First, harmonisation of national trade mark laws in the Union has been “tight”. In this regard, for reasons that make sense in the European political and legal theatre, it is different from the international model of so-called “minimum rights” harmonisation that drove convergence of trade mark norms for the preceding century. Second, the Trade Mark Directive (with the aid of the Court of Justice and national courts) has effected almost total harmonisation of substantive trade mark law, belying the claims of limited harmonisation that are found in the recital to that Directive. Third, the Court of Justice has exhibited a tendency to limit any room for member state manoeuvre, for example, finding that even optional provisions of the Directive must be given a single European meaning. The Court has paid little attention to concerns of subsidiarity. Finally, additional pressure to find single European solutions results from the existence of a Trade Mark Regulation that creates a counterpart unitary regional right (the Community Trade Mark, or CTM) and attendant EU institutions to administer and enforce that right.

This parallel EU-level regime tends to reinforce the idea that trade mark law has been wholly Europeanised, and exerts pressure on the content of Directive-driven national law because the demands of vertical coherence have trumped the potential benefits of regulatory competition between national and regional regimes. In short, there has over the past twenty years been an extensive and deep Europeanisation of trade mark law. But this in turn raises ongoing questions about how best to develop harmonised European principles, because harmonisation is a dynamic lawmaking process and not a static legislative instrument.

In this Chapter, I consider how the process of harmonisation has affected the development of optimal principles of trade mark law, an analysis that is necessarily informed by substantive preferences. It is only by measuring harmonisation by reference to some form of substantive metric that a full assessment of the harmonisation process can be made. I suggest that it important to recognise the important role of national courts in the development of EU trade mark law (in part because of the nature of the field of law), and that paying attention to national legal traditions would assist the Court of Justice in improving the quality of European trade mark law.

Featured here is the beginning of a paper by , Professor of Intellectual Property and Information Technology Law at the University of Oxford. He is also Director of the Oxford Intellectual Property Research Centre, and a Professorial Fellow of St. Peter’s College. The full article can be found .

The post The Europeanisation of Trade Mark Law appeared first on IPOsgoode.

This past Thursday, IP Osgoode hosted Professor Graeme Dinwoodie, Oxford Professor and Director of the Oxford Intellectual Property Research Centre, who gave a lecture on recent developments in keyword advertising cases. Titled “A Comparative Analysis of Liability for Keyword Advertising”, his lecture focused on a detailed analysis of a set of recent keyword advertising cases arising in the US and in the EU.

Dinwoodie first focused on recent developments in American case law. The first case brought to attention was a recent U.S. Second Circuit decision in (April 3, 2009), where it was held that Google’s Adwords sale of �������ܱ𳦴dz�’s trademark as keyword triggers to advertisers may constitute an actionable claim under the Lanham Act.

In analyzing �������ܱ𳦴dz�’s decision, Dinwoodie first looked at a prior Second Circuit decision on pop-up ads, where the court ruled that the defendant did not engage in trademark use under Section 45 of the  (“use in commerce”).  In essence, the Second Circuit court held that the main issue of trademark infringement claims was whether the trademark was used to brand any goods (trademark use requirement), leaving any confusion arguments to the wayside. As such, subsequent cases in the Second Circuit were being decided on the basis of ‘no trademark use’ only. Such a trademark use requirement would effectively immunize any search engines from liability.

Dinwoodie noted that the Second Circuit’s grounds for decision on keywords departed from the rest of the Circuits, where some claimants still prevailed on the confusion argument. However, due to the high threshold for proof of confusion, it resulted in most of the cases in all the Circuits coming to similar judgments, albeit for different reasons. Dinwoodie also addressed some policy concerns arising from the disparate grounds utilized by the Circuits, including the uncertainty of compliance for search engines, the difficulty for trademark owners to police keyword infringement and damages, and the implications for future trademark law.

The Rescuecom decision sought to accept the policy critiques of the trademark use requirement arising out in the 1-800-Contacts decision. Dinwoodie noted that the judgment itself was in a somewhat peculiar form, with a notably short judgment followed by an Appendix. While the Appendix was considered to just be “dictum”, it explained that the 1-800-Contacts decision had just been misunderstood by the district court. The Appendix also included a notation that it had the agreement of the 1-800-Contacts judges.

The immediate significance of the decision is that the Second Circuit has now chosen to focus on confusion as the overall barometer of liability, leaving search engines potentially liable.

Dinwoodie’s second focus in the lecture was on recent developments in the EU.  The Advocate-General (AG) of the European Court of Justice (ECJ) had recently completed a recommendation report regarding three cases referred by the French Supreme Court involving Google’s system and Louis Vuitton, Viaticum, and Tiger (this topic was covered in a previous post here).  From the referral, three questions were raised to the AG.

The first question referred to the AG was whether search engines were liable for the sale and display of keyword advertising under Article 5(1) and 5(2) of the EU Trade Mark Directive.  Article 5(1)(a) provides protection for trademark owners against the use of identical signs in relation to goods or services, while 5(1)(b) provides protection against the use of similar signs for similar goods or services that is likely to cause confusion on the part of the public.  Article 5(2) protects mark holders whose trademark has a reputation, from the use of signs that take unfair advantage of, or is detrimental to, the distinctive character or reputation of that trademark.

With regards to Article 5(1)(b), the AG made the distinction between the sale of ads (including the keyword suggestion tool) and the display of sponsored links. As defined in Article 5(1)(b), the use must have a link with similar goods and services of the mark owner. The AG stated that the sale of ads by Google involved no services sold to the public and did not involve similar goods and services, and thus carried no liability.

The AG continued to state that the display of sponsored links rested on the essential function of the display and that there is no judgment about the trademark until the link is clicked. Thus, based on the trigger alone, there would be no confusion on the part of the public. Here, Dinwoodie noted that the AG’s underlying policy concern was that his analysis should not make Google liable for the creation of organic results, as it would effectively kill the search engine market. In doing so, Dinwoodie questioned whether the AG’s assessment of public confusion should have been factually determined by national courts.

In addressing Article 5(1)(a) and 5(2), the AG stated that the protection offered under these provisions was not absolute and ought to be balanced against other values, including freedom of expression, freedom of commerce, and fundamental rights, among others. The AG found keywords as essential instruments by which information is organized and made accessible to internet users.

The second question referred to the AG was whether search engines were immune under the E-Commerce Directive, absent any actual knowledge. Under Article 14, Google’s organic search results are immune to liability. However, there is a distinction between organic and paid search results. Because Google profits from its AdWords service, it no longer provides “neutral” search results and is barred from the immunity provisions of Article 14.

The third question was whether advertisers were liable to trademark owners for the purchase of keywords. The AG responded with a similar policy concern as in the previous questions, stating that the allowance of advertiser liability would undermine the policies behind search engine immunity. Dinwoodie noted here that the AG had only addressed ‘trigger-only’ keywords (i.e. ads that are triggered by keywords, but  with no actual reuse of the keyword in the ad) and did not address ‘trigger plus ad text’ (i.e. ads that are triggered by keywords and the keyword also appears in the ad itself).

The report of the AG is only a recommendation for the ECJ and is not binding. While the ECJ upholds about 80 per cent of AG recommendations, the most recent recommendations in this field have not been so readily accepted. Dinwoodie does not predict how the ECJ will respond to the AG’s report, but he does provide a couple of issues with the AG’s report that the ECJ may struggle with in deciding whether to accept the recommendations.

One of his concerns was the fact that the AG had made summary judgment with regards to his assessment of public confusion. Dinwoodie thinks it would be more readily received if the AG had deferred the issue to the national courts.

Another concern is that in addressing the extent of trademark protection and the balancing of other values, the ECJ may find that the AG’s list of balancing values may be too broad in scope (e.g. freedom of expression).

The ECJ will not be making its decision until early next year, but with the serious changes to the EU trademark law landscape proposed by the AG, the ECJ may wish to take its time in making this decision.

Powerpoint slides and video coverage of Professor Dinwoodie's talk, including the questions and answer period, may be found on IP Osgoode's .

The post IP Osgoode Speaks: Graeme Dinwoodie “A Comparative Analysis of Liability for Keyword Advertising” appeared first on IPOsgoode.

The well-known marks doctrine provides an exception to the general rule of territoriality and will protect a foreign mark that is well-known but not used in the United States.  Although it has long been assumed that U.S. federal law contained such an exception, this assumption was cast into doubt last year by the Court of Appeals for the Second Circuit.  In ITC Ltd. v. Punchgini, Inc., 482 F.3d 135 (2d Cir. 2007), the Second Circuit held that federal trademark and unfair competition law did not recognize a "well-known marks" exception to the principle of territoriality.  The court acknowledged that the doctrine was mandated by international law but found that the Lanham Act provided no basis on which to recognize it.  The court was unwilling to develop the doctrine as a matter of policy notwithstanding that "a persuasive policy argument can be advanced in [its] support."

Punchgini conflicts directly with the decision of the Court of Appeals for the Ninth Circuit in Grupo Gigante S.A. De CV  v. Dallo & Co., Inc., 391 F.3d 1088 (9^th Cir. 2004), which recognized the well-known marks doctrine under federal law as a matter of trademark policy.  This split prevents uniform national enforcement of trademark law, which is one of the basic purposes of the Lanham Act.  Moreover, failure to protect well-known marks is inconsistent with international obligations of the United States, at a time when U.S. negotiators have been keen to secure equivalent protection for U.S. mark owners abroad.

The complexities of the situation in the United States were only exacerbated by the fact that in Punchgini, the Second Circuit concluded that, although the doctrine did not exist under federal law, the unfair competition law of the State of New 91��ɫ might recognize the doctrine.  Seeking guidance, the court certified questions about state law to the New 91��ɫ Court of Appeals, and the answers from that court effectively offered some protection to well-known marks.  See ITC Limited v. Punchgini, 9 N.Y.3d 467 (2007).  The New 91��ɫ Court of Appeals rephrased the certified questions somewhat, allowing it to characterize the New 91��ɫ case law in its own terms, and held that "New 91��ɫ recognizes common law unfair competition claims [that might protect certain foreign marks against those who compete unfairly in New 91��ɫ], but not the ‘famous' or ‘well-known' marks doctrine."

The New 91��ɫ court held that protecting well-known marks under New 91��ɫ unfair competition law required that the defendants deliberately copied plaintiff's mark and that the relevant consumers in New 91��ɫ "primarily associate" the plaintiff's mark with its goods or services. However, when the case returned to the federal courts, the Court of Appeals for the Second Circuit affirmed the summary dismissal of the plaintiff's claims under New 91��ɫ law.  See ITC Limited v. Punchgini, 518 F.3d 159 (2d Cir. 2008).  The court held that although the plaintiff had shown sufficient evidence of deliberate copying, it had failed to offer proof of the necessary association.  Nor would the Second Circuit allow the plaintiff to rely on the first element of the test (deliberate copying) to prove the second (consumer association).

Last month, however, in Empresa Cubana Del Tabaco v. Culbro Corp., __ F. Supp.2d __ (S.D.N.Y. Nov. 19, 2008), a federal district court in New 91��ɫ granted judgment to a foreign mark owner on a claim brought under New 91��ɫ principles of unfair competition.  A Cuban company successfully argued that a U.S. cigar manufacturer had deliberately copied the COHIBA mark for cigars, and that that mark was primarily associated with the goods of the foreign plaintiff.

It could be argued that, despite the disclaimer by the New 91��ɫ Court of Appeals in Punchgini, the cause of action endorsed by the court under New 91��ɫ law does not radically differ from the protection offered well-known marks by the Grupo Gigante court.  Indeed, the Empresa Cubana decision suggests that in certain respects the New 91��ɫ cause of action may be more liberal than the federal doctrine recognized in Grupo Gigante.  For example, in Grupo Gigante, the Ninth Circuit held that, to receive protection as a well-known mark, the plaintiff's mark must satisfy an "intermediate standard" of renown between secondary meaning and the dilution-based concept of fame.   The New 91��ɫ cause of action requires only secondary meaning.

Ironically, however, in order for the New 91��ɫ claim of the Cuban company to proceed despite the Cuban embargo regulations (which the Second Circuit had previously held barred the Cuban company's well-known mark claims under federal law pre-Punchigini), the court had to be satisfied that the New 91��ɫ claim was not essentially the same as that which had been barred under federal law.  Otherwise, the state law would simply be being used to circumvent the embargo regulations.  The court concluded that the New 91��ɫ claim was not the same: "New 91��ɫ's common law of unfair competition by misappropriation does not provide the foreign party with the right to exclude others from using the mark and does not, like a trademark, reserve a designation for the foreign party's exclusive use."

Uniform national treatment of well-known marks, and U.S. compliance with international obligations, requires Congress to clarify that well-known marks are protected under federal law.  Empresa Cubana only highlights that reliance on New 91��ɫ state law is a wholly inadequate solution theoretically as well as practically.

The post Continuing Uncertainty Over the Protection of Well-Known Marks in the United States appeared first on IPOsgoode.