• Court Wields the Knife in SEBATIER

    The High Court recently handed a victory to the manufacturers of SABATIER knives, invalidating U.K. trade mark registrations for marks incorporating SABATIER for knives in Class 8 [Rousselon Freres et Cie v Horwood Homewares Ltd. [2008] EWHC 881 (Ch)]. At the same time, the Court highlighted the dangers lurking for practitioners arguing similarity of word marks and composite marks: the test is not easy, and if the IPO is not guided to the right result at the outset, the trouble and cost of putting it right on appeal can be immense.

    At Daggers Drawn

    Rousselon Freres owned U.K. trade mark registrations for inter alia SABATIER for knives in Class 8 dating from 8 April 1969. Horwood Homewares later registered a JUDGE SABATIER Logo and STELLAR SABATIER Logo for cutlery, knives and kitchenware in Classes 8 and 21, dating from 10 March 2000.
    Rousselon applied to invalidate the Horwood marks on the grounds that they were similar to SABATIER and other SABATIER composite marks, that the goods were identical or similar, and that there was a likelihood of confusion.

    Although the hearing officer accepted that Horwood’s Class 8 goods were identical to those of Rousselon, he was not persuaded that the Class 21 goods were similar because the trade appeared to regard “knives” and “kitchenware” as separate categories. In this, he was influenced by evidence that such goods were displayed in different sections at trade shows.
    Most critically, though, the hearing officer found that the later marks were not similar to SABATIER despite containing that word. He reasoned that the dominant elements of the later marks were the words JUDGE and STELLAR. Although SABATIER too was distinctive, the hearing officer did not regard it as the dominant distinctive element of the later marks. He considered that the average consumer was likely to take greater notice of the JUDGE and STELLAR elements, and that there would therefore be no likelihood of confusion.

    Rousselon appealed to the High Court, arguing that the hearing officer had failed to apply the correct test for determining similarity of the goods in Classes 8 and 21 because he gave excessive weight to the trade categorisation of the goods.

    Rousselon further argued that the hearing officer had wrongly concluded that the marks were not similar because he had failed to apply the ECJ’s guidance in Medion v Thomson Multimedia (Case C-208/06) and OHIM v Shaker di L Laudato & Co. SAS (Case C-334/05 P). The former had been handed down before the original hearing, and the latter before the decision.

    Similarity of Knives and Kitchenware

    The judge disagreed that the hearing officer had made an error of principle on similarity of goods. The hearing officer had been entitled to take account of the way in which the trade categorised particular goods as an indicator of how the average consumer perceived the goods. Rousselon’s evidence had not shown convincingly that knives and kitchenware were, as a rule, sold in close proximity, or that when they were, they would be regarded as similar by the average consumer. Much of the evidence showed kitchen knives being sold through catalogues and lists in which knives appeared near many products, such as pepper mills, chairs and cleaning products, that would not be regarded as similar.

    The judge remarked, “common sense…tells us that a cooker is not similar to a knife. I suppose that evidence could be so strong that one is driven to the conclusion that a cooker is similar to a knife for trade mark purposes, but one is entitled to start with the common sense…and self-evident…proposition that a cooker is not similar to a knife. …In the absence of evidence either way it would be unsurprising to find a hearing officer…reaching the conclusion that knives and cookware are not similar.”

    Although the judge sensed a certain complementarity between knives and cookware in Class 21, the connection was not strong enough to render the hearing officer’s decision fundamentally flawed. In the judge’s view, that connection was even weaker in respect of many other Class 21 goods.

    The judge conceded that had there been evidence to point the other way, the trade categorisation evidence would have lost a great deal of its strength, given that the shows were very large and not necessarily representative of how the average consumer would encounter goods. However, the evidence before the hearing officer was not so strong as to point decisively the other way, and the hearing officer had been entitled to find that Horwood’s Class 21 goods were not similar to those of Rousselon in Class 8.

    Similarity of Marks

    In assessing similarity of marks, the hearing officer had applied the guidance in inter alia Sabel BV v Puma AG [1998] RPC 199, which required him to assess similarity based on overall impressions, bearing in mind distinctive and dominant components.

    The High Court reluctantly disagreed with the hearing officer’s approach. Although showing great deference for the hearing officer’s experience and sympathy for the fact that later relevant decisions were not drawn to his attention before the hearing or decision, the Court nonetheless considered that the hearing officer’s approach was contrary to the ECJ’s decisions in Medion and Shaker, and that, as a result, his decision was vitiated by an error of principle.

    In particular, the judge noted that the test in Medion and Shaker required a finding of a likelihood of confusion if a later composite mark included an earlier word mark in a manner such that the word mark retained an independent distinctive role.

    It was clear in this case that Horwood’s composite marks contained SABATIER. It was also clear that the hearing officer regarded SABATIER as distinctive. The hearing officer had clearly found the decision a difficult one, but in the judge’s view it could not have been difficult if SABATIER had retained no distinctive role in Horwood’s composite marks. Clearly the SABATIER word did retain such a role, and the judge found that the hearing officer had implicitly recognised that.

    Consequently, the guidance in Medion left no choice but to find that there was a likelihood of confusion. The hearing officer should have found accordingly, and the Court reversed his decision.

    Comment

    The judgment in this case was clearly written with care. The judge was cautious not to substitute his own views for those of the hearing officer, except where it was proved that, on the evidence, the hearing officer had made an error of principle in applying the law. The ambivalence of the judge’s finding upholding the hearing officer on similarity of goods suggests that he himself may have found differently if he had been permitted to do so.

    In any event, the main issue in this case was always similarity of marks. There was never any doubt that JUDGE, STELLAR and SABATIER were all distinctive in respect of the Class 8 and 21 goods at issue. Medion and Shaker were clear, however, that where an earlier distinctive mark is included in a later composite mark and retains an independent distinctive role in it, there is a likelihood of confusion. The fact that the hearing officer found otherwise probably owes a great deal to the fact that the parties themselves did not draw his attention to these very pertinent decisions before judgment was handed down.

    Even now, though, Medion and Shaker present difficulties that cannot always be easily resolved. In marks like JUDGE SABATIER and STELLAR SABATIER, the words do not hang together to form a phrase and SABATIER clearly retains a separate distinctive role. However, it may not always be so easy to assess this in other cases.

    The recent decision of the Appointed Person on appeal from the UK-IPO in Inbev SA v Air Parts Europe (O-160-08) demonstrates this. In that case, the earlier mark BECK’S was held not to be similar to SOUTH BECK for alcoholic drinks, on the basis that the geographical qualifier SOUTH had a considerable impact on the meaning and understanding of the mark, making it likely to be understood in a geographical context, regardless whether the average consumer was aware that one meaning of “beck” was “stream.” The fact that certain alcoholic drinks were commonly sold by reference to geographical names made that interpretation more likely. BECK’S, on the other hand, was more likely to be understood in a surnominal sense, and there was therefore no likelihood of confusion.

    In this case, the fact that the later composite mark included BECK rather than BECK’S played an important role in the decision against confusing similarity. However, if the earlier mark had been BECK, then it is difficult to avoid the conclusion following Medion that there must be a likelihood of confusion, even though so many other elements point the other way.

    The ECJ’s “independent distinctive role” test resists clear and consistent interpretation and can lead to results that appear to defy common sense. Until such time as it is clarified, the unpredictability it creates will hamper legal certainty and the early resolution of trade mark disputes. The sooner the ECJ refines and clarifies the Medion approach, the better.