Lessons from the Trademark Use Debate

• Author: Graeme B. Dinwoodie/Mark D. Janis
• Position: Professor of Law/Professor of Law and H. Blair & Joan V. White Chair in Intellectual Property Law
• Pages: 1704-1721

Professor of Law, Associate Dean, and Director, Program in Intellectual Property Law, Chicago-Kent College of Law; Chair in Intellectual Property Law, Queen Mary College, University of London.

Professor of Law and H. Blair & Joan V. White Chair in Intellectual Property Law, University of Iowa College of Law. Copyright 2007, Graeme B. Dinwoodie and Mark D. Janis.

Page 1704

In their thoughtful article,¹ Professors Dogan and Lemley discard more all-encompassing versions of the trademark use requirement (in which every infringement allegation would be subjected to a threshold inquiry into whether defendant's use constituted a use as a mark). Instead, they seek to delineate and defend a "more surgical form" of trademark use doctrine,²meaning one that is intended only to protect defendants that are employing marks for something other than the branding of their own products or services.³ Thus, as Dogan and Lemley now say, trademark use should not be viewed as "a panacea, a silver bullet, or a wonder theory."⁴ We applaud Dogan and Lemley for seeking to advance the trademark use debate by narrowing it and for attempting to stuff the trademark use genie partly back into its bottle. But we remain unconvinced that the trademark use doctrine will serve the goals of the trademark system, regardless of whether the doctrine can be cabined successfully in accord with the Dogan and Lemley reformulation.

Although our analysis diverges from that of Dogan and Lemley in many respects, we have selected two general points of disagreement on which to focus in this brief Article. First, Dogan and Lemley's latest arguments haven't altered our views as to the correct interpretation of the Lanham Act, as detailed below in Part I. We still aren't persuaded that the language of the Lanham Act imposes a trademark use requirement, even when that requirement is defined "surgically" and sections 32 and 43(a) are read "fluidly," as Dogan and Lemley suggest. Moreover, this more nuanced interpretation still renders section 33(b)(4) redundant and unduly limits appropriate common law development of trademark law.⁵

Second, as we discuss in Part II, we are unmoved by Dogan and Lemley's additional normative arguments for deploying trademark use to shield defendants from even potential liability for various commercial uses of marks, especially in connection with online contextual advertising. Dogan and Lemley would minimize direct liability for search engines by applying the trademark use doctrine but would reintroduce some marginal amount of regulation by leaving the threat of contributory infringement liability. WePage 1705 still believe that this would fail to create sufficient oversight of either the presentation of search results or advertising sales practices.⁶ In addition, we disagree that marginalizing trademark law will best encourage intermediaries to structure their business arrangements in ways that promote reliable information flow to consumers while balancing the rights of trademark owners and their competitors.⁷ In fact, the opposite is true: in the shadow of the threat of potential liability, some search engines have begun to restructure their business practices, undertaking what we perceive are good faith efforts to respect trademark rights. Lastly, we disagree with Dogan and Lemley that offline analogies should necessarily direct the outcomes of trademark disputes over online practices.⁸ In particular, we don't think that the practice of grocery stores placing rival products in close proximity to one another on store shelves tells us much about whether Internet search engines should be liable for keyword sales, and we worry that the rhetorical potency of grocery store metaphors (and the like) will lure courts into assimilating online and offline conduct without carefully assessing the differences in online and offline contexts.

Our disagreements on these points also highlight broader differences about methodological approaches to trademark law, some of which we identified in Confusion Over Use,⁹ others of which we discuss throughout this Article. In particular, our distaste for limiting the potential scope of the Lanham Act reveals our greater willingness to see trademark and unfair competition law as a market regulator. Relatedly, we are more firmly committed to judicial development of both potential liability and potential defenses; Dogan and Lemley want courts to focus only on the latter. Our view is supported both by the theory of the Lanham Act as a common law system operating within a statutory framework and by our faith in the institutional benefits of common law lawmaking at a time of political tumult in intellectual property law.

Indeed, the insights to be gained from this broader methodological debate may prove to be the most useful byproduct of the trademark use controversy. They may provide lessons for trademark law that will be valuable long after the current spate of litigation over online contextual advertising has subsided.

Page 1706

I Trademark Use and the Lanham Act's Ragged Edges

A Abuse of the "Use in Commerce" Definition

As we outlined in Confusion Over Use,¹⁰ trademark use proponents have invoked (1) section 45's definition of "use in commerce" (a phrase that also appears in sections 32 and 43(a)), and (2) the "in connection with" clauses in sections 32 and 43(a), in their efforts to establish Lanham Act grounding for their notions of trademark use. Dogan and Lemley provide a particularly nuanced analysis that only partially endorses these prior interpretations while also exposing numerous problems with them.¹¹ Ultimately, they prove more about the Lanham Act's imperfections than about trademark use.¹²

On the appropriateness of relying on section 45's "use in commerce" definition, Dogan and Lemley's Article yields significant ground. Dogan and Lemley concede that use concepts appearing in different parts of the Lanham Act need not have "complete consistency in meaning,"¹³ that the differences in meaning "turn[] on the context,"¹⁴ and that "the scope of use sufficient to establish a trademark as a source-identifier in the minds of the public is [different] than the scope of use sufficient to tamper with that role and thus commit infringement."¹⁵ Yet Dogan and Lemley attempt to salvage something from the inclusion of the phrase "use in commerce" in the infringementPage 1707 provisions. Under their approach, the "strict definition" in section 45 would be "inapplicable to infringement" determinations.¹⁶ However, courts deciding infringement could take a "more fluid approach" to the language of the section 45 definition in order to give it some relevance in infringement determinations.¹⁷

This selective fidelity to statutory text may validate the very interpretation that Dogan and Lemley repudiate. In a number of keyword-advertising cases, courts grappling with the trademark use doctrine have cited the section 45 definition and have included language in their opinions suggesting that they are accepting uncritically the proposition that the use required to establish rights is the same as that required to infringe rights.¹⁸For example, courts have attempted to analyze whether selling or buying a keyword is an act of "placing" a mark on goods or an act of "displaying" a mark in connection with services.¹⁹ Thus, some courts appear to be leaning towards treating use as a monolithic concept-the very conclusion that Dogan and Lemley themselves now disown.²⁰ This illustrates what mightPage 1708 happen when courts embrace the so-called "fluid" approach to interpreting the infringement provisions.

B The Section 33(b)(4) Defense

Even setting these problems aside, the Dogan and Lemley reformulation for trademark use still doesn't square with section 33(b)(4).²¹Given the fact that section 33(b)(4) expressly preserves a defense for some non-trademark uses,²² interpreting the infringement provisions as impliedly filtering out all non-trademark uses would render the defense superfluous.²³This conflict occurs whether trademark use is defined broadly or surgically; it's no answer to claim, as Dogan and Lemley do, that by defining trademark use restrictively, their trademark use doctrine doesn't "tread into the territory" of the section 33(b)(4) defense.²⁴

C The Role For Common Law Development of Trademark Law

This debate about the relationship between trademark use doctrines and the language of section 33(b)(4) implicates broader concerns about the extent to which the Lanham Act allows for further common law development of trademark doctrines.²⁵ Dogan and Lemley say that the fact that Congress has codified a specific defense (such as the section 33(b)(4) defense) should not preclude courts from developing others²⁶ and that we must be rejecting this proposition (i.e., that we must be arguing that section 33(b) freezes the development of the common law) because we favor interpreting the Lanham Act so as to give some meaning to section 33(b)(4). However, as we pointed out in Confusion Over Use, our contextual approach contemplates that courts will continue to develop defenses as they are called upon to balance confusion-avoidance values against other valuesPage 1709 in new contexts.²⁷ And we believe that the trademark statute provides them plenty of room do so.²⁸

We do recognize that section 33(b) may limit judicial capacity to develop new defenses as to...