As mentioned above, the access approach adopted by the CJEU in copyright cases online has been met with a number of criticisms. From the perspective of private international law and the Brussels I Regulation (Recast) in particular, the access approach seems to contradict both general and specific goals set under the Brussels I Regulation (Recast).  [20] First, predictability and foreseeability principles, as entrenched in the Brussels I Regulation (Recast),  [21] seem to be undermined. The access approach allows a defendant to be sued in any place where the infringing content could be accessed online, that is, in virtually every EU Member State (assuming that no geo-location measures are applied). Thus, a defendant acting in one state cannot predict the place of the suit in advance. The situation gets even worse if a defendant is actually and clearly targeting some countries but the website can still be accessed in other countries (“overspill” effects). In such a situation, the case can still be initiated in the courts of countries that are not targeted.  [22] Secondly, an underlying purpose of the Brussels I Regulation (Recast) and CJEU practice is to curb forum shopping. The access criterion enables the broadest possible forum shopping by the right holder who can now initiate a case in any country with access to the website.  [23]

Thirdly, as the CJEU itself has highlighted, “the rule of special jurisdiction laid down in Article 5(3) of the Regulation is based on the existence of a particularly close connecting factor between the dispute and the courts of the place where the harmful event occurred” (emphasis added).  [24] However, it is very doubtful whether courts of the place where the content can merely be accessed have a “particularly close connection” with the dispute. The state of the court would be one of the many states where content can be accessed. The fact that the content can be accessed does not automatically lead to the fact that it was (or will be) accessed and in this way caused (or will cause) damage in that state. Thus, it is questionable whether the court of a place of mere access would be “the court objectively best placed to determine whether the elements establishing the liability of the person sued are present”.  [25]

Similarly, it is also doubtful whether the access approach helps to attain the goal of “sound administration of justice and the efficacious conduct of proceedings”.  [26] As the Advocate General in the Hejduk case noted, where the work is merely accessible for viewing and downloading in a country, it is impossible to estimate the exact damage that occurs in that country. Then, it is likely that the court would either underestimate the damage in the jurisdiction or overestimate it, and in this way cross the limits of court jurisdiction.  [27] Furthermore, if the same activity is adjudicated by several courts on a territorial basis, this may lead to conflicting outcomes or excessive damages.

From the perspective of a balance of interest approach underlying copyright law, one may question what implication the access approach, as applied when determining the jurisdiction under ex article 5(3) Brussels I (art 7(2) Brussels I Recast), has on the parties involved in the dispute.  [28] The permission for right holders to sue for copyright infringement online in each country from where the content can be accessed gives great advantages to right holders.  [29] This strengthens the possibilities of copyright enforcement online and copyright protection in general. On the other hand, such forum shopping possibilities cause legal insecurity and a lack of foreseeability for users, in particular, online service providers. As mentioned, such a situation is even more unreasonable from the perspective of online service providers if they are targeting only a specific country (or several countries) and access in other countries is a mere “overspill” effect. For instance, an online service provider runs a Lithuanian website in a Lithuanian language, with advertising targeting a Lithuanian audience and members residing essentially in Lithuania. The website operator makes a French work available on the website without authorisation from the right holders. It would be quite unreasonable from the perspective of the online service provider if the French right holder was able to sue this Lithuanian website in France merely because the illegal content is accessible in France.

Thus, the access approach may further discourage online service providers from offering EU-wide services. Actually, it invites them to apply geolocation measures technically restricting the access to certain EU Member States only.  [30] Here one should ask how such practices are likely to affect the EU digital market and whether they are compatible with recent EU policies on copyright and creative markets online. If defendants choose to use geolocation or other measures and geographically limit access to their websites, this would lead to further territorial segmentation of the EU online market. In recent studies  [31] and Commission policy papers  [32] the geographical segmentation of creative markets online has been highlighted as one of the most significant problems and the Commission is expected to take measures to address this single market problem in the near future.  [33] The “access” approach adopted by the CJEU reinforces the territorial nature of copyright on the Internet and in this way contributes to the problem.

Despite these criticisms, there are also arguments that justify the establishment of jurisdiction of a local court based on mere access to the website in that country. First, contrary to what some commentators argue,  [34] the Pinckney and Hejduk decisions are consistent with the previous CJEU rulings on jurisdiction issues online. In the eDate Advertising case,  [35] dealing with the violation of personality rights online, the CJEU applied article ex 5(3) of Brussels I and allowed the case to be brought in any country where the content was accessed (scope of jurisdiction limited to the territory of the state).  [36] In the Wintersteiger case,  [37] which concerned the infringement of national trade mark rights online, the Court seems to have followed the same access approach. However, since trade mark protection is limited only to those countries where the trade mark is registered, the Court logically restricted jurisdiction only to those Member States where the trade mark is registered. Copyright, on the contrary, is automatically protected in all Member States without registration; therefore, the general access criterion was applied.  [38]

As both the Advocate General in the Pinckney case and some commentators have pointed out, in a number of previous cases the CJEU has followed another, targeting, approach.  [39] However, it has been applied in cases analysing the scope of exclusive rights at the level of substantive law,  [40] rather than when determining jurisdiction of the court. Both the Advocate General and the Court highlighted that the jurisdiction and localisation questions should not be mixed up.  [41] Although the Advocate General suggested also applying the targeting approach to the analysis of jurisdiction questions, the CJEU took a different approach. The Court decided that there is no need to apply the targeting approach when analysing the question of jurisdiction; rather, broader grounds of jurisdiction (i.e. any place of access) shall be allowed. This different approach towards the targeting doctrine at the level of jurisdiction and when determining the scope of exclusive right can be explained by the different objectives of PIL and substantive law provisions. The localisation of a protected act will determine not only the question of infringement but also for which territories a copyright licence should be obtained. The criteria to establish jurisdiction may be broader and consider various principles of PIL, such as the sound administration of justice.  [42]

Furthermore, although the access approach allows forum shopping by right holders, this threat is minimal since other legal mechanisms, if applied together with the access approach, will to a large extent discourage right holders from forum shopping. First, as the CJEU highlighted, the courts of the place where the content can merely be accessed would have a limited scope of jurisdiction. That is, they will only be able to grant damages that occurred in the territory of that country. Since damages in a country with mere access would be minimal or equal to zero, suing in that country would not be the most optimal solution for the right holder.  [43] Thus, the right holder that wishes to obtain EU-wide relief is encouraged to approach the court of the state where at least some damage could be established. Alternatively, the right holder could obtain entire cross-border damages in the courts of the defendant’s domicile (art 4(1) of Brussels I Recast; ex art 2(1) Brussels I) or in the courts of the place where the causal activity leading to harm took place (art 7(2) Brussels I Recast; ex art 5(3) Brussels I).

Another way to solve the forum shopping problem caused by the access approach would be to apply the targeting doctrine at the level of substantive law. As will be discussed later, the targeting doctrine has already been extensively applied when localising various uses online and offline. It is suggested here that it should apply when localising copyright infringement online as well.  [44] According to this doctrine, if the courts find that the country where the infringing content was accessible was not targeted by the website, they will dismiss the claim of copyright infringement in that country. That is, even if the access approach allows right holders to sue defendants in the countries where the website can merely be accessed, according to the targeting doctrine, the claim of infringement in such countries would fail at a substantive law level. As a result, right holders would have little incentive to sue infringers in such “empty jurisdictions”.  [45] Meanwhile, only the courts of the targeted Member State will be able to establish infringement in the state and award damages.  [46] Such an approach has been followed inter alia by the Austrian court in the abovementioned Wintersteiger case. As a consequence of the CJEU decision in the Wintersteiger case, the Austrian court accepted jurisdiction in the case of a trade mark infringement online. However, it rejected a finding of trade mark infringement under Austrian law since no targeting or commercial effects had been established in Austria.  [47]

Thus, it is true that allowing jurisdiction on the basis of mere access to the content provides right holders with extensive forum shopping opportunities. However, this problem is reduced since courts of the place of access have only territorially limited jurisdiction; in the absence of targeting and marketing effects in the country, they will not be able to establish damages. Further, if the targeting doctrine applies when allocating the place of copyright infringement at the level of substantive law, the courts of the place of mere access would not be able to establish copyright infringement in their territory at all. The combination of these rules would result in a rather balanced outcome. Although the access approach provides right holders with broad jurisdiction choices, they would be effectively limited since in jurisdictions with mere access, right holders would not be able to get any relief. Meanwhile, users would get more legal security since they can foresee that only the courts of targeted countries would eventually be able to establish infringement and that the damages such courts grant would be limited to the forum territory only. In order to get EU-wide remedies, right holders would need to approach courts in the country of the defendant’s residence.

As an alternative to the access approach, the targeting doctrine has been broadly discussed. It has been applied in a number of the CJEU cases, either at the level of substantive law when localising the activity in a cross-border scenario, or in relation to jurisdiction based on art 17 (ex art 15) of the Brussels I Regulation (Recast).  [48] It was also proposed by the Advocate General in the Pinckney case.  [49] A similar market effect approach has been recommended by the World Intellectual Property Organization (WIPO) when determining infringement of trademarks online.  [50] This market effect approach was followed by some national courts  [51] and suggested by commentators.  [52] The European CLIP Group also suggests limiting unreasonable forum shopping by right holders by applying a targeting approach.  [53] Unsurprisingly, lawyers expected that the targeting doctrine would be the approach that the CJEU adopted in the Pinckney case.  [54]

The underlying idea of the targeting doctrine, as suggested at the level of jurisdiction, is welcomed. It intends to limit the number of courts that could hear the case by excluding jurisdiction of courts in the countries that were not targeted and not substantially affected by the online activity. It means more predictability for online users, including online service providers. By targeting their conduct to some countries only, they can be sure that they may be sued in those countries only, and not in any other country where the website can merely be accessed. This means less possibility of forum shopping for the right holders since the number of potential forums decreases. Also, the courts that would be granted jurisdiction under these rules would have a closer connection to the dispute since the activity would have (actual or potential) effects in that country; they would be better able to estimate the harm done, which would also lead to sounder administration of court proceedings.

From a copyright law perspective, the balance between right holders’ and users’ (potential infringers’) interests in IP enforcement procedures is also better maintained. By applying the targeting doctrine, right holders’ possibilities of enforcement may be reduced (e.g. they may not be able to sue in their home country just because the content is accessible there). Meanwhile, this solution would better serve the legal certainty and foreseeability interests of online service providers. In addition, the balance of interests here can also be adjusted by defining which party would have to prove whether the country at stake is targeted. According to the CLIP proposal, the burden should be on a defendant to prove that a forum country has not been targeted, since the defendant possesses more information about the access and usage of the website.  [55] Such an allocation of the burden of proof would decrease the disadvantages that the targeting doctrine causes to right holders.

On the other hand, a few problems would arise if the targeting doctrine is applied when determining jurisdiction in copyright infringement cases online. One of the most difficult tasks is finding a proper definition of “targeting”. What criteria should be taken into account when determining whether the country was targeted? What criteria are sufficient to establish targeting? What activities by the internet service provider are sufficient to claim that it did not target particular countries? How high should the standard be? For instance, if a Lithuanian news website primarily targeting a Lithuanian audience is also being accessed by a number of Lithuanian immigrants in France, would France be “targeted” by this website? Does the answer change if the online publisher “wanted” it to be read by Lithuanian emigrants abroad? Would the answer differ if a service provider simply “did not care”? Another example could be the case scenario in the Hejduk case: would the German architects’ website under the domain name .de also be targeting German speaking architects in neighbouring Austria? How explicitly should the website include or exclude Austria as a potential audience? Even if the website explicitly excludes the Austrian audience (e.g. by using disclaimers) but the website is still widely accessed by the specialist audience in Austria, would the website still be considered as targeting Austria?

Several guidelines could be useful in determining targeting. The 2001 WIPO Recommendation for the use of signs on the Internet  [56] sets a number of factors that could be mutatis mutandis taken into consideration when determining targeting in copyright cases. Also, the CJEU and national courts have already been applying the targeting doctrine at the level of substantive law, and despite the lack of a clear definition, they came to some reasonable solutions.  [57] The question, however, remains whether such a fact-based analysis should be undertaken when establishing jurisdiction. It seems to be more reasonable at the level of substantive law. Further, let us assume that in order to establish targeting at a jurisdictional level, a more general evaluation of the situation is sufficient. Then, the question emerges as to how the “targeting” at the level of jurisdiction and “targeting” when allocating the conduct online (at the level of substantive law) should be distinguished; the same term would refer to different standards. Overall, the difficulty in defining the term “targeting” speaks in favour of applying it at the level of substantive law rather than when determining jurisdiction of the court.

The other problem is when no countries are specifically targeted by online conduct or, alternatively, all countries are targeted.  [58] This would often happen with websites using a broadly spoken language, such as English. For instance, these might be the websites that have popular content or services in which users from across the EU are interested (such as file-sharing websites), or news websites that target specialist audiences across the EU (e.g. the IPKat blog). Also, one may imagine that under the Pinckney scenario, UK online companies are interested in making any possible sales not only in the UK but also across the EU. In such cases the targeting doctrine would be difficult to apply. Would that mean that those Internet service providers who have not undertaken careful measures to target just specific markets could be sued in any country where the content may be accessed? On the one hand, this might be quite reasonable. If a service is intended for an EU-wide audience, the service provider is exposing itself to all jurisdictions across the EU. On the other hand, such a legal situation discourages online service providers from offering EU-wide services.  [59] In order to avoid risks, they are likely to restrict their online services to several markets and this is not compatible with the EU goal of creating a single digital market for creative content.

Also, the recent national court practice in applying the targeting doctrine when allocating copyright infringements online shows that it is applied by courts quite broadly. Often, very few connecting factors are sufficient to establish targeting in the forum country.  [60] In order to make this doctrine effective in restricting the forum shopping by right holders, the CJEU would need to establish a more rigid standard of what constitutes targeting.

Thus, the targeting approach could be helpful in deciding jurisdiction of the court when websites clearly target one or several countries and right holders sue in countries clearly not targeted by these websites.  [61] This would prevent right holders from suing in a jurisdiction where the website is merely accessible.  [62] However, the application of the targeting doctrine at the level of jurisdiction would often require a careful factual analysis. Such analysis could be better carried out at the level of substantive law when localising the infringing activity.  [63]

The third solution could be subjecting copyright infringement online to the jurisdiction of the court where the infringing conduct originated. As the Advocate General proposed in the Hejduk case, under ex article 5(3) Brussels I (art 7(2) Brussels I Recast), jurisdiction could be granted only to the courts of the state where the “causal event” took place.  [64] This would mean that the place of a harmful event under Article 7(2) of the Brussels I Regulation (Recast) would be only the place from where the defendant acted to initiate the allegedly infringing activity. It would often coincide with the place of residence of the defendant, but this would not necessarily always be the case. As will be discussed later, a similar “country of origin” approach has already been applied at the level of substantive law.  [65]

On the one hand, such a “causal event” rule would guarantee great legal certainty for online service providers. They could be sued only in the country from where they acted, which would normally be the country of their domicile. On the other hand, such a rule, especially if applied at the level of jurisdiction, would lead to unreasonable disadvantages to right holders and may significantly weaken the enforcement of copyright online. Right holders would be required to sue foreign infringers outside their jurisdiction; they would not have a chance to get local remedies in any other country, including their home country, even if damage is felt there. It may be unreasonably difficult for right holders to enforce their rights against websites operating abroad. This would weaken the enforceability of copyright law online. Also, such an approach would encourage forum shopping by online service providers; they may want to provide their services from countries where the court system is not functioning well or where copyright protection is weakest. Further on, this rule would normally lead to the same results as the “defendant’s domicile” rule. This means that an additional ground of jurisdiction based on the place of a harmful event would be largely eliminated. For this reason, applying the “causal event” rule for determining jurisdiction in copyright cases online does not seem to be a viable option.

To conclude, the above discussion has demonstrated that the access approach, as adopted by the CJEU, is not as unreasonable as some commentators argue. It is true that the access approach per se may encourage forum shopping by right holders and could undermine the legal certainty and foreseeability interests of online service providers. However, this danger is diminished to a minimum by two legal mechanisms if they are applied in parallel to the access approach. First, the targeting approach could be applied to allocate the online activity at the level of substantive law, which means that the courts in the countries with mere access to the infringing content will not be able to establish infringement and grant remedies in their territories. Secondly, as the CJEU has clarified, courts in the place of mere access have jurisdiction to grant territorial remedies only. These two rules would discourage right holders from bringing cases in countries with mere access to the illegal content. Furthermore, although the targeting doctrine may seem, at a first glance, to be a more suitable alternative than the accessibility approach, the analysis shows that it has its own problems. Namely, the establishment of targeting would often require a detailed factual analysis, which could be better carried out when allocating the online conduct at the level of substantive law. Finally, the “causal event” rule, as suggested by the Advocate General in the Hejduk case, is the least suitable option for determining jurisdiction in copyright cases online. It would give overly significant advantages to the online users (potential defendants) by unreasonably weakening the position of right holders.

The lex loci protectionis rule is implemented by Article 8(1) of the Rome II Regulation and requires subjecting each IP infringement to the law of the country for which protection is sought.  [67] This applicable law rule is based on the territoriality principle that still underlies EU copyright law. The territoriality principle generally means that copyright law of one state applies only in the territory of that state and does not extend to any acts outside that state’s territory.  [68] The result of this is that the same work is protected under multiple national laws that may grant different scope of protection and enforcement measures. Lex loci protectionis, as an applicable law rule that mirrors the territoriality principle, requests that the infringement of an IP right is adjudicated by the law of the country that grants that right. If the right is infringed in several states, the law of each state would apply with respect to the (part of) infringement occurring in that state.

The territoriality approach and lex loci protectionis rule have traditionally been justified by the need for states to maintain sovereignty over their own national copyright laws, which constitute a part of national cultural policies.  [69] However, with the increasing globalisation of creative markets and especially with the emergence of the Internet, it has become increasingly difficult to justify a strict territoriality approach in copyright.  [70] First, although infringements online have EU-wide (or worldwide) reach, the lex loci protectionis rule does not allow a right holder to acquire, under a single applicable law, EU-wide relief. Instead, even if the case is brought before a court having international jurisdiction over the dispute, the court would have to apply 28 national copyright laws for the same online infringement.  [71] Secondly, for users, such as Internet service providers, this means that they have to comply with the law of each Member State in which their service is available. This encourages service providers to limit their services to certain territories, which is not compatible with the EU single digital market policy.  [72] Therefore, the European Commission has several times raised the question of abandoning or limiting the territoriality principle, at least in an online environment.  [73] Similarly, a number of academic proposals have suggested introducing a single-law approach at least to multi-state copyright disputes.  [74] Therefore, the alternatives to the territoriality approach and lex loci protectionis rule need to be discussed.

The main alternative to the lex loci protectionis rule has traditionally been the country of origin, or lex originis, rule. Generally, it suggests subjecting copyright disputes to the law of the country from where the work originates. That is, instead of the territoriality approach, it promotes a universality approach and suggests that a single law should apply in respect of a single work, regardless of where it has been exploited. Although it has been advocated by some commentators  [75] , it has been accepted in few jurisdictions  [76] , and often only to a limited extent  [77] . The reasons against the introduction of the lex originis rule for copyright infringements online have been discussed by a number of authors  [78] . In short, lex originis is not compatible with the national treatment provision implemented in article 5(2) of the Berne Convention  [79] . As a result of lex originis, works used in the same country would be subject to different rules of different countries (depending on the origin of the work). Furthermore, lex originis does not ensure more legal certainty for users of works, as compared to lex loci protectionis. When commercial users want to use a number of different works (e.g. communicate works over the radio, TV or on the Internet), each of these works might be subject to a different national law, depending on its place of origin. In addition, lex originis prevents states from pursuing their territorial copyright policies in their own territories in respect of works originating abroad. That is, according to lex originis, national laws would apply in respect of works created inside the country, but not in respect of works created in other countries but used inside the forum country  [80] . These and other reasons have led to the general rejection of lex originis as a possible alternative to the lex loci protectionis rule.

On the other hand, European policy makers have been discussing the idea of introducing a similar country of origin approach at the level of substantive law.  [81] Although this would be a solution to the territoriality problem at the level of substantive law rather than private international law, it is worth a short discussion here.

The country of origin doctrine was implemented as a substantive law rule in the EU Cable and Satellite Directive,  [82] where communication to the public by satellite was defined as occurring only in the country where the broadcasting signals were emitted. Outside copyright law, a similar concept was used in the E-commerce directive  [83] where e-service providers were subjected only to the laws of the country of their establishment.  [84] Following this approach, in the case of online communications, the act of making available works online could be defined as taking place only in the country where the defendant is established (E-commerce directive approach) or in the country where the act leading to the making available online originates (Satellite and Cable directive approach).  [85]

I argue here that such an approach would not be the most suitable one. On the one hand, it is true that the country of origin doctrine, if implemented at a substantive law level, would ensure much greater legal certainty for Internet service providers. They would need to comply only with the law of the country from where the activity originates. The single-law approach that this doctrine follows would also enable right holders to get EU-wide relief under a single law, namely, the law of the country from where the infringing activities originate. For instance, in the Hejduk case, after the jurisdiction of the court is established (be it in Austria or elsewhere), the court would have to apply German law with respect to the making available of photos online; German law would apply in determining EU-wide remedies, and there would be no possibility whatsoever of applying Austrian law.  [86]

On the other hand, the definition of the act of making available as taking place only in the country of origin would lead to several problems. First, this may lead to forum shopping by service providers where they establish themselves in (or provide their services from) a country with the lowest copyright protection standards.  [87] It is true that national copyright laws of Member States are harmonised to a certain extent.  [88] However, a number of issues remain non-harmonised (most importantly, copyright exceptions), which means that the scope of protection in different Member States still differs.  [89] Thus, the country of origin rule could be suitable when the full harmonisation of copyright laws has been achieved,  [90] which is unlikely to happen in the near future.  [91] Secondly, when the defendant is situated abroad, right holders would be forced to go to a foreign court (of country of origin) to adjudicate the dispute; that court would apply their own law. Alternatively, if jurisdiction rules allow, a right holder may sue a foreign defendant in the right holder’s own home jurisdiction; however, these courts would need to apply foreign copyright law to establish EU-wide infringement and grant remedies. For instance, following the scenario in the Pinckney case, if the jurisdiction of a French court was established  [92] and the making available is defined as taking place in the UK (a country from where making available originates), French courts would have to apply UK copyright law to establish the EU-wide infringement.  [93]

Thirdly, such a definition of the making available right would contradict the targeting doctrine that has been recently established by the CJEU in defining/localising other intellectual property rights in the online environment. A country of origin approach would replace the targeting doctrine and entirely eliminate territoriality with respect to the making available right online. Fourthly, a country of origin approach would prevent right holders from adjudicating a copyright infringement originating abroad in a home country and getting at least a territorially limited relief. This has been the most popular practice up to now. Right holders may prefer territorially-limited national relief that they could get in their local court, instead of the EU-wide relief which they would need to acquire in a foreign court or by proving foreign law. Fifthly, the law of the country of origin might be in some cases not sufficiently related to the entire dispute. For instance, the activity may originate in one country, but the internet service may target and/or have its main impact in other countries.  [94] In such cases, it may be unreasonable to apply the law of the country of origin and entirely ignore the laws of the countries impacted, especially if the laws of these countries provide for different legal solutions. In addition, the application of a country of origin approach would lead to the situation where players in the same market would be subject to different laws which means the infringement of par condition concurrentium principle.  [95]

As a result, the country of origin approach, even if implemented at the level of substantive law, does not seem to be a viable solution, at least so long as there is no full harmonisation of copyright law across the EU.

Another way to restrict the negative effects of territoriality in online copyright cases is to introduce a special applicable law rule for ubiquitous copyright infringements. The rule suggested in art 3:603 of the CLIP Proposal could serve as an example:  [96]

Article 3:603: Ubiquitous infringement

(1) In disputes concerned with infringement carried out through ubiquitous media such as the Internet, the court may apply the law or the laws of the State or the States having the closest connection with the infringement, if the infringement arguably takes place in every State in which the signals can be received.

(2) In determining which State has the closest connection with the infringement, the court shall take all the relevant factors into account, in particular the following:

(a) the infringer’s habitual residence;

(b) the infringer’s principal place of business;

(c) the place where substantial activities in furthering of the infringement in its entirety have been carried out;

(d) the place where the harm caused by the infringement is substantial in relation to the infringement in its entirety.

(3) Notwithstanding the law applicable pursuant to paragraph 2, any party may prove that the rules applying in a State or States covered by the dispute differ from the law applicable to the dispute in aspects which are essential for the decision. The court shall apply the different national laws unless this leads to inconsistent judgments, in which case the differences shall be taken into account in fashioning the remedy.

The rule means that in the case of ubiquitous (online) infringements of copyright, the court may apply to the entire cross-border dispute the single law that has the closest connection with the dispute. In addition, the CLIP proposal suggests an exception which allows any party to prove that the law of any country covered by the claim provides for a different solution; this has to be taken into account when granting a remedy.

For instance, if in the Pinckney scenario a French right holder wanted to get international relief against the UK online service providers, he would have to go to a competent court with an EU-wide jurisdiction and ask for the application of the single law with the closest connection (which would probably be UK law). The UK defendants could claim, for instance, that their activity is not infringing some of the laws in the EU or that damages in some countries would be lower than damages under UK law. This would have to be taken into account by the court when designing the remedy.  [97] Similar rules have been suggested by the ALI,  [98] Joint Japanese-Korean Proposal  [99] and are currently discussed as a possible international guideline by the Committee on Intellectual Property and Private International Law at the International Law Association.  [100]

This rule has a number of advantages and some disadvantages. The latter, however, could still be eliminated. First, under this ubiquitous infringement rule the right holder gets possibility to request an EU-wide relief, even if this may need to be done in a foreign court and applying a foreign law. At the same time, s/he remains a possibility to request the relief in his/her own country (if there are sufficient grounds of jurisdiction and infringement could be established here); however, this relief would be territorially limited. Second, this special applicable law rule for ubiquitous copyright infringements could be applied only to cases where EU-wide infringements are obvious,  [101] i.e. prima facie infringement cases, but not in cases when it is clear that infringement is possible in some states but potentially not possible in other states.  [102] This may happen when the case concerns the issues that have not been harmonized at the EU level. For instance, in Pinckney case, where the UK companies were selling unauthorised DVDs online, the prima-facie infringement is likely to be established across the EU since making available right has been harmonized at the EU level and no exceptions would arguably apply to the case. However, this would not be so obvious in less straightforward cases. For instance, in the German Thumbnail case  [103] the courts struggled with the question whether showing of pictures in a form of thumbnails in Google search engine would infringe copyright or, rather, whether this use would be covered by any of the copyright exceptions. Since copyright exceptions are not fully harmonized at the EU level, it would be meaningful to limit German court decision in this case to the single territory and allow courts of other Member States to give their say on the matter.

Third, this rule envisages that the law “with the closest connection” would apply. This helps to avoid the potential problem of a country of origin rule, namely, that online service providers decide to run their services from the state with laws most favourable to them. Also, it helps to deal with scenarios when an online service has more connection to the country other than the country where activity originated. However, the flexible “closest connection” factor leads to certain legal uncertainty.  [104] Although the CLIP proposal has a number of factors that should be taken into account when determining the closest connection,  [105] for Internet service providers it may still be difficult to in advance foresee which law has the closest connection and what law they have to comply with. Therefore, in order to ensure more legal certainty, it is advisable to introduce some presumption in the ubiquitous infringement rule. For instance, it could be presumed that the country from where the activity originated or where the defendant is established has the closest connection to the dispute. However, this presumption could be rebutted by proving that other country is more related to the dispute (e.g. when website clearly targets an audience in another country).  [106]

Fourthly, this ubiquitous infringement rule also allows for taking into account the possible divergences among national copyright laws. As mentioned, this rule would apply only to prima facie infringement cases where the issues are essentially harmonised at EU level. However, if certain issues remain divergent (e.g. calculation of damages or type of relief available) the parties may refer to these differences and the court should take these differences into account when determining the remedy.  [107]

Thus, such a special applicable law rule for ubiquitous infringements would have a number of advantages compared with the country of origin doctrine as discussed above. Right holders would get an opportunity in prima facie online copyright infringement cases to get EU-wide remedies under a single applicable law. If right holders do not wish to adjudicate such EU-wide infringement in a foreign court or under a foreign law, they still retain the opportunity to request relief in their home jurisdiction;  [108] this relief would be limited to the forum territory. Further on, legal security for users, including online service providers, would be ensured by allowing such cross-border relief only in the jurisdiction that is most closely related to the dispute and only in cases where prima facie infringement can be established across the EU. Cases that address issues not yet harmonised would be left to the discretion of national courts. Also, it is likely that the courts, applying the ubiquitous infringement rule, would often have to apply forum law which would make court proceedings easier and more efficient. Currently, only the courts of the defendant’s domicile seem to have EU-wide jurisdiction (and, thus, the jurisdiction to apply this ubiquitous infringement rule).  [109] At the same time, it is likely that in most cases the defendant’s domicile will have the closest connection to the dispute.

One of the potential problems with the ubiquitous infringement rule could be that national courts are not used to engaging in analysis and application of such a flexible applicable law rule. The ubiquitous infringement rule would require a more careful discussion in selecting the applicable law, and the weighting of different factors when establishing the closest connection. However, a similarly flexible rule has been introduced in the Rome I Regulation.  [110] Also, some lower instance courts in Europe have already demonstrated some attempts to consider more carefully applicable law issues in copyright.  [111]

Thus, the special applicable law rule for ubiquitous infringements, as suggested in the CLIP and other international proposals, could be quite a reasonable rule that would contribute to the effective and balanced enforcement of EU-wide copyright infringements online. It would not only provide improved enforcement possibilities for right holders but also, if slightly amended, would adequately take into account the legal certainty and foreseeability interests of users. Also, parties would be left with the possibility of claiming infringement under national copyright law and acquiring territorial remedies.