Guest editorial: Confusion surveys and confused judges

The author of this month's guest editorial is Anna Carboni. This is what she writes:

Confusion surveys and confused judges 

Anna Carboni

Should we trust judges to rely on their own experience of the world to decide whether the reasonably observant average consumer is likely to be confused by a particular trade mark or get-up? English judges tell us that we should, except in cases involving particularly unusual goods or services. So a bald male judge will opine on how a ladies' salon haircare brand will be perceived, and a judge who has never owned a pet may decide a case about the get-up of cat food packaging.

The traditional explanation for this state of affairs is that the presence or absence of a likelihood of confusion is ultimately a ‘jury question’, to be determined by the judge using his or (very occasionally) her common sense and experience of the world. But one of the main points of a jury is to represent the common public and ensure that a case is not decided on the basis of one person's possibly idiosyncratic view of the world. Why then, in a case where there is no evidence of actual confusion, is it preferable for a judge to rule on the likelihood of confusion without the assistance of anyone other than the parties themselves, who ‘one can assume’ will have a biased view?

It is not suggested that we should return to the bad old days of allowing parties to select the most-likely-to-be-confused witnesses from the respondents to a market survey questionnaire to come to court and explain the answers that they gave, and inviting the judge to substitute his personal perception with the perceptions of a group of handpicked and well-prepared individuals. Lord Justice Lewison was right to put a stop to that practice in the Interflora v Marks & Spencer cases. But the long-standing and continued approach of English judges making it ultra-difficult for a party to adduce survey evidence on the issue of confusion is an illogical one that deserves further consideration. That is not to say that surveys for use in court should not comply with strict rules. Such rules have been set down in many jurisdictions: surveys must target a statistically significant number of the relevant public; they must be conducted fairly; they must avoid asking leading questions; they must be carried out under careful instruction and supervision by appropriately qualified individuals; and the answers given must be accurately recorded, coded and disclosed.

The cost of a survey exercise that meets these criteria is considerable when one includes the fees of the organisation that carries it out, the individual expert who explains it to the court, and the lawyers preparing the case. But the developing approach in the High Court for England and Wales of holding lengthy preliminary hearings on the question of whether survey evidence will be “of real value” to a particular case and is justified by the cost involved is leading to a disproportionate increase in the overall costs of running a trade mark case, and needs to be stopped.

The tail appears to be wagging the dog. While controlling costs is of course important, adding an interim stage in litigation that racks up a layer of additional costs cannot be the right way to go about it. The costs are undoubtedly increased in part due to the efforts that a legal team feels it has to go to overcome the built-in prejudice against survey evidence on the issue of confusion. Bearing in mind that such evidence is readily accepted as helpful to judges (and even juries) in other jurisdictions such as the United States and many countries in Europe, the current approach leads to huge frustration on the part of litigants.

So let's finish with some questions. First, will Judge A, with no experience of a particular market (even if it is not an unusual one) be assisted by a properly constructed and conducted survey of consumers who do have such experience? Secondly, is it safe to assume that Judge B, who regularly buys the trade mark owner's products and would not be confused by the defendant's lookalike, would not change her mind if survey evidence showed that her response is unusual? And thirdly, will new Judge C, who was in practice as a barrister for 25 years (which involved neither managing cases nor giving cost quotations), have the skills to do a cost benefit analysis in relation to a survey for a case that in all probability he will not even ultimately hear? Answers on a postcard.