumors of the Italian Torpedo’s Death Have Been Greatly Exaggerated

Patentees have long seen the Italian torpedo as something of a “dodge” by alleged infringers: a way of blocking (or torpedoing) infringement actions by taking advantage of technicalities and inefficiencies in the system. Perhaps understandably, some courts have agreed. For instance, the District Court of Düsseldorf indicated that it was apt to refuse to stay proceedings where the jurisdiction of the court first attached in (for example) Italy, i.e., where the torpedo action clearly constituted an abuse of process.7

Italian courts themselves have had something to say about the tactic of taking advantage of perceived inefficiency in their court system. In the Windmöller case,8 heard by the Italian Supreme Court in 2003, the court found an interesting way to deny jurisdiction with respect to torpedo actions: it held that a potential infringer could not claim jurisdiction in Italy on the basis that Italy is the place where the harmful event occurred, because asking for a declaration of noninfringement inherently involves contending that a harmful event has not occurred. This judgment was seen by many as the death-knell for the Italian torpedo. However, the judgment left some problems in its wake.

First, what about an action in Italy that only seeks a declaration of noninfringement of the Italian designation of a patent (and therefore is not a “torpedo” at all)? It seems that, under the rule in Windmöller, the Italian courts will not have jurisdiction, at least under the rule that gives jurisdiction to the courts of the place where the harmful event occurred. Unless the patentee happens to be domiciled in Italy, Italian courts may have no jurisdiction to hear actions of noninfringement, even with respect to the Italian patent.

Linked to the above, the Windmöller decision does not help patentees who are domiciled in Italy: they can still be sued for a declaration of noninfringement under the general rule that parties can be sued in their country of domicile, and the potential infringer would still be able to seek declarations with respect to foreign designations of patents as part of that action.

Also, although the Italian Supreme Court held in Windmöller that Italian courts will not have jurisdiction in a typical torpedo action, it will still take those courts some time in each case to decide and rule that they do not have jurisdiction over infringement of foreign patents.

Finally, there is the small matter that lower courts in Italy are not bound by decisions of the Supreme Court. The lower courts tend to follow Supreme Court decisions, particularly if those decisions are well-articulated and argued, but they do not have to.

In 2006, the CJEU published two important decisions on jurisdiction in cross-border patent cases: GAT v. LuK9 and Roche v. Primus.10 At the time, some commentators expected these judgments to be the final nail in the coffin for strategies like the Italian torpedo. However, it wasn’t quite the case. GAT v. LuK ruled that exclusive jurisdiction rests in national courts in cases involving patent validity—it did not relate to actions for declarations of noninfringement where validity is not likely in issue, which is the case in a torpedo action. Roche v. Primus concerned jurisdiction to hear proceedings against different regional companies in the same group structure in one country on the basis that there was an alleged risk of irreconcilable judgments11—it did not relate to a case where a patentee is sued in one country on the basis that that country is where the harmful event occurred,12 which is the basis typically argued in torpedo actions.

Apart from these legal considerations, it did not take long for practitioners and the courts to provide hard evidence that the Italian torpedo was not dead after all. In late December 2006, only a short time after the supposed death of the Italian torpedo at the hands of Windmöller and the CJEU, BlackBerry manufacturer Research in Motion (RIM) launched a torpedo against Visto in the courts of Milan.13 As part of its global battle, RIM sued Visto for revocation of the U.K. designation of Visto’s mobile e-mail patent, then launched a further action in Italy, seeking revocation of the Italian designation of the patent and asking for a declaration of noninfringement with respect to the Italian, Belgian, Dutch, French, German, and Spanish designations of the same patent. Visto retaliated in the United Kingdom by claiming that RIM had committed an abuse of process by commencing the Italian action.14 The case went to the U.K. Court of Appeal, which found in favor of RIM.15 The English appeal court judges pointed out:

Much ingenuity is expended on all this elaborate game playing. Despite the temptation to do otherwise, it is not easy to criticise the parties or their lawyers for this. They have to take the current system as it is and are entitled (and can only be expected) to jockey for what they conceive to be the best position from their or their client’s point of view. . . .

Again a party who fires an Italian torpedo may stand to gain much commercially from it. It would be wrong to say that he is “abusing” the system just because he fires the torpedo or tries to. Things may be different if he oversteps the line (e.g. abuses the process of a court) but he cannot and should not be condemned unless he has gone that far.16

In other words, while torpedoes might bend the rules, they are still within the rules.

However, while the English court seemed to say that torpedo actions were fine in principle, the Italian court was starting to clamp down on them in practice. In the same litigation between RIM and Visto, the Milan court declined jurisdiction over the non-Italian designations of the patent in April 2008,17 16 months after the torpedo had been fired and one month after the U.K. Court of Appeal’s ruling. The Italian decision provided some encouragement that, in jurisdictional matters at least, the Italian courts were willing to reach a decision in a reasonably short time and were taking a rather dim view of torpedo actions.

In fact, in view of Windmöller, the CJEU decisions in GAT v. LuK and Roche v. Primus, the increasingly speedy Italian court (at least for matters of jurisdiction), the reluctance of German courts to stay proceedings in clear cases of abuse, and the fact that relatively few torpedo actions were actually being filed, by the early 2010s the issue of Italian torpedoes in patent litigation had perhaps become something of a historical, or at least academic, curiosity.

The Italian Torpedo Reloaded

Then, in 2013, just when everyone thought that the Italian torpedo had been firmly put in its place (if not exactly killed), the Italian courts changed the landscape once again. In General Hospital v. Asclepion,18 the Italian Supreme Court overturned the earlier decision in Windmöller. The defendant (Asclepion) applied to the Court of Rome for a declaration that it did not infringe the Italian or German designations of the European patent owned by General Hospital. General Hospital challenged jurisdiction on the basis that Asclepion claimed no harmful event had occurred, on the same basis as Windmöller. However, in a dramatic about-turn, the Italian Supreme Court decided that Windmöller should no longer apply, and held that the Court of Rome had jurisdiction to grant declarations of noninfringement with respect to both the Italian and the German designations of the patent.

Whether the Italian Supreme Court was mindful of any of the problems with Windmöller is not clear. However, the Italian court referenced the CJEU decision in Fischer,19 which held that the Brussels Regulation permitting a tort defendant to be sued in a member state wherein the harm or damage occurred must be interpreted as including actions for negative declarations seeking to establish the absence of liability in tort.

So, the Italian Supreme Court has dispensed with Windmöller and the arguably unfortunate consequences of that ruling. However, with respect to the Italian torpedo, we are back to square one. Even the recent trend of the Italian courts to speed up deciding whether they have jurisdiction to hear torpedo actions may not help patentees because now, after General Hospital, the Italian courts may decide that they do have jurisdiction to hear them.

So, what is a patentee to do? Does every patentee face the prospect of having infringement actions blocked by the defendant and not being able to do anything about it? Can patentees work the system to their advantage as well?

How Patentees Might Defend against the Torpedo (and Avoid a Sunk Battleship)

Fortunately, under the rules governing jurisdiction, there is some opportunity for the patentee.

The Italian torpedo is based on the European rules governing lis alibi pendens, the rule requiring later courts to stay their actions and decline jurisdiction in cases involving the same parties and the same cause of action.20 Therefore, one way a patentee might avoid a torpedo is by granting an exclusive license to a third party that could sue the infringer.21 Even if an infringer had previously filed a torpedo action in Italy against the patentee, the new infringement case would not involve the same parties, and therefore the lis alibi pendens rule should not apply.

Another key element of the Italian torpedo is that the Italian court must be first. Therefore, a patentee could simply file its infringement cases first. The patentee has some advantages over the potential infringer because in some countries (notably Germany), it is possible to file an infringement action before the patent is granted. Specifically, an action can be filed after the intended grant of the patent has been announced by the European Patent Office. Of course, this can only be done if the patentee (or patent applicant, in this case) has a patent coming up to grant which it knows (or has good reason to suspect) is infringed by a third party. But while the patentee may sue for infringement before grant, a potential infringer cannot sue for a declaration of noninfringement until after grant. In other words, if done correctly, it should not be possible for a potential infringer to preempt the preemption of a torpedo.


Whether the Italian torpedo will reemerge from the ashes as a commonly used tactic by potential patent infringers remains to be seen. Similarly, at present we can only speculate regarding the defensive (and offensive) strategies that patentees might use to redress the balance. But one thing is comforting: the courts have shown a willingness to recognize the imperfections of the system, and also a welcome reluctance to blame litigants and lawyers for trying take advantage of them. One can hope that, with fierce litigants and talented lawyers on both sides, at the end of the day the outcome will be fair for all concerned.


1. 19 Eur. Intell. Prop. Rev. [EIPR] 382 (1997).

2. Council Regulation 44/2001, of 22 December 2000 on Jurisdiction and the Recognition and Enforcement of Judgments in Civil and Commercial Matters (EC) [hereinafter Brussels Regulation].

3. Id. at art. 2.1.

4. Id. at art. 5.3.

5. Id. at art. 27.1.

6. Id. at art. 27.2.

7. Landgericht [LG] [District Court] Düsseldorf, Dec. 19, 2002, InstGE 3, 8 (Ger.).

8. Cass., 19 dicembre 2003, n. 19550, Giur. it. 2004, 264, BL Macchine Automatiche S.p.a. v. Windmöller & Hölscher KG (It.).

9. Case C-4/03, Gesellschaft für Antriebstechnik mbH & Co. KG v. Lamellen und Kupplungsbau Beteiligungs KG, 2006 E.C.R. I-6509.

10. Case C-539/03, Roche Nederland BV v. Primus, 2006 E.C.R. I-6535.

11. See Brussels Regulation, supra note 2, at art. 6.1.

12. See id. at art. 5.3.

13. See Research in Motion UK Limited v. Visto Corporation, [2007] EWHC (Ch) 900 (Eng.).

14. Id.

15. Research in Motion UK Limited v. Visto Corporation, [2008] EWCA (Civ) 153 (Eng.).

16. Id. ¶¶ 15, 16.

17. Tribunale di Milano, 18 aprile 2008, n. 5111/08, Research in Motion Limited v. Visto Corporation (It.).

18. Cass., sez. un., 10 giugno 2013, n. 14508, General Hospital Corporation (Massachusetts General Hospital) v. Asclepion Laser Technologies GmbH (It.).

19. Case C-133/11, Folien Fischer AG v. Ritrama SpA (Oct. 25, 2012).

20. Brussels Regulation, supra note 2, at art. 27.1.

21. See, e.g., Oberlandesgericht [OLG] [Court of Appeal] Düsseldorf, July 20, 2009, n. I-2 W 35/09 (Ger.).

Matthew R. Jones

Matthew R. Jones is a partner at the firm of EIP in London, where he specializes in multijurisdictional patent litigation in Europe. The author would like to thank Gabriel Cuonzo and his colleagues at the firm of Trevisan & Cuonzo, and Benjamin Grzimek of EIP, for invaluable assistance and useful discussions in the course of preparing this article.