Trademark law is in the midst of an identity crisis. The prevailing economic account of the law has come under sustained attack by scholars, who have both challenged its descriptive accuracy and blamed it for many of the expansions of trademark rights in recent decades. The likelihood of confusion test – long the nucleus of infringement analysis – has been roundly condemned as indeterminate, incoherent, and normatively empty. No one seems to agree about why we have trademark law and how best to implement it. Scholars have cast about for explanations of how we got here and ideas of how to get out of this mess.
In this article, Bob Bone tries his hand at both diagnosis and cure. His focus is the much-maligned likelihood of confusion standard. The article (1) explains how we ended up with such a curious test for infringement, (2) critiques the standard, finding it internally inconsistent, normatively deficient, and dangerous, and (3) proposes an alternative that Bone views as more consistent with trademark law’s goals. To do all of this, of course, Bone must endorse some version of those goals. It all adds up to an ambitious undertaking, to put it mildly. And while the article inevitably falls short of fulfilling all of these ambitions, it offers some wonderful insights and enriches the conversation about the values that shape our trademark laws.
After a brief overview of trademark doctrine, the article turns to an engaging account of the history behind the likelihood-of-confusion test. Bone focuses on the Second Circuit in the mid-twentieth century, given the centrality of that circuit’s Polaroid test in the trademark infringement landscape. He contends that the Second Circuit crafted the likelihood-of-confusion test as a sort of compromise between two competing visions of trademark law’s goals. One faction of the Second Circuit had insisted on a showing of either harm or moral culpability as a threshold requirement in trademark cases, while another faction viewed consumer confusion, alone, as an evil to avoid. The Polaroid test for infringement, Bone contends, reflects an amalgam of these two approaches, incorporating harm and bad faith as factors rather than a threshold requirement, and making likelihood of confusion the ultimate question in trademark infringement suits. The result is a rudderless, normatively vacuous standard. I found this portion of the article terrific. It brings to life the intellectual and legal history of the period, and offers a compelling explanation for some of the standard’s most puzzling factors. More generally, it powerfully demonstrates the reality that legal rules result as often from happenstance and patchwork as from thoughtful, considered evolution.
The next chunk of the article exposes the inadequacies of the likelihood-of-confusion standard. Bone contends that the standard “produces bad results, … is doctrinally incoherent, … and is inadequately supported on normative grounds.” He explains and defends each of these points, drawing upon existing commentary and adding some insightful critique. One comes away from this discussion convinced that the likelihood-of-confusion test is badly in need of repair.
Having persuaded the reader that trademark law needs better normative footing and doctrine that reflects its goals, Bone turns in the latter half of the article to his alternative formulation. The article’s essential premise is that trademark law should protect against confusion only when it either causes real harm or results from “morally blameworthy conduct.” Confusion, Bone declares, “is not a problem in itself. People are confused all the time and the law does nothing to lend assistance. Thus, preventing consumer confusion is justified only if the confusion is associated with morally culpable conduct or causes sufficient harm.” Infringement, in other words, can be proven through one of two things: a showing of morally blameworthy acts, or a likelihood of confusion that will cause “sufficient harm.”
The notion that trademark law should focus more on harm is a popular one; Bone generally endorses the idea but proposes a somewhat different approach than other harm-requirement advocates. To me, the more intriguing aspect of his framework is the morality-based alternative. I found this portion of the article thought provoking, although it left me eager for more. In developing his moral arguments – i.e., that “morally blameworthy conduct” deserves condemnation regardless of any proof of harm – Bone assumes that society has an interest in punishing those engaged in deliberate deception, without regard to whether their deception causes any harm. This assumption raised a whole host of questions. How, for example, do we assess whether someone is engaged in purposeful deception? Why should we punish deliberate but non-material deception in the trademark context, when the law does not generally punish those who harmlessly lie? At the other extreme, one might ask why we limit “morally blameworthy conduct” to deliberate deception, and exclude free riding and other acts that at least some people view as morally wrong? In parts of Europe, for example, it’s viewed as morally problematic to take “unfair advantage” of a mark’s reputation. If we base the law strictly on morals, whose morality should dictate the rules? By adopting a non-consequentialist approach to the “moral” branch of trademark infringement, Bone invites questions about the source of his moral compass. Bone touches upon some of these issues, but they could well absorb a full article by themselves.
All in all, this article sheds important new light on the source and the shortcomings of the likelihood of confusion standard. Just as importantly, it opens a fascinating conversation about the appropriate role of “morality” in trademark law. I look forward to the next chapter.