Linguistic Battles in Trademark Disputes

Linguistic Battles in Trademark Disputes Roger Shuy Linguistic Battles in Trademark Disputes By the same author BUR...

Author: Roger Shuy

233 downloads 1201 Views 692KB Size Report

This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form

DOWNLOAD PDF

Linguistic Battles in Trademark Disputes

Roger Shuy


By the same author BUREAUCRATIC LANGUAGE IN GOVERNMENT AND BUSINESS THE LANGUAGE OF CONFESSION, INTERROGATION, AND DECEPTION LANGUAGE CRIMES


Roger Shuy Distinguished Research Professor of Linguistics, Emeritus Georgetown University

© Roger Shuy 2002 All rights reserved. No reproduction, copy or transmission of this publication may be made without written permission. No paragraph of this publication may be reproduced, copied or transmitted save with written permission or in accordance with the provisions of the Copyright, Designs and Patents Act 1988, or under the terms of any licence permitting limited copying issued by the Copyright Licensing Agency, 90 Tottenham Court Road, London W1T 4LP. Any person who does any unauthorised act in relation to this publication may be liable to criminal prosecution and civil claims for damages. The author has asserted his right to be identified as the author of this work in accordance with the Copyright, Designs and Patents Act 1988. First published 2002 by PALGRAVE MACMILLAN Houndmills, Basingstoke, Hampshire RG21 6XS and 175 Fifth Avenue, New York, N.Y. 10010 Companies and representatives throughout the world PALGRAVE MACMILLAN is the global academic imprint of the Palgrave Macmillan division of St. Martin’s Press, LLC and of Palgrave Macmillan Ltd. Macmillan® is a registered trademark in the United States, United Kingdom and other countries. Palgrave is a registered trademark in the European Union and other countries. ISBN 0–333–99758–1 This book is printed on paper suitable for recycling and made from fully managed and sustained forest sources. A catalogue record for this book is available from the British Library. Library of Congress Cataloging-in-Publication Data Shuy, Roger W. Linguistic battles in trademark disputes/Roger W. Shuy. p. cm. Includes bibliographical references and index. ISBN 0-333-99758-1 (cloth) 1 Trademarks–Law and legislation–United States. 2. Secondary meaning (Trademark law)–United States. 3. Forensic linguistics–United States. I. Title. KF3193 .S55 2002 346.7304¢88–dc21 2002072635 10 11

9 10

8 09

7 08

6 5 4 07 06 05

3 04

2 03

1 02

Printed and bound in Great Britain by Antony Rowe Ltd, Chippenham and Eastbourne

Dedicated to my parents Reo W. Shuy (1904–1978) and Gladys Day Shuy (1906–1979)

This page intentionally left blank

Contents Acknowledgements

ix

Preface

x

1 The Battles over Linguistics and Law

1

2 A Very Brief Introduction to Linguistics for Lawyers

17

3 A Very Brief Introduction to Trademarks for Linguists

28

4 Generic vs. Secondary Meaning: Registry Hotel v. Hospitality Management

46

5 Teaching a Jury about Meaning: Warren v. Prestone

56

6 Sounding Alike and Meaning Alike: ConAgra v. Hormell

69

7 Descriptiveness: Nouns and Modifiers: Woodroast Systems v. Restaurants Unlimited

81

8 The Meaning of a Patronymic Prefix: McDonald’s v. Quality Inns

95

9 Sounding, Looking, and Meaning Different: AMR Pharm v. American Home Products 10 Differences in the Ingredients, Qualities, and Characteristics of the Products: Pyewacket v. Mattel

110

116

11 Going beyond Competing Company and Product Names: AutoNation v. CarMax

125

12 Using Foreign Language Words in Trademarks: Alixandre Furs v. Alexandros Furs

144

13 Disclaiming Dealership Authorization: Matrix Essentials v. F & M Distributors

150

14 Using Linguistic Tools and Thinking in Trademark Disputes vii

159

viii Contents

15 Some Suggestions for Linguists

171

16 Some Suggestions for Attorneys

182

17 Power, Control, and the Ownership of Language

190

References

201

Cases Discussed

204

Index

205

Acknowledgements The rapidly emerging intersection of linguistics and legal issues has many important leaders, all of whom have affected my thinking in many ways. In the US I am particularly indebted to my good friends and colleagues, especially Larry Solan, Ron Butters, Peter Tiersma, Bethany Dumas, and Bruce Fraser with whom I have had the great privilege to exchange ideas and from whom I have learned much. I am particularly indebted to Larry Solan for his helpful comments on an early draft of this book, and to Ron Butters for his suggestions on the chapter that describes a case in which we were expert witnesses on opposite sides. I have also learned much from my colleages in other countries, including especially Malcolm Coulthard and Janet Cotterill in the UK, Hannes Kniffka in Germany, and Weiping Wu in Hong Kong. The attorneys who invited me to help them on the cases described in this book also receive my deepest appreciation: Larry Hefter, Bart McLeay, Bruce Little, Lou Colombo, John Alex, Gilbert Schill, and Richard Knoth all believed in the value of linguistic analysis.

ix

Preface This book has three intended audiences: attorneys who may not be familiar with the usefulness of linguistics in trademark cases, social scientists who may have concerns about how the legal community enforces its own language policy and planning, and linguists who may not be familiar with the way trademark law works and how their specialization can be used in such cases. It does not claim to provide everything that there is to know about trademark law or linguistics, however. Linguists who are asked to work on trademark cases would do well to give consideration to the book cited most in judicial decisions in such matters: J. McCarthy’s Trademarks and Unfair Competition. Attorneys wanting to learn more about the potential usefulness of linguistics in their cases have not had available an equivalent source of information or set of examples about how linguistics can be used in trademark cases. This book attempts to reveal at least some of that potential usefulness to them. Social scientists interested in institutional authority, power, and control of language may find the discussion of how these factors contrast of some interest. It should be admitted that there is no completely satisfactory way to address three different audiences at the same time. One would obviously write differently when the separate readers are linguists, attorneys, or social scientists concerned with the language planning and power of law. Any linguist who tries to show attorneys how linguistics is, or can be, helpful to them cannot, and need not, say everything there is to say about linguistics. Likewise, there is no good way to explain all the intricacies of trademark law in a book intended for an audience of linguists. Some of the issues that the individual fields may find important are likely to be overlooked, oversimplified, or glossed over too quickly, and this book allows for such possibilities from the start. But it was felt that some kind of first crossover step should be made, one that gives some basics to these fields at the same time. For this reason considerable tolerance on the part of all readers is humbly requested. This work grows out of my own experiences in trademark cases over the past two decades. Other linguists and attorneys may have had similar or different experiences, of course, and it is hoped that by making my own work available, others may find it useful and build upon it. x

1 The Battles of Linguistics and Law

Trademark battles conventionally begin with two parties claiming the right to use certain words or expressions. Such cases end either with one party gaining the legal right to use the words and expressions and the other party being prohibited from doing so, or with both parties authorized to have full access to the marks. The authority of law is then granted accordingly. Experts, including linguists, are commonly called upon in such trials to bring the knowledge of their fields to bear on the outcomes of the cases. Such expertise also carries the authority of their respective disciplines. Linguistics is based on the truth about how language works. Such knowledge is especially important in trademark cases, since trademark disputes are largely about language. Differences in where authority comes from and how it should be used bubble beneath the surface. Since knowledge about how language is structured and how it changes is important, even essential, to trademark attorneys, it is useful to address these differences about where authority comes from at the outset. It can be equally helpful for linguists who are asked to work on such cases to have some level of understanding about how the field of law views and uses its authority.

The battle over authority to own words and expressions What happens when one kind of authority conflicts with another? Popular television programs sometimes dramatize the jurisdictional battles that take place when both national and local law enforcement agencies try to capture the same criminals in the same investigation at the same time. Similarly, the authorities of the medical profession sometimes do battle with psychologists and educators over how to deal with 1

2


children said to suffer from such things as attention deficit disorder. Each field has its own identifiable authority, which is not the same as the authority of that of the other field. We tend to have bloodless battles when such conflicts occur. Does the authority of law control all other kinds of authority? On the surface, this would seem to be the case, since certain laws control the behavior of other fields, such as medicine, education, therapy, and business. After all, a society must have regulations and rules if it wants to preserve the orderly behavior of its citizens. Law can be considered prescriptive as it attempts to bring standardization into practice, generally regarded as a good thing.

Trademark law as a form of language planning From the start it should be recognized that trademark law is about the right to monopolize the use of language. This is an important social issue, extending recently into such things as the right of a corporation to keep others from using domain names on the internet. Who can own language? Who has the authority to determine what can be said or written, and who does not? What results when one segment of a speech community tries to change the rest of that community’s use of language? Such questions have led to battles over language planning and policy in many areas of the world. An entire branch of sociolinguistics trains and equips language planners, scholars who analyze and lay out plans to address the social and political problems raised by multilingual situations in such countries as Canada, Indonesia, Gabon, Norway and even by the efforts of some people to establish English as the official language of the United States. Trademark law in many ways can be seen as an effort to create and enforce its own language planning and policy, as the case studies in this book will demonstrate. The legal community claims that the battle over ownership of words or expressions is authorized and empowered by trademark laws. So a restaurant owner named McDonald is prevented from naming his business after himself. Even a hotel company is blocked from using the patronymic prefix Mc- in the name of its proposed hotel chain, McSleep, since McDonald’s was determined to have the sole legal right to that prefix, Mc- (see chapter 8). A man named A.R. Mani is challenged by the fashion designer, Armani, for his longstanding use of “armani” on his web-site. Some corporations believe that they own sole rights to such common words as “hospitality management,” “long life,” and “woodroasted.” Other companies

The Battles of Linguistics and Law 3

oppose them, denying that they, or anyone else, have the right to own these words. One kind of language planning battle exists within the field of trademark law itself. Perhaps unbeknownst to most people, there are scholars who study and work in this area of social science called language planning, somewhat like the more familiar area of urban planning. Language planners assess situations where more than one language is used and plot out future possibilities and optimal solutions. One of the major principles of language planning, a future-oriented intervention that tries to influence language use, is that its decisions and implementation are motivated by nonlinguistic variables (Garvin 1973: 24), usually sociopolitical ones (Daoust 1997). Whether or not they are planned, changes in language policy are frequently accomplished by groups bringing pressure on governments to advance the causes of their own speech communities. One example of this is Israel’s Hebrew Academy, established in 1953, and there are also many others. In the case of trademark law, the motivation for determining language usage appears to be less political than a genuine effort to bring order to a potentially chaotic business practice. It is doubtful, in fact, that practitioners in the field of trademark law are even aware that they are engaging in what linguists call language planning and policy issues. Such unawareness also abounds in the de facto, tacit language policies in many democratic countries. For example, there is no legislation mandating English as the official language of the United States, yet for all intents and purposes that is clearly the case. In multilingual nations the goals of language planning are commonly to achieve a peaceful pluralism (as with Bahasa Indonesian), to gain the assimilation of certain speakers (as with Russian), to yield a kind of purism that eliminates deviant varieties of language (as in the ban on speaking Catalan in Spain during the Franco years), to help more than one language attain official status (as with French in Canada), to internationalize languages to facilitate sociocultural, economic, and political communication, or to enable a country to get greater access to science, technology, and other world issues (as with the teaching of English in many African countries). In such cases, emotions run high and physical wars are sometimes fought over rights to language use. The language planning of trademark law can claim only one of the goals noted above: the attempt to create a purism that tries to eliminate what it considers to be the deviant varieties of language, a goal that is consistent with law’s innate need to be prescriptive. Language planning and policy throughout the world deal with one

4


issue that language planning and policy of trademark law seem to ignore – language loyalty. Once a language policy is developed, or even legislated, it is often accompanied by negative reactions from those who feel oppressed by it. Language is a central part of a user’s sense of identity. This is not to say that the public is currently ready to rebel against the language policies created by trademark laws. But there may well be a sleeping giant out there somewhere, eager to react against the use of laws telling it what it can and cannot say and write. Dictionary makers have already experienced dissatisfaction, and the recent disputes over who can own domain names on the Web may heat up the argument even more. People tend to fight for their rights once they realize they have been restricted or challenged. A comparison of law’s authority with the many language issues faced by multilingual societies may be instructive. A speech community is not defined as one in which everyone uses exactly the same linguistic forms, but rather as one in which social forces dictate the acceptable and agreed-upon range of language use (Clyne 1997). When social prestige and power are unevenly distributed in a society, each separate language, or form of it, is assigned a different social role. The “high” variety monopolizes the official use and public functions. The “low” variety is less valued and is reserved for purportedly less important, private domains of use (Ferguson 1959). Here the association with trademark law is obvious, since law seems to be deciding just who the “highs” and “lows” of society really are. But even if agreement is reached that law represents the “high” language and the rest of the population represents the “low” users, problems remain. For one thing, trademark law demands that the “low” language users be prohibited from using the “low” forms. This goes far beyond the definitions and use of “high” and “low” languages in multilingual settings, where both versions are allowed to coexist, albeit with different social evaluations. Using socially stigmatized grammar or uttering politically incorrect words may identify a person as “low,” but such action is not legally prohibited and does not normally lead to lawsuits or fines. Secondly, law self-identifies itself as the “high” segment of society. When this happens, a negative reaction is apt to follow. For example, for many years the medical profession was arguably considered a very “high” segment of society. In recent years we have seen challenges to this status. Political leaders often suffer a similar fate. Those in power run a constant risk of an insurrection. This book attempts to help the fields of law and linguistics work


together in harmony, despite different views about authority, so we begin with their different underlying assumptions about authority and try to explain the differences in the context of trademark cases. Most of this book is about the ways linguistics can help attorneys with their cases, but their underlying differences about what authority means and where it comes from are constantly present. Understanding each other and working from the subtext of each other’s premises can make the marriage of the two fields work efficiently. This chapter deals with how authority is conceived by law, linguistics, the trial attorney, dictionaries, and money, each of which represents a potential battle of its own kind.

The battle over the authority of law One question addressed in this book is whether or not the legal community has the authority and right to consider itself the “high” segment of society that has control over the non-legal community, which must then be considered the “low” segment in matters of language policy and use. Associated with this question is how to arbitrate between the authority of law and the authority that linguistics derives from the natural processes of language structure and change, especially when these different views of authority conflict with each other or, for that matter, even when they go to battle together, on the same side of a trademark dispute. Since trademark battles are largely about language, this book’s goal is to assist attorneys and linguists in their combined efforts to address specific problems. At the same time, it is necessary to recognize the potential conflicts that can emerge in this marriage of disciplines. It is clear that a society needs to depend on laws to regulate certain types of actions and behaviors. This book is not against laws, or even against trademark laws per se. It is more an effort to stimulate further discussion about what appears to be happening in societies that traditionally assign a high value to the belief that certain language rights belong to all users of that language and not just to a single segment of it. Laws protecting people from physical aggression or pain are obviously needed, as are laws that punish those who steal, commit fraud, defame, and commit other acts of selfishness and greed. When the law attempts to restrict or control what a citizen can say or write, however, things can get very dicey. Perhaps the battle over authority grows out of two very different sets of attitudes, values, and beliefs of the fields involved: law and lin-

6


guistics. Law is, by definition, prescriptive and autocratically final, asserting that once a law is decided, everyone is subject to it. In contrast, linguistics is, by definition, descriptive and democratic, asserting that acceptable language behavior is determined by the way it is used by the people who use it. These two different guiding principles sometimes face each other, head-on, when language issues are debated in the legal setting. The battle over whether or not a language should, like law, be prescriptive is evident in popular debates about language as well. In many ways the linguistic battle over trademarks fits into a long tradition of prescriptivism and authoritarianism in language use. In the eighteenth century, Jonathan Swift recommended that England establish an official authority, like the Academies of France and Italy, to serve as the final arbiter about language variation. In the US, President John Adams suggested a language academy, “instituted by the authority of Congress, for correcting, improving, and fixing the English language which would strike all the World with Admiration and Great Britain with Envy” (cited in Heath 1992: 27). Needless to say, these proposals for prescriptive authority never happened in quite that way, although an ideology of standardization did emerge. Such an ideology’s chief characteristic was the suppression of optional variation at all levels of language (Milroy and Milroy 1999). Along with this prohibition came an effort to prevent language change, to establish a single, correct way of using the language, and, in the process, to define those who varied from such a standard as deviant, stupid, ignorant, or even immoral. But linguistic research continues to show that those who vary from the norms of such ideology are indeed not deviant, stupid, ignorant or immoral (Labov 1970). The notion of establishing an academy having the sole authority to prescribe language use has never worked in the US or in many other countries. Even the ideology of standardization has many problems, as any specialist in language change will attest. Sometimes language change that is ultimately accepted comes from the so-called “low” elements of a society, the poor, the minority, or even the imprisoned. The “high” elements simply do not have the power to stop it. Milroy and Milroy consider the “highs’ ” attempt to control the “lows” as a social process of discrimination based on language traits, noting that evidence from newspaper reports, television programs, classrooms, and courtrooms support such legitimized discrimination, enabling the “high” segments of society to be gatekeepers, restricting access to goods and services and affect people’s lives in many domains (1999: 152).


Trademark law practice is not the only area in which the legal profession challenges the language authority of citizens outside the legal profession. The very culture and structure of court appearances by witnesses, including expert witnesses, give evidence of similar prescriptive inequality about how language can be used. For example, witnesses are not permitted to introduce new topics. They are constrained to answer only the specific questions asked. Yes/no questions must be answered with a “yes” or a “no,” despite the fact that many such questions are impossible to answer in that way. Nor are witnesses allowed to ask questions (with the exception of requests for clarification); only attorneys have the authority to exercise that power. Then there is the unfamiliar structure and culture of the courtroom experience itself, in which most witnesses feel that they are in a foreign country. Such practices may well enhance the efficiency and authority of the trial, but they are new, confusing, and difficult for the witness. This is not to say that all such legal language policies must be abandoned, but it does point out how they can be problematic for the average citizen (Conley and O’Barr 1998; Tiersma 1999). In terms of openness of information and communication, the field of law sometimes also takes it upon itself the authority of preventing citizens from knowing details of the outcomes of trials. Chapters 7, 9, and 12 describe cases that ended with sealed decisions, ones in which only the litigating parties know exactly how the case was settled. Such a closed system may have important economic consequences for possible future litigation, but it runs counter to the principles of most democratic societies, and in the sciences, including linguistics, it would never be acceptable to withhold information about procedures or conclusions.

The battle over the authority of linguists The authority of linguistic testimony in a trademark case rests on the scientific facts about how language works and how it changes. These facts grow out of years of research and theory-building about a given language’s system of speech sounds, grammar, and meaning. These facts are relevant to language use in any context, including the legal arena and, in this case, trademark disputes. In recent years linguists have been actively participating in the broader intersections of language and law and have published books on various aspects of this intersection, including the use of tape recordings in criminal cases (Shuy 1993; 1996), courtroom language (Berk-Seligson 1990; Matoesian 1993; Stygall 1994; Conley and O’Barr 1998), police

8


interrogation (Shuy 1998), the language of judges (Solan 1993; Philips 1998), legal language (Tiersma 1999; Mellinkoff 1963), voice identification (Baldwin and French 1990; Hollien 1990), authorship identification of written messages (McMenamin 1993), aircraft communication (Cushing 1994), and bureaucratic writing (Shuy 1998). There is, at the same time, a growing number of scholarly articles on these and other topics that intersect linguistics and legal issues as well, many published in the journal Forensic Linguistics, and many in books that contain collected articles on language and law (Levi and Walker 1990; Rieber and Stewart 1990; Gibbons 1994). A brief introduction to forensic linguistics can be found in The Handbook of Linguistics (Aranoff and Rees-Miller 2001). Although these books and papers cover a great deal of what linguists have to say about the relationship of their field to the concerns of law, there is still much to be done. For example, at present there are no books that apply linguistics to many of the major areas of civil law such as defamation, product liability, fraud, contract disputes, and copyright infringement. Nor has there been a single book dealing with at least one linguist’s approach to trademark cases, a gap which this book attempts to address. When one speaks about the kind of linguistics most helpful to a trademark attorney, one must first introduce the types of specializations that are most likely to be appropriate for a trademark case. There are at least three: forensic linguistics, applied linguistics, and the so-called hyphenated linguistics. Forensic linguistics Although linguists who work with data and issues in the legal context are sometimes referred to as “forensic linguists,” I have tended to steer away from labeling my work in this way. I prefer to think of myself as a linguist who works on the issues of a specific domain of life, in this case, civil and criminal law. As such, the term “applied linguistics” seems more appropriate. A person who does such work is, first and foremost, a linguist and should be referred to in this manner, especially in the courtroom. Linguists working in the context of medical communication are not, as far as I know, called “medicolinguists.” The same can be said about the work of linguists as they address language issues in the field of law. There is little reason for the data on which a linguist works to have the right to name that work. This is not to say that the work of linguists in law cases is not, in some ways, forensic. Common dictionary definitions of “forensic”


include the fact that it means “suitable to courts of judicature or to public discussion for debate” and “relating to or dealing with the application of scientific knowledge to legal problems.” Certainly, the analysis of linguists in law cases is suitable for public discussion and debate. Their analysis is certainly the application of scientific knowledge to legal problems, as long as “legal problems” means the language data used in a case. But it is the attorneys who determine when and how this analysis is used in cases of their own construction. The expert’s application of linguistics to the language data in the case is made by the linguist, not by the attorney. The application of the linguist’s findings in the ultimate issues of the case is made by the attorney, not the linguist. These appear to be more separate tasks than we find in the mutual knowledge and directionality required of the “hyphenated” fields of psycholinguistics, sociolinguistics, or neurolinguistics. In the legal setting, the separate work of linguistics is applied by attorneys in their separate work of lawyering. Unlike psycholinguistics, where it is necessary to know a great deal about both psychology and linguistics, the intersection of linguistics and law can be and is advantageously carried out separately and independently by both fields of expertise. Whereas both elements of the hyphenation in the “hyphenated” fields are sciences, the application to law is one of a relatively neutral, objective science to the area of legal representation and advocacy, a quite different thing. Applied linguistics Applied linguistics has been a recognized and respected part of linguistics proper for decades. Applied linguists are knowledgeable about the essential tool areas of linguistic theory, including phonetics, phonology, morphology, syntax, semantics, pragmatics, discourse analysis, language change, and language variation. They apply this knowledge to the specific contexts in which they work, including education, medical discourse, diplomatic discourse, advertising, therapy, and law. It is not surprising that general introductions to linguistics include applied linguistics as a major component of the field’s definition. A recent example is the Handbook of Linguistics (Aranoff and Rees-Miller 2001), in which seven of the 32 chapters are devoted to applied linguistics topics. Hyphenated linguistics In the so-called hyphenated areas, such as psycholinguistics, sociolinguistics, and neurolinguistics, these labels seem appropriate. These areas of linguistics have developed over time as a way of marking off the territory covered by those who actively engage in such fields. There is jus-

10 Linguistic Battles in Trademark Disputes

tification for such marking, especially since such specialists indeed need to have extensive knowledge of both academic areas and are required to have experience using the kinds of data commonly studied in those different fields. Psycholinguists are either linguists with more than a passive knowledge of psychology, or psychologists with more than a passive knowledge of linguistics. Sociolinguists need to have knowledge of both linguistics and the social sciences. Neurolinguists need to know a great deal about both neurology and linguistics in order to do their work. There is reason to question whether or not the same needs to be true of linguists who apply their knowledge to the arena of legal issues. In my experience, at least, attorneys want outside help in resolving their cases. They require experts in other fields to provide a different expertise for them. One does not have to know the law intimately to do this. Lawyers need a linguist, not necessarily another person well versed in law. This is not to say that there are no linguists who are knowledgeable about law. In fact, some have degrees in both law and linguistics. This may give them an edge in some ways, but it is certainly not a requirement. I point out these distinctions more for the attorneys who may read this book than for the linguists. If an attorney should happen to call a university linguistics department and ask to speak to a forensic linguist, the question may be met with surprise. There are a few universities now offering courses relating linguistics to law, but they are not necessarily called “forensic linguistics,” possibly for the reasons given above. If attorneys calling a linguistics department ask to speak to an applied linguist, they may be shocked to be connected to a specialist in teaching English as a second language or an expert in language testing, for most applied linguists do exactly these things, not work connected specifically with legal cases. These are not the only designations used by linguists. Some call themselves “theoretical linguists,” indicating that their research and concerns are about how language works, but not on how such principles can be applied to real world problems. Chapter 8 describes what can happen when a linguistic expert defines himself as a theoretical linguist. Some linguists refer to themselves as “cognitive linguists,” indicating that their concerns are for the workings of the mind in organizing, processing, and conveying information. Cognitive linguistics adopts points of view from various disciplines interested in cognition, including psychology, philosophy, and computer simulation. Chapter 6 describes a case in which a linguist described himself in this way.


Whether called forensic, applied, hyphenated, theoretical, or cognitive, all linguistic work is authorized by accepted linguistic theory and research. As such, it is often technical and complex, making it difficult for non-linguists to absorb easily. For example, in several cases described here an effort was made to make use of a type of phonological analysis called “distinctive feature analysis” (Chomsky and Halle 1964). This analysis is helpful in pointing out the similarity or difference in the trademark names in dispute. It allows linguists to analyze each sound in terms of its components, permitting more detailed, scientific comparisons than would otherwise be possible. Most people know that speech sounds are either vowels or consonants, and some even know that certain speech sounds can be either voiced or voiceless. Apart from linguists, however, few know that there also can be contrasts between other acoustic characteristics, such as nasality (air passing through the nasal passage), stridency (the degree of harshness of those consonants represented by the spelled letters f, v, s, z, ch, j, sh, and zh), continuation (sounds that can be continued as long as the speaker has breath to continue them), and a few other such features. Technical? Definitely. Many attorneys consider distinctive feature analysis too technical to be effective in court. The same can be said for technical descriptions of syntactic or semantic differences in trademark cases. At trial, the precise presentation of linguistic theory is sometimes replaced by a simpler, less technical version of it, one that the jury or the judge may be more likely to understand. The trick for the linguist is to keep as much of the truth and authority of the theory as is palatable and appropriate in the applied situation. No matter which specialization one speaks of, linguistics offers the truth about how language works and changes. Not surprisingly, most people, including attorneys, don’t know this truth or the basis behind it. The cases described in this book are intended to demonstrate ways in which the issue of linguistic knowledge is instanced.

The battle over the authority of the trial attorney In most of my experiences in trademark cases, the attorneys have been intelligent, professional, and efficient. They have listened to the analyses I presented and have worked in concert with me to make my analyses effective at trial. They have genuinely understood that the role of their linguistic consultants is to help them learn how language works in ways that will assist their cases. But the pecking order is pretty clear.


The attorney is in charge of the case; the linguist’s role is to use his or her linguistic authority to help the attorney make the best case possible, but without compromising the knowledge and authority of their field. It should be possible for the linguist to be on either side of the case and still come up with the same analysis. Advocacy is the attorney’s role, not the linguist’s. Mutual recognition of the two different authorities and roles is crucial any time a linguist works with an attorney. Compromises on either part can lead to serious problems. Chapter 9 describes how the attorneys essentially left me in the dark about how the case was going, what the reasons were for stopping my involvement, what other legal issues got in the way of the trial in which I was geared up to testify, and even how the case ended. Although it is undeniable that it is attorneys who are authorized to run the show, it can be disrespectful, if not inefficient and counterproductive, for this authority to be so strong that those who work with them are easily ignored or forgotten. Experts also live in a world of schedules and priorities. Lack of communication is one of the major causes of breakdown in many areas of life, including marriage and work. Attorneys who work side by side with experts in a case, and are married to them in their combined efforts, do best when there is a strong sense of partnership rather than a top-down authority hierarchy. Chapter 10 describes a different abuse of the attorney’s authority, where the lawyers for whom I had prepared a draft report ignored much of what it said and informed the court in writing about what I would be “expected to say” at trial. Unfortunately, these expectations included several conclusions that I had never agreed to, either in my written report or in oral discussions. Not surprisingly, what the lawyers “expected” me to testify about was consistent with the way they may have wished I would testify, but it was clearly inconsistent with what my analysis showed. It’s not possible to know whether this was the result of a misunderstanding of what I had written and said (I find this difficult to believe), or whether the lawyers felt that they could eventually convince me to testify the way they hoped I would. Then again, it may have been simply carelessness on their part. Regardless of the reason, it seems authoritarian and presumptuous not to have informed me or shown me their document before submitting it to the Court. The battle over the authority of dictionaries It should be pointed out that a turf war between the authority and language policy of law and the authority and language policy of lexicography plays a role in some of the cases described here (see, for example,


chapter 4). Does the field of law have the authority to prevent lexicographers from including the generic uses of trademarked words as entries in their dictionaries? Should dictionaries that scientifically gather their data of actual usage and build their definitions from such files be overridden by the language policy of trademark law? Who owns the language anyway – lawyers or the people who use it everyday? As Landau observes, lawyers and dictionary makers are in a constant conflict that grows out of their respective differences in task and goal (1989: 298–302). Oddly enough, trademark cases sometimes admit the authority of dictionaries, allegedly to resolve issues of meaning and pronunciation. Yet when the authority of dictionaries comes face to face with the authority of law, it is the latter that seems to prevail. If one examines current, commonly used, and respected dictionaries, one can find evidence that lawyers actually control language policy. The most recent editions of Merriam Webster’s Collegiate Dictionary and the American Heritage Dictionary, for example, contain the entries “Kleenex” and “Band-aid,” both capitalized and defined only as trademarked products. Although the files of these dictionaries may contain many generic uses of these words, such information does not occur in their entries. Recent immigrants who want to know the meaning or spelling of certain words they have heard will learn only that these words are owned exclusively by a specific commercial enterprise. There is no accompanying citation of the generic uses of these entries. The issue of final authority appears to belong to law.

The battle over the authority of money The battle of the authority of law vs. the ordinary use of language by the public is not the only language battle in the trademark war. As several of the cases in this book illustrate, there is also another powerful authority at work – money, especially in the US, but also in other countries, where trademark battles are essentially waged over financial matters. The most common role that money plays is the obvious one: the larger, better financed company can object to a little business’s name and threaten it with expensive litigation that it can ill afford. To avoid such litigation, the smaller company often surrenders the right to use its name. The opposite is also true. A smaller company that believes its name is being infringed by a giant corporation must decide whether or not it has the finances to carry out trademark litigation. Since losing


the case (or even waging the battle, for that matter) may be financially disastrous, the smaller company may simply give up what it believes to be a just cause. Money is often the single most important factor in either decision. A second money factor comes into play as well. Trademark law has created a financial caveat – secondary meaning. If a name, even an otherwise weak descriptive name, can be shown to be identified with an individual manufacturer and the public is motivated to buy it for that reason, its use by another company is said to cause likely confusion in the marketplace. Promotion, advertisement, and use of that name leads to consumers associating it with only one seller, regardless of any claims of registration or use by another allegedly offended party. Thus, even if a mark is descriptive, it can be registered and protected if there is proof of secondary meaning and has become distinctive. So how does such strength of a descriptive mark develop and grow? Through the expenditure of money on promotion and advertising. The company with the greatest resources clearly has the advantage here, so much so, in fact, that its expenditures meet the requirements and the authority of law. The case of McDonald’s vs. Quality Inns, described in chapter 8, illustrates how secondary meaning was crucial in a trademark case. In a sense, the authority of money influences, or complements, the authority of law.

The roles of attorneys and linguists in legal battles With good reason, lawyers consider themselves experts in the use of language. Why then, should they need a linguist to help them with their cases? Again, because trademark cases are as much about language as they are about law. It would be unthinkable for medical malpractice attorneys to fail to use physicians as consultants and expert witnesses in their cases. It could be disastrous not to use an engineering expert in a product liability case involving the collapse of a bridge. In most cases, lawyers are good at knowing what they don’t know as well as where they can get technical assistance. In the case of linguistics, however, attorneys are not always as clear about what they don’t know. They often don’t even know that there is such a specialization that will help them find the reasons why there are difference or similarities in sounds and looks of a given trademark. They are often unfamiliar with the principles of semantics and pragmatics that can help them explain differences or similarities of the meanings of trademarked words and expressions. And even if they realize some of this, they may not know where to go to find such help or how to ask for it.


The cases described in this book are intended to show how linguistic principles can help trademark attorneys. Like most humans, attorneys know how to use language without knowing the details of how they are doing it. In trademark cases, for example, lawyers can know that two different marks do not sound alike, but without specialized knowledge of linguistics, they cannot specify the scientific differences nor the reasons for them. English syllable stress rules explain the differences in the names of the pharmaceutical products, B-OVal and OVRAL, as chapter 9 illustrates. The phonological rules permitting the phonetic realization of the “o” sound in AutoMation became an important issue in the case described in chapter 11. Few attorneys could explain such language behavior without a linguist to help them. Unfortunately, it is common for attorneys to give only limited information about trademark law to the linguists whom they call on for help. In this, the role of a linguist may be somewhat different from that of physician or engineer consultants, whose tasks may be more discrete. All experts need to be sensitized to the legal system in which they are expected to participate. Linguists in particular need to be informed about the legal theories to which their testimony will be relevant. They need to be informed about what the law means by concepts such as “merely descriptive,” “confusion,” “sufficiently similar,” “the doctrine of foreign equivalents,” “the imagination test,” “strong and weak marks,” among other things. It is equally important for attorneys to tell their linguistic experts about any other type of evidence they may be presenting, especially that regarding surveys of marketplace recognition, so that the linguists won’t try to extend themselves beyond what is appropriate or necessary, and thereby risk becoming potentially damaging to the case. It is also important for linguists to be informed about how the law treats concepts such as generic, descriptive, suggestive, and arbitrary marks, so that the linguist’s work can be made relevant to such definitions. Nor are linguists always helpful to attorneys in their clarity about what their field can and cannot do for the case; this sometimes leads to a waste of time and resources. In this book, I attempt to address some of these situations. Perhaps most important of all, attorneys must be willing to spend enough time with their linguistic experts, a practice which, in my experience, is not commonly followed. These issues are addressed more fully in chapter 16. Lawyers do not often give their linguistic experts sufficient knowledge about the pace and culture of the world of litigation. They are not often clear enough about the special aspects of the law that can help linguists focus their efforts on the specific areas to be analyzed. In one criminal


case, I once spent a great deal of time showing how the suspect was entrapped, only later to learn that the legal interpretation of entrapment was more technical and specific than my semantic analysis of it. In the case of trademark law, there appear to be certain essentials that are useful for linguists to know as they perform their analyses. It is important for them to understand the limits of what may be needed of their linguistic analysis. For example, linguists should know that market survey experts tend to deal with issues about what the consumer actually knows, removing this area for linguists to try to deal with. Chapter 2 presents some of the legal issues and concepts that linguists might want to know about, just in case their attorneys neglect to tell them. Linguistic experts are not always kept abreast of the rapid changes that occur during the trial itself. On several occasions, most of what I had prepared as testimony was dropped by the day I came to court, since developments in the case had taken care of any need for it. Points that had previously had low salience now were emphasized. Issues that had seemed central now became less so. Linguists, like other academics, need to know not only what is being changed but also why this is happening. Since many trademark cases are bench trials, this too suggests considerable preparation of experts by the attorney. Testifying before judges is very unlike testifying before a jury. Attorneys often know the ways that certain judges prefer to conduct their trials and would do well to share such information with their experts. Since they are the trier of the facts, judges may themselves ask questions during testimony. Linguists who have appeared only before juries should be prepared for this. In summary, linguists working in institutional settings such as education, health care, and law always find it helpful to be given a working knowledge of the terminology, traditions, culture, pace, and general practices of those institutions. Those linguists who have worked with attorneys discover that, like linguistics, law has a number of specializations, each of which is somewhat different. To have done language analysis in a criminal case with tape-recorded conversation evidence can be quite different from working with the data of a trademark case. This book makes an effort to introduce linguists to such issues as well, in ten case studies and with suggestions about how to deal with them effectively.

2 A Very Brief Introduction to Linguistics for Lawyers

Many people think linguists are people who speak a lot of languages. That’s one definition, but certainly not the only one. The type of linguistics that this book describes refers to scientifically studying languages rather than merely speaking them. Linguists analyze the components that make up spoken language, such as sounds, prefixes, suffixes, words, meaning, sentences, and discourse. Written language shares most of these components as well, except that the sounds of spoken language are represented by letters in written scripts. In addition, all living languages continuously change under the influences of time, space, and social interaction, which create variations within a language that grow out of group differences based on age, gender, race, location, occupation, education, ethnicity, and even religious orientation. Linguists refer to these core components as phonology (the system of sounds), morphology (the system of word parts), lexicon (the vocabulary), syntax (the system from which sentences are constructed), and discourse analysis (the language’s system for putting sentences together to make a unified discourse). But we still haven’t mentioned meaning, one of the major ingredients of linguistic analysis. The meaning conveyed in either written or spoken language is called semantics (referring to the direct, core meaning of words in isolation, as in a dictionary) and pragmatics (the way meaning is conveyed indirectly or through context). In each core component, linguists discover the system, how it works, and what this can say about predicting future use. This constitutes the authority of linguistics which is relevant, in this case, to trademark disputes. It is important for attorneys, judges, and, ultimately, juries, to have this broad understanding of what linguistics is, for one of the initial obstacles that expert witness linguists face is that 17


of a possible misconception of the field. Effective expert reports, depositions, and testimony usually begin by describing the whole field as broadly and as simply as possible, then focus more deeply on those specific aspects of linguistics that are used in a given case. If attorneys are sufficiently motivated, one easily accessible source where they can find readable descriptions of virtually every aspect of current linguistic knowledge is the Handbook of Linguistics (Aronoff and Rees-Miller 2001). For an even more condensed and thumbnail description of what linguists have in their tool bag and how these tools might be able to be used to provide assistance to their cases, the following brief glossary of core linguistic terminology and concepts may be in order.

Phonetics and phonology Phonetics is the study of the nature, production, and perception of the sounds of speech. Phoneticians are trained to hear minute distinctions between sounds and to transcribe those sounds in a phonetic alphabet which is much larger and more complex than the alphabet used in conventional writing. Phonetic symbols represent not only the minute differences in the sounds of spoken speech, but also the pauses, intonation, junctures, and stress as the sounds are spoken. Although phonetics is a way of describing in detail the sounds of language, it tells us nothing about how those sounds are organized by native speakers, who use speech sounds in a systematic way that may be quite different from the way speakers of other languages use those same sounds. The study of the ways speakers of a language organize and use the sounds of individual languages and the general nature and rules of speech sounds is called phonology. Both phonetics and phonology are tools that a linguist can find helpful in trademark cases. It is often useful to distinguish the ways that different trademarks can be pronounced, but it is also important to show where the stress in a syllable may impact these differences, how the seemingly small timing breaks between parts of words can show a contrast, how the influence of a commonly used word can influence the pronunciation of a made-up trademark word, how the English phonological system does not allow certain consonant clusters without inserting a vowel sound between those consonants, how English spelling is used to represent speech sounds, and how the quantity and quality of distinctive phonological features can be used to determine similarity or difference between trademarks.

A Very Brief Introduction to Linguistics for Lawyers 19

When a linguist is called upon in a trademark case, the issue is often whether or not the competing trademarks sound alike. At such times linguists commonly call on their knowledge of phonetics and phonology in their analyses. Some of the actual trademark cases cited in this book illustrate how this can be accomplished (see chapters 6, 9, and 12 in particular). Sometimes phonetic differences are found in syllable stress. Sometimes a useful comparison can be made of the quantity of phonetic sounds (not spellings) that are the same versus the quantity that are different. Phonological differences often are found when foreign words are part of the trademark disputes.

Morphology Linguists refer to the units of lexicon (words) or grammatical units (such as the past tense marker -ed) that are smaller than a word as morphemes. Thus the word “unthoughtful” is considered to be made up of the morphemes “un-,” “thought,” and “-ful.” The base form, “thought,” is called a free morpheme because it can occur by itself as an independent word. But “un-” and “-ful” are derivational affixes, called bound morphemes. They do not normally occur by themselves and are bound to a base form, the morpheme that can occur by itself (called the free morpheme). In the word “cars,” the free morpheme is “car” and the bound morpheme, this time a grammatical inflection, is the “-s” plural marker. The study of the grammatical structure and the categories realized by them is called morphology. Morphology also often has relevance in trademark cases. For example, does the plural “-s” form significantly contrast with a singular form without an “-s” in the name of two different marks? (see chapter 11). Does a noun used in one trademark significantly contrast with that same noun in another mark when a “-y” morpheme is added to change that noun into an adjective? (see chapter 6). When a fanciful trademarked noun begins to be used as an adjective or verb, as when the trademarked proper noun “Xerox” becomes used as a verb (“xeroxed”; “xeroxing”) does this morphological flexibility indicate that the original mark has become so well known that its original uniqueness no longer obtains? (see chapters 8 and 11). Can commercial use cause bound form morphemes such as the Mc- in the name, McDonald’s, to change from their original meaning to the extent that they take on new meanings that are in the public domain? (see chapter 8). These and other morphological issues are explored by linguists who are called upon to assist in a trademark case and actual cases illustrate this in this book.


Lexicography That aspect of language which focuses on individual words is called the study of lexicon, and its specialists are lexicographers, those who often make dictionaries. Some linguists consider the lexicon as a component of the grammar system with lexical rules shared at different levels. Others consider the lexicon as an unstructured list of independent units of language, much in the way a dictionary does. It is common in trademark cases for dictionaries to be used to determine the pronunciation, etymology, and meaning of various marks. The linguist’s job is to assist attorneys in selecting the best lexicographic practices and to evaluate the quality of the citations found in those dictionaries. Although there is much overlap in the entries of different dictionaries, some will contain more accurate and complete information than others. Linguists familiar with dictionary practice can be helpful to trademark attorneys in determining which dictionary to use in a given case. The three major criteria for evaluating dictionaries were stated most clearly by McMillan (1949) and are still applicable: quantity, quality, and effective presentation. Quality is probably most important criterion in trademark cases. There is simply no such thing as “the” dictionary. The attorneys who cite a “compact” dictionary do so at their own risk (see chapter 5) and are likely to be trumped by the attorney who cites a good unabridged one. It is also wise to cite a dictionary that has a large and up-to-date citation file upon which its entries are based. In the case of technical terms, one guide to the accuracy and completeness of definitions may come from the quality of the panel of experts used by that dictionary. Another guide can come from simply comparing that dictionary’s entries with those of more specialized dictionaries. Obviously, dictionaries specializing in engineering, medicine, or other sciences and technology are of the highest qualitative usefulness. The criterion of quantity can cut both ways. Large might seem better than small, but this is not always true. Webster’s Second New International Dictionary (1934) contained 600,000 vocabulary entries while its third edition (1961) contained only 450,000. If size were the only criterion of excellence, one might prefer the older edition. What the measure of quantity alone does not tell us, however, is that the 1961 edition trimmed 250,000 entries from its 1934 version, all of which were judged to be archaic. The newer edition then added 100,000 new terms, clearly making it more useful to the modern reader, even though shorter.


The criteria of effective communication, usually thought of as simplicity and clarity, are a bit more subjective. Dictionary makers, like all writers of expository prose, must choose between giving precise definitions and producing ones that a broad readership is likely to understand. Some feel that it is more important to give the general reader clear, direct, and simple definitions, even at the expense of technical accuracy on all points. The widespread belief that a given dictionary is the final authority on the meanings of words can be challenged if that dictionary’s practice of finding authority is questionable. Some dictionaries have usage panels that are alleged to provide such information but, as Landau points out, most of the usage designations are actually provided by in-house editors (1989: 254). On the whole, despite claims of being authoritatively prescriptive rather than democratically descriptive (as the American Heritage Dictionary proclaimed when it emerged in 1979), most dictionaries are based on the usage of the people who speak the language natively. Good dictionaries update the treasure house of citations compiled by the monumental Oxford English Dictionary produced between 1882 and 1928 and continue to accumulate modern citations of actual, current usage. In trademark disputes it is often the case that the modernity of the words or expressions in question or the specialized contexts in which they are used make it necessary to find more current citations of usage. The emergence of electronic search engines used with present-day newspapers and magazines makes this possible. Several of the cases described in this book show how such searches were made and used to help determine the current usage of the media for trademarks and trademark associated words and expressions (see chapters 4, 6, 7, 8, and 11).

Syntax The way that a language systematically arranges words within sentences is called syntax. The traditional way of defining a sentence, “the expression of a complete thought,” has not proved very useful since scholars have never been able to agree on what a thought is, much less a complete one. Another traditional definition says that a sentence is made up of a subject (topic) and a predicate (what is said about the topic). This simplification also creates problems, as when the sentence begins with “It,” as in “It’s a lovely day,” for we are then faced with the question of what the topic “it” really is. Sometimes it is believed that the conventional definition of a sentence depends on the punctuation of


writing: “A sentence begins with a capital letter and ends with a period.” This definition turns out to be useless for analyzing speech, of course, but even for written representations of speech the exact places to insert sentence breaks can be very arbitrary, as arguments over written transcripts of tape-recorded conversations in criminal cases have clearly shown (Shuy 1993). Crystal (1987) estimates that there are some 200 definitions of a sentence in the literature to date. In spite of the past difficulties in defining a sentence, the concept is very useful. Modern linguists continue to address the issue by examining the linguistic constructions and recognizing that the most independent of them can be considered sentences. It is generally agreed that a sentence is the largest unit to which syntactic rules apply, a statement that works pretty well except perhaps for elliptical sentences such as the response, “With a spoon,” to the question, “How should I eat this dessert?” Linguists look for sets of words or morphemes that hang together as a unit. These units are most commonly noun phrases and verb phrases, but also include adjective phrases and adverb phrases. Phrases can be separated into their constituent words, and words can be divided into their constituent morphemes. This concept of sentence structure includes a hierarchy of levels, working upward from morphemes to clause-level constructions. Thus an utterance such as “The man quickly ate the pie” is judged to have a determiner (“The”), a noun (“man”), which together form a noun phrase. To this noun phrase the verb phrase is added, made up of an adverb modifier (“quickly”), a verb (“ate”), and the direct object (“the pie”). Together these phrase make a sentence. The phrases are separated into their constituent parts. Linguists refer to this process as phrase structure grammar, often illustrated with a tree drawing to show these relationships. Although modern linguistic theory favors different versions of generative grammar, when dealing with linguistic lay persons it is sometimes found that elements of more out-of-date syntactic theory may lead to easier understanding. For example, an approach called tagmemics, now largely discarded as a syntactic method, contains the easily understandable concept of syntactic “slots” and “fillers.” That is, a sentence slot can be obligatory or optional, as with the adverb or adjective slots. Other syntactic slots are obligatory, as with the verb slot. On occasion I have found this a useful way to present complex information clearly to juries, as will be seen in chapter 7, where I use it to visualize the syntactic structure of menu items in a trademark case.


Semantic and pragmatic meaning The traditional definition of meaning was that it is “something that is expressed by a sentence.” Modern linguistics has gone far beyond this over-simple statement and takes many different approaches to meaning. Some linguists distinguish between meaning found at the sentence level, independent of context, and meaning found in the discourse context. Other linguists are more concerned about the way that sentence meaning exists as part of the overall language system. Some focus only on the way representations of meaning are derived from representations at the syntax level by means of semantic rules in the lexicon. Others try to derive the meanings intended by the speaker and those understood by the listener, the domain of what has come to be called pragmatics. I have found that attorneys and judges are typically unimpressed with the term “semantics.” It seems to conjure up the impression of verbal quibbles to manipulate or mislead. For this reason in court testimony I have tended to speak of “meaning” as much as possible and have tried to avoid the otherwise perfectly respectable term “semantics,” even though that is exactly what the study of meaning is. If meaning refers to anything, it signifies that words and expressions mean something as they are used in a certain way in a text. To be sure, we can define the meaning of words individually, using dictionaries, other references, and citations files that we develop on our own, but such a definition may or may not parallel the meaning found in an actual context. Issues of synonyms, hyponyms, and antonyms must also be considered, as must polysemy (where the word or expression has more than one meaning) and homonymy (words that have two or more different written shapes). The implications of these distinctions for trademark cases should be immediately obvious. Pragmatics is an important part of the study of meaning as it occurs in the context of social interaction. Pragmatics deals with the effect of the choice of an expression on those who hear or read it. There are norms of formality, intimacy, and politeness that most people assimilate and use differently in their communication with various individuals or social groups. Pragmatics is seen as bound up with structural matters of phonology and grammar while not actually being a part of them. There is some disagreement about the boundaries of pragmatic meaning, as chapter 11 illustrates. Whether or not it extends to conventional areas of semiotics (the study of signs, like traffic lights, colors, and Christmas trees), is still debated.


Since humans cannot always refer to things by pointing at them, we have developed many ways of using words and expressions to provide the sense of the meaning that we want to convey. We can rely on such words and expressions without any reference to the real world objects that they represent, perhaps one of humanity’s most defining and endearing characteristics. Trademark law contains many areas of potential semantic and pragmatic disagreement. Just how does one define the generic quality of a name? (chapter 4). To what extent is a word actually descriptive? (chapter 7). What are the linguistic processes that are at work in the suggestiveness of a trademark? (chapter 5). Is there really anything that can be considered a synonym? (chapter 6). * It is difficult even to imagine an area of life in which these linguistic core components do not play an extremely important role, for it is the nature of human beings to talk and write, whether the topic is formal or casual, or whether the field is medicine, politics, diplomacy, education, business, sports, gossip, family, engineering, the hard and social sciences, the humanities, or, in this case, trademark law. During the past couple of decades, awareness of the importance of the relationship between language and law has grown rapidly. Lawyers have always used language as the major tool of their trade and they are very good at it. But it is also useful to be able to articulate and explain exactly what it is that one is being good at. Specialists in any field often have something useful to contribute to lawyers as they try their cases. For many years, medical doctors, psychiatrists, engineers, and others have been called on to testify many times in civil or criminal law cases. But it is only in recent decades that attorneys have seen fit to call on linguists to describe the intricacies of language in court. Several factors have led to the fact that this development of the intersection of linguistics and law. It is an oversimplification to blame attorneys for being ignorant of the potential usefulness of linguistics. When the evidence in a case is a written contract, a set of jury instructions, a potentially libelous speech, or an ineptly written warning label, it might seem obvious that a linguist might help shed light on the language issues. Unfortunately, most attorneys are so busy with their litigation that they don’t have time to pause and think about the ways that another field might help them. This is especially true about a field such as linguistics, which is largely unknown, misunderstood, or almost invisible to most people.


One of the amazing things about language is that we can use it without knowing how we do it. If we were to stop and think about each sound, word, or sentence, we would probably stumble and mumble. One of the older Charlie Brown comic strips showed the hapless Charlie unable to speak. His nemesis, Lucy, asked him why he was so silent. Once he retained his composure, Charlie was finally able to reply that he was thinking about his tongue and it made him mute. This is not unlike thinking about our feet as we ascend a stairway. Doing so can be dangerous to our equilibrium. We do habitual things without thinking. Language is also like this. So it is not surprising that attorneys have not been quick to see how linguistic analysis might help them in their work. But linguists must share most of the blame. Before a person is offered something, it is hard for that person to know that there is something there to be offered. Until recently, linguists have not been very active in reaching out and offering their services to the many areas in which linguistic analysis might be useful. It should be noted that linguistics has been characterized by two types of interests. One type is composed of linguists whose concern is about how language works, especially in relationship to the mind. Like any science, linguistics describes the field, finds principles and rules of language behavior, and then makes generalizations and predictions about how language can be used. Scholars who focus entirely on these matters are often referred to as theoretical linguists. Then there is the other type of linguists, those who take this knowledge and theory and attempt to relate it to real-life contexts, such as education, medical discourse, mediation, therapy, and law. They are commonly referred to as applied linguists. Most university linguistics departments focus on theoretical issues, and rightly so, since any effort to make applications first requires a thorough knowledge of theory. But those departments that also train for applied concerns commonly tend to focus primarily on second language learning and second language teaching, giving little or no attention to the myriad of other possible applications of the field. For example, there are very few university courses in linguistics and medical communication, advertising, or diplomacy. Disputes over trademarks provide one arena for the useful application of linguistics to real-world issues. To participate in this arena, however, it is necessary for linguists to be willing and able to spring into action and for attorneys to be able to see the usefulness of this field to their practice. It is hoped that this book will help bring both of these to pass. The value and usefulness of linguistic analysis in trademark cases must


first be clarified to the courts. Linguists have come to accept the fact that the rest of the world has little idea of what we do. I have learned that judges sometimes think that linguists do nothing more than teach and speak foreign languages or specialize in types of jargon. Others consider linguists mere arbiters of good grammar or usage. In this, judges are no different from the population as a whole, nor can they be expected to be. Evidence of this misperception of linguistics can be seen in a widely quoted treatise on trademark law, Jerome Gilson and Anne Gilson Lalonde’s Trademark Protection and Practice (1999). Of expert testimony in trademark cases, these authors observe: Expert testimony on the issue of whether a trademark is understood by the purchasing public to be a generic name is of questionable merit. It is the understanding of the public and not that of the linguist or other language expert which is controlling. However some courts have placed considerable reliance on this type of testimony, and one has even found expert testimony to be virtually controlling. What Gilson and Lalonde misunderstand here is simple and common. They are quite right about the need in trademark disputes to discover the public understanding rather than the linguist’s. But linguists do not even attempt to assess the public’s actual understanding. Attorneys for both sides usually obtain market survey specialists to do this kind of work. The job of the linguist is not to discover the intentions or understandings of either the sender of the message or the receiver of it. Linguists analyze the text of the message for what is lexically and grammatically possible for what the sender could have meant by it and what the receiver could have understood it to mean. The linguist also searches for and analyzes third party uses of the words and expressions found in trademark litigation, helping to determine current usage and practice outside the primary evidence presented in the case. Linguists call on their expertise in phonology to measure the degree to which competing marks sound the same or different. Morphology and syntax competence enable linguists to shed light on competing word parts and sentence structures in trademark expressions. Linguists’ knowledge of language variation and change over time helps them provide useful information about a trademark’s past and current usage. But in none of these analyses does (or should) the linguist claim what specific senders or receivers of the message actually meant or understood.


The often critical area of the meaning conveyed by a trademark is enhanced by the linguist’s knowledge of semantics and pragmatics, including the way meaning is conveyed indirectly as well as directly. Dictionaries are often used in trademark cases, but their adoption of acceptable lexicographic practice is not generally known by the public. Linguists can enlighten on such matters as well. This book provides ten cases in which linguistic analysis has been used in trademark disputes. It is intended to provide linguists with some information and insights about how they might deal with attorneys who ask for their help in their own trademark cases and to encourage attorneys to make use of this resource. The cases represented here all made use of linguistic analysis and, in some cases, expert witness testimony at trial. In five of the cases, opposing linguists were employed, providing a contrast of somewhat different linguistic approaches (see chapters 6, 8, 11, 12, and 13). Chapters 15 and 16 provide lists of suggestions, first to linguists who participate in trademark cases, and then to attorneys who make use of them.

3 A Very Brief Introduction to Trademarks for Linguists

Archeologists report that some 3,500 years ago, potters put marks on their products to identify the source of their clay pots. Over the years we have become more systematic about such marks and today a complex legal system has developed to insure the rights of origin of goods and services. We have created trademarks, a word, expression, logo, or other marking device that identifies the source of the goods, product or service, such as Kodak camera equipment or Exxon gasoline. Service marks resemble trademarks and carry the same legal principles for such things as services of various kinds, including entertainment. The shortened, term, “mark,” has come to represent both trade and service trademarks. The owner of a trademark has the exclusive right to use that mark in the owner’s scope of products or services or to license that mark to others for use in different fields. Most important, owners can prevent others from cashing in on their marks or creating confusion on the part of consumers about the true origin of the product or service. Not all countries of the world consider confusion in the same way. In the US and the UK, for example, potential confusion is determined by the potential of language to confuse and by surveys showing the actual confusion of consumers. In Japan, on the other hand, the likelihood of confusion is determined by how the law is read, not by linguistic analysis or market survey research (Okawara 1999). In the US, trademarks are obtained only through use. A trademark must be affixed to the goods as it is displayed and advertised. This is not necessarily the case in many other countries, where different standards exist. In the US it is usually the first person to use a mark who has exclusive right to it, as long as the product is sold in the geographic location in which the mark is commercially used. The US requirements, 28

A Very Brief Introduction to Trademarks for Linguists 29

unlike those of Britain or Spain, do not make it mandatory to register a trademark first. Even if the mark is unregistered, it is protected against infringement in the user’s area of commerce, reputation, and potential expansion, if it can be shown to be distinctive. Unregistered designs or words that are meant to function as a mark, but might not otherwise be recognized as such, universally can be marked with the informal symbol “TM,” along with a circle with an “R” inside it, ®. In the US it is important to indicate that a trademark is registered by using the words, “Reg. U.S. Pat. & Tm. Off.” The British Trade Marks Act of 1994 has been adopted by most of the other European Union members. Since that time, trademark registration includes anything represented graphically and capable of distinguishing goods and services – words, designs, letters, numbers, shapes, packaging, colors, sounds, smells. (For a comparison of international differences for registering and examining trademarks, see Ladas 1975 and Blackett 1998.) The US has both state and federal trademark registrations and laws. Marks can be registered in any state in the US and can be protected as long as the service or products are confined to that state. When commerce goes outside that state, however, it comes under federal jurisdiction, and state protection no longer obtains, a very good reason for the owner to register it federally rather than at the state level. In the US, trademark rights continue indefinitely as long as the owner uses the mark properly, renews the registration every 20 years, and files a declaration of use during the sixth year after it is registered. Non-use for two consecutive years is considered abandonment of the mark in the US, while the time limit is five years in the UK. Laws governing the timing of renewals differ in other countries. A large number of countries have relatively uniform trademark laws, many modeled after the British Trade Marks Act of 1938. Although some differences exist in trademark laws throughout the world, there is a growing tendency toward harmonization of international laws. For example, in 1988, the European Council of Ministers proposed such harmonization to European Union member states. In 1993 the Council of the European Union of Community Trade Mark Regulation adopted the Community Trade Mark (CTM), enforceable in all EC countries. Major remaining international trademark differences center on how the actual registration of marks is accomplished. Some countries, such as France, Germany, and Ethiopia, examine registrability for what is called “absolute grounds of non-registrability,” such as the mark’s descriptiveness and/or deceptiveness. Other countries, such


as Kuwait, Yemen, and Turkey, examine registrability for the likelihood of confusion with previously registered marks. The US, Japan, Hungary, Jordan, and the UK examine for both. As we shall see, these registrability concepts, especially “descriptiveness” and “confusion” often frame ensuing trademark disputes. The growing number of competing producers and distributors of goods has resulted in a shift over time of the concept of trademarks as objects of property to a means of protecting honest and fair dealings in trade. The purpose of trademark law can be summarized as the effort to protect the public against confusion and deception by identifying the source of the products’ origins as opposed to other products of another source. An additional purpose is to protect the trademark owners’ businesses and their advantageous relations with the public symbolized by their trademarks. Recently, the internet is beginning to have an impact on international implications of trademarks and former geographical boundaries are beginning to vanish. In 2001, a British judge awarded the use of the internet initials, WWF to the World Wildlife Fund, despite the claims to those initials by the World Wrestling Federation. Similarly, a Canadian graphics designer named Anand Ramnath Mani, using his internet domain name, www.armani.com to promote his business, rejected the offer of the Italian fashion giant, Georgio Armani, of $750 plus an new Armani suit to give up his domain name. Mr. Mani refused the offer, taking it to the Swiss-based World Intellectual Property Organization for arbitration. This body ruled that Armani could not stop Mr. Mani from using his own name, since he had used it for years with no intention of hurting the men’s fashion company. Even though trademarks are generally considered a local matter and are registered on a country-by-country basis, geographical boundaries are coming more and more into dispute, and, with the global nature of the internet, illustrate the need for even greater international cooperation and understanding. In an important sense, trademark law is about the ethics of honesty and deceit. Infringement of trademark rights can occur when a mark is adopted that is the same or sufficiently similar to the mark of another and makes use of that mark for sufficiently related products or services. The key word here is often “sufficiently.” Controversy usually arises over whether or not consumers will be sufficiently confused about the source of the sponsorship, just how similar the mark must be to rank as “sufficiently” similar, and how related the products must be to fit the category of “sufficiently” related.


Marks that are federally registered in the US are protected under section 32 of the federal trademark law known as the Lanham Act. Beginning with the Trademark Registration Act of 1905 and continuing with the Lanham Act of 1946, the US Patent and Trademark Office and the US Courts have applied certain standards in deciding whether or not a company’s name and the names of its goods are protected from use by others. Some relevant parts of the Lanham Act include the following passages: No trademark by which the goods of the applicant may be distinguished from the goods of others shall be refused registration on the principal register on account of its nature unless . . . (e) it consists of a mark which, (1) when applied to the goods of the applicant is merely descriptive or deceptively misdescriptive of them . . . One of the key expressions here is “merely descriptive.” Rephrased positively, these words mean that a protected mark must be more than just a description of the contents of the product or service. For example, a manufacturer making a product named Chap Stick cannot prevent other manufacturers from producing a product by the same name, since Chap Stick is said to merely describe a stick that is used to prevent chapping and chapping is a term used primarily in relationship to lips. The name is said merely to describe the function of the product. The Lanham Act continues: (f) . . . nothing herein shall prevent the registration of a mark used by the applicant which has become distinctive of the applicant’s goods in commerce . . . Here the key word is “distinctive.” Nothing shall prevent registration of distinctive, as opposed to merely descriptive, marks. Obviously, there is much dispute about how distinctive a product or service name considered “distinctive” really is. Likewise, names that seem “merely descriptive” are sometimes argued to be more than that. To deal with such issues, trademark law offers further categories and definitions. One of the most quoted sources for understanding trademark law is J. Thomas McCarthy’s Trademarks and Unfair Competition (1984 with regular updating supplements). Scarcely a trademark trial goes by without reference to this book or to one or more of the continuing updates of it.


McCarthy sets out four categories of marks: arbitrary or fanciful, suggestive, descriptive, and generic. Linguists who analyze language in trademark cases should be aware that whatever such an analysis yields, the lawyers with whom they are working conventionally set such analysis into the context of these four categories. 1. Generic names From the biological sciences we learn that “generic” means that a name is related to or characteristic of an entire group or class of things. Generic drugs, for example, are said to contain the same ingredients as equivalent trade named-products. “Cola” is generic; “Pepsi-Cola” is not. Generic names are not protected from use by others. Thus generic words such as “salt,” “computer,” or “bran flakes” are available for anyone to use freely, without fear of being challenged in court. An important linguistic question concerns when and how a word can become generic. Whether or not the first use of the mark makes it generic was questioned in Netherlands trademark law as early as 1962. 2. Descriptive names As the name denotes, descriptive words are ones that describe the qualities, ingredients, or characteristics of a product, making them difficult to protect as trademarks. McCarthy explains: A mark is descriptive if it is descriptive of: the intended purpose, function, or use of the goods; of the size of the goods, of the class of uses of the goods, of a desirable characteristic of the goods, or the end effect upon the user. (1984: 442–3) Under both the Lanham Act and common law, one test to determine whether a mark is “merely descriptive” is what that mark, when applied to the goods or services it represents, would mean to the potential buyer. The word “merely” apparently means that if the mark tells the customer only what the goods are, their function, characteristics, use, or ingredients, then the mark is descriptive. In the case of a mark that has more than one word that is descriptive in all of its name, the mark as a whole is also merely descriptive. The descriptive designation includes names that are laudatory and descriptive of the alleged merit of the product, such as Gold Medal or Blue Ribbon, since self-laudatory or “puffing” marks are regarded as a condensed form of describing the character or quality of the goods. Descriptive marks are said to impart information directly, as in Tasty


Bread, Yellow Pages (a telephone directory), Food Fair (a grocery store chain), Holiday Inn (a hotel chain), Joy (a kitchen soap), Raisin-Bran (a breakfast cereal product), and Tender Vittles (a brand of pet food). Even a clearly descriptive name, however, can be protected if it has what trademark law refers to as “secondary meaning.” When a business has used a word exclusively for a long time or has promoted its product to the extent that the public does not register its literal meaning in their mind and associates it only with that business or product, the words are said to have attained secondary meaning. When secondary meaning is proved, the issue of whether or not the words are descriptive is no longer important. An illustration of this is found in chapter 8, in which McDonald’s power and authority of millions of dollars of advertising trumped a linguistic analysis showing that the prefix Mc- had gone through the process of lexical generalization, currently meaning basic, convenient, inexpensive, and standardized rather than referring only to a hamburger restaurant chain. In this case, secondary meaning illustrated clearly the vast recognition that the Mc- mark had achieved among consumers. This descriptive prefix has become accepted by the public as denoting only one seller or source through secondary meaning, in spite of its descriptive nature. Black’s Law Dictionary offers the following definition of secondary meaning: A special sense that a trademark or tradename for a business, good, or services has acquired even though the trademark or tradename was not originally protectable – also termed special meaning; trade meaning. By establishing secondary meaning, descriptive names, such as Holiday Inn or Tender Vittles, also have trademark protection even though they are otherwise considered descriptive. The Lanham Act requires that a mark be judged descriptive before proof of secondary meaning is even argued. Interesting linguistic issues grow out of the very designation “descriptive,” since the more descriptive the term, the greater burden there is on the plaintiff to prove secondary meaning. The descriptive category is not monolithic. Some words are regarded as only slightly descriptive. Others are called highly descriptive and require much more evidence of secondary meaning in order to be protected. Finding the degrees of descriptiveness is a task that invites linguistic analysis. Another interesting aspect of the descriptive category is that a combination of two or more admittedly descriptive elements may some-


times result in a composite mark that is not merely descriptive. For example, for Mouse Seed (a rodent exterminator) both “mouse” and “seed” can be said to be descriptive but, when combined, the result is not. Sugar and Spice (bakery products) and Season-All (storm windows), despite the fact that the two words composing them are descriptive, are also said to have produced a composite mark that is not merely descriptive. Some marks are also laudatory (or “self-laudatory,” also referred to as “puffing”). These are names that ascribe positive properties about products or services either directly or indirectly, to those who buy them. Laudatory marks are held to be descriptive (i.e., Gold Medal flour and Blue Ribbon beer). Efforts to change the form of an otherwise descriptive term, for example by adding a prefix or affix, will not automatically remove the mark from the descriptive category (McCarthy 1984: p. 459). “Flexitized,” a name on a flexible collar, and “Nylonized,” on nylon-treated fabrics, were both considered descriptive despite their addition of the suffix, “-ize.” Nor will misspellings of a word generally change a descriptive word into the non-descriptive category. Spelling “new” as “nu” has not prevented a court from considering the effect of the word as it is properly spelled. Even the spelling of “Al-kol” for “alcohol” was found not to have removed the phonetic identity between the mark and a term descriptive of the product. The phonetic skill of one court was shown clearly in its decision that the product, “Kwix-tart,” an electronic storage battery, was the equivalent of “quick start,” even though the spellings of the two words had crossed word boundaries. “Savon Gas” was considered to be the same as “save on gas,” despite blending two of these words together in the proposed name. On the other hand, if the misspelling results in a word that is not recognizably descriptive of the product, such as “Lektronic,” a name for electric shavers, it can be treated as an arbitrary or fanciful mark. In this case, although the shaver was indeed, “electronic,” that misspelled word was judged to be nondescriptive of the shaver itself. Abbreviations of a descriptive term that still convey the descriptive connotation of the original source term are also held to be descriptive. “B-100” was held to be descriptive of vitamins with 100 mg of vitamin B. But sometimes even abbreviations can reach the category of arbitrary and fanciful. For example, “V-8” was found to be non-descriptive of the juice made from eight different vegetables. The issue of foreign words translated into English poses another inter-


esting linguistic issue. “The doctrine of foreign equivalents,” used in US trademark disputes, is complex and a bit curious. This doctrine tests whether or not the word would have a descriptive connotation to American buyers. Words from modern languages, such as Italian, French, Spanish, German, Hungarian, Polish, Serbian, Japanese, and Yiddish, are judged to be descriptive to that portion of the consuming public that is familiar with the appropriate language. Words from dead languages, such as Classical Greek, or from languages allegedly unfamiliar to the American public, such as Yoruba, Finnish, or Quechua, are excluded from this category. Thus US courts ruled that “Volkswagen” or “VW” means “people’s car” in German and is therefore descriptive of a low-priced, popular car. In this case, even though Volkswagen and VW were ruled descriptive, the marks were eventually protected since they had acquired secondary meaning that denoted them as unique to their maker. The doctrine of foreign equivalents is not without its critics. McCarthy makes the following comments on it: A rigid, unthinking application of the ‘doctrine’ of foreign equivalents can result in a finding quite out of phase with the reality of customer perception. The ‘doctrine’ should be viewed merely as a guideline and applied only when it is likely that the ordinary United States purchaser would ‘stop and translate’ the word into its English equivalent. (1984: 466) To the linguist, the doctrine of foreign equivalents can be something of an puzzle. For one thing, US citizens are sadly lacking in foreign language knowledge and skills, an embarrassment of its own kind. But just how the courts can determine whether or not a foreign language word would have a descriptive connotation to American purchasers is problematic on several counts. Some purchasers may have such a descriptive connotation and others not, depending on the context of their ethnic origins, their schooling, their socioeconomic status, and their acquaintance with the rest of the world. Second, even in the languages that this doctrine identifies as those to which Americans allegedly might relate, there is abundant evidence that American manufacturers are themselves unaware of the linguistic issues involved. Chevrolet’s South American fiasco with an automobile named Nova (translated: “it will not go”) is a poignant example, as was Ford’s problem with its Pinto in Brazil (translated: “tiny male genitals”). And one can only imagine what English speakers thought when they saw


advertisements for Japan’s Kinki Nippon Tourist Company, the Cypriot soft drink called Zit, the Swedish toilet paper called Krapp, the Finnish product for unfreezing car doors called Super Piss, or the Scandinavian Electrolux vacuum cleaner’s slogan, “Nothing sucks like Electrolux.” Third, the US, like other countries, has always been a land of immigrants and today is no exception. Should that list be expanded to include Vietnamese, Hmong, and Lao, for example? And what about Mandarin, Portuguese, and Cantonese? Fourth, the connotations of the use of certain foreign language words differ widely. An expensive product such as perfume or furs may be enhanced with a French name. A technology product, such as computers, may carry a positive connotation by using a Japanese language name. It may prove desirable to use a Finnish or Swedish language name for a furniture product. These associations, real or imagined, can impact on the selection of trademarks. McCarthy is surely correct when he states that application of the doctrine of foreign equivalents can result in a finding quite out of phase with reality. Linguists may also be interested in a sub-category of descriptiveness called “deceptively misdescriptive marks” in the US. These are marks that are descriptive of some characteristic or quality of their products which do not actually contain them (i.e. are “misdescriptive,” with little or no negative connotation). An important further distinction is made between “misdescriptive” and “deceptively misdescriptive.” Misdescriptive marks can be illustrated by the product Glass Wax, a glass and metal cleaner, which contains no wax at all. The court reasoned that this mark was not deceptive even though it is misdescriptive since there is no evidence that the public was influenced to buy it because it contained wax. Some glass and metal cleaners do contain wax and some customers might even believe this product contained wax, but these issues were held to be of no significance in the customers’ decisions to purchase. The metaphoric use of language in this case suggests another venue for future linguistic expertise. Metaphor is a kind of deception that is common to users of language. The category of descriptiveness is a productive source of linguistic analysis. What makes something descriptive? How descriptive is enough? Are there features of descriptiveness that are more salient than other features? What does the modifier “merely” do to the term “descriptive”? How can linguistics enlighten the definition and use of “deceptive” language? What are the linguisitc dimensions of being “laudatory”? And is meaning actually part of what trademark law calls “secondary meaning”?


3. Suggestive names Suggestive marks are ones that fall somewhere on an undefined continuum between poles of merely descriptive and arbitrary/fanciful. Suggestive marks are said to require some operation of the imagination to connect the name with the product. The trademark, Greyhound (a bus line), for example, suggests sleekness and speed, and Tide (a laundry soap) is suggestive to the extent that it pleasantly suggests water. In many ways, Suggestive marks are difficult to distinguish from Arbitrary marks, as McCarthy warns: For legal purposes, there is little, if any, reason to make the distinction, and the cases hardly ever bother to do so. Arbitrary and Suggestive marks fall within the same legal pigeon hole of classification in that neither category requires proof of secondary meaning for legal protection and registration. (1984: 440) One difficulty that this category brings with it is that some names can be categorized only when seen in the context of the products that they represent. “Brilliant” may be a descriptive name for “Brilliant Diamonds” but a suggestive name for “Brilliant Furniture Polish.” Here the role of the context in which the word is used becomes crucial. It is very difficult to draw a bright line between descriptive and suggestive marks in the abstract. It takes the context, in this case the product, to make that clear. Further complicating matters is the fact that there remains some element of description in any suggestion. The suggestive process would be difficult to begin without it. It is also quite likely that all marks are suggestive in some sense, otherwise they wouldn’t work at all. The legal tests for suggestive marks are four: the imagination test, the competitors’ need test, the competitors use test, and sometimes the dictionary test. The most commonly used is the imagination test. The more imaginativeness that is required by the consumer to get some direct description of the product from the mark, the more likely the term is to be considered suggestive. Whereas descriptive terms convey directly and clearly the ingredients, qualities or characteristics of the product or service, suggestive marks indirectly suggest these things. Applying this test requires using judgment about just how immediately and directly the consumer’s mind works to get from the mark to the product. If the purchaser has to make use of mature thought and engage in a multistage reasoning process to connect the mark to the product or service,


the term is judged to be suggestive. If the mental leap is immediate, the mark is usually held to be descriptive. This test is very interesting but doubtfully provable, at least with our present understandings about how the mind works. According to linguist Steven Pinker: [Language is] needed to help get information from one head to another by way of the mouth and the ear, a slow channel, but they are not needed inside a single head where information can be transmitted directly by thick bundles of neurons. So the statements in a knowledge system are not sentences in English but rather inscriptions in a richer language of thought, ‘mentalese.’ (1997: 70) Just how we can get at these thick bundles of neurons, this richer language of thought, this mentalese, is still imperfectly understood. But that the courts make use of such a test at all is something of a tribute to their wisdom, if not their science. We do not have complete measurements of “multi-stage” operations of the mind as yet, although it seems obviously true, as Pinker says, that such exist at some level. Just how many actual steps or stages a given leap makes would be wonderful to know, however, and suggests an avenue for linguists to follow, if this is in any way possible. The competitor’s need test grows out of the imagination test and logically follows it. Its thesis is that the more imagination that is required to associate the mark with a product, the less likely the words will be needed by competitors to describe their own products. If competitors don’t use it, this is said to indicate that the mark is suggestive, not descriptive. If the message of the mark is so direct that competitors are likely to use the term for their own goods, this indicates that the mark is descriptive. For the linguist, the issue of synonyms and degrees of similarity appear to be relevant here. The competitor’s use test attempts to determine the extent of the employment of another product’s term when it is used to describe their own products. If the public understands that the term identifies or describes the product itself rather than the source of the product, the mark is judged descriptive, not suggestive. Obviously, dictionaries and third party registrations of a mark are evidence of the meaning of a word. Despite these tests to distinguish descriptive from suggestive marks, there remains a large amount of what the courts have called “intuition” in the ultimate judgment. One judge lamented this fact and commented


that it would be better to have a logical analysis susceptible of articulation. The difficulty that triers of the fact have appears to be amply evident in a past decisions in which the name of a food store chain, “Family Market” was held to be a suggestive mark, while the names other food store chains, “Shoppers Fair” and “Food Fair,” were held as descriptive. McCarthy warns that lawyers and judges should not deceive themselves that the distinction between descriptive and suggestive is an objective and concrete system: Rather, in making this difficult classification, they should examine the mark in question in light of the basic rationale for the distinction and try to follow something more objective than a spontaneous, ‘gut reaction’ test. (1984: 497) In that linguistics is concerned with meaning, whether by inference, indirectness, or directness, it would seem that this field holds some promise for helping clarify some of the admittedly intuitive judgments made in trademark cases. Another linguistic challenge includes the apparent need of the courts to define steps in the continuum between descriptive and suggestive, to say nothing of the “imagination test.” And could it be that descriptiveness intrudes into suggestiveness in any way? This category appears to be another area in which linguistic analysis can prove fruitful to the trademark attorney. 4. Fanciful and arbitrary names McCarthy defines a “fanciful” mark as “a word which is coined for the express purpose of functioning as a trademark” (1984: 436). Fanciful words are ones that are either totally unknown in the language or are completely out of the current common usage, as with obsolete or scientific terms that are not familiar to the ordinary consumer. Wellrecognized (and often cited) examples of fanciful trademarks include Kodak (photography equipment), Clorox (bleach), Cutex (fingernail products), Polaroid (photography equipment), Sanka (coffee), Yuban (coffee), and Xerox (photocopying equipment). A problem that occurs when products with certain fanciful marks become very popular is that the general public may adopt that name as though it were a generic term, despite its fanciful origins. This, indeed, has happened with products such as Cellophane, Escalator, Aspirin and, more recently, Xerox (although the latter is still a matter of dispute). Fanciful marks are considered the strongest of all marks. They are novel enough to make a strong impact on the buyer’s mind. As such,


fanciful marks receive the highest protection against infringement. Confusion by the consumer when other companies use a fanciful mark is held to be very likely in such cases. Still, fanciful marks are relatively rare, perhaps because they are so linguistically distant from the product or service they represent. A company takes a chance when creating a fanciful mark. Consumers may find the term so unusual that they do not associate it with the product at all. An arbitrary mark, again according to McCarthy, “consists of a word or symbol which is in common usage in the language, but which is arbitrarily applied to the goods or services in question in such a way that it is not descriptive or suggestive” (1984: 439). For example, it is said that none of the following marks conjures up any mental reaction which is connected with the product at all: Black and White (scotch whisky), Nova (television series), Camel (cigarettes), Shell (gasoline), Apple (computers), Arm and Hammer (baking soda). Arbitrary marks are difficult to distinguish from suggestive marks. They fall within the same parameters of judgment in that neither requires proof of secondary meaning. Legal battles are almost never fought over this distinction. Arbitrary marks, though strong, are not as strong as fanciful marks. They may not suggest or describe anything about their products, but they may be in common use as a mark in their own or in other product fields, making them potentially weak. The major court battles over trademark infringement do not frequently include names that are generic, fanciful, or arbitrary. In cases involving names that are descriptive, the battle is usually over whether or not secondary meaning has been achieved. Some arguments exist over suggestive names, but if suggestivity can be shown to be present in the product’s title, such challenges usually fail. The prefix, “omni-,” for example, is defined in most dictionaries as Latin-based, meaning “all,” found in such words as “omniscient,” “omnivorous,” and others. It has a nice ring to it and is a likely candidate for a trademark. This indeed happened. Omni Hotels registered the name as a trademark in 1984, having used it since 1975. But then along came the Omni Travel Agency in Virginia, and Omni Travel Inc. in New Jersey. Many other travel agencies also began to use “omni” as part of their name. The lawyers for Omni Hotels informed all other “omni-” users that they should stop using that name, and destroy all signage and advertising materials with that name on it. Furthermore, the travel “omnis” were to inform all their clients that their name would no longer be Omni. Logic would tell us that the first company to register the name, Omni,


had the exclusive right to it. But this is not how trademark law works. Registration, in the US at least, is nothing more than a signal that the company is claiming the right. However, the actual right comes not from the registration itself, but from using the trademark in business. And if the company does a lot of advertising and the name becomes identified with that company, it has a very good chance of winning a trademark dispute by proving secondary meaning. The more money a company spends on advertising and building name recognition, the better chance it has of prevailing in a trademark dispute. The more money one spends, the less confusion there is about who Omni really is. But why would anyone confuse a hotel with a travel agency anyway? The Omni case is illustrative. It is possible that mammoth Omni Hotels might have ignored the various little Omni travel companies except for the fact that they discovered that the US Postal Service had begun to investigate a travel operation in Texas using the name Omni International. This company was suspected of fraud in the sale of bargainpriced vacation packages. Omni Hotels feared that their hotel patrons might confuse their own hotel operation with that of the travel company under investigation. Therefore, Omni Hotels got an injunction barring Omni International from using the name while, at the same time, it began to search for other users of the Omni name, eventually leading them to the little travel agencies such as Omni Travel in Virginia. This still does not say all that must be said about potential (or actual) consumer confusion in the overall context of the travel and hospitality business. For example, large airlines now own hotels and hotels now own their own travel companies, further supporting Omni Hotels in their concern about the Omni name being tarnished by a US Postal Service investigation of a travel agency. So when customers use a travel agency in Virginia, do they believe they are dealing with Omni Hotels? And while we are at it, is the name Omni descriptive, therefore not allowing it to be able to claim the protection of a trademark at all? Or is it so arbitrary and fanciful that it can be protected? Does the fact that it is Latin (at least in origin) and not English help or hurt its case for trademark protection? Or is secondary meaning the most powerful factor here? It is of no small interest to linguists that virtually all of what is said about the four categories of trademark law centers on the use of language. Establishing what is and what is not generic or descriptive may seem simple for non-linguist lawyers to accomplish in litigation


(although it frequently is not), but issues of suggestivity, arbitrariness, and fancifulness can require considerable linguistic expertise. Principles of syntax, semantics, pragmatics, inferencing, presupposition, lexicography, phonology, morphology are often central to a resolution of such a dispute. As Tiersma (1999: 121) points out, the law of trademarks is largely about reference: “To be effective, a trademark must refer unambiguously to the maker of a product.” Just exactly how this referencing is made is the subject of most litigation. Opposing parties often argue over three major questions: Does it look the same? Does it sound the same? Does it mean the same? From the linguist’s perspective, these questions translate into common linguistic tools or components. The question of whether or not it “looks the same” is a matter of syntax, morphology, and graphemics (the study of letters and symbols). The question of whether or not it “sounds the same” is a matter of phonetics and phonology (the study of speech sounds and their structures). The question of whether or not it “means the same” is a matter of semantics and pragmatics (the study of word and discourse meaning). The following chapters describe ten individual trademark cases in which these questions, among others, play a central role. The owner of a trademark can sue any actual or threatened infringer whether or not the mark has been registered. Civil remedies for unfair competition, which is what trademark infringement really is, include injunction against future infringement, a substantial portion of the infringer’s profits, damages for past infringement (which can be tripled under US federal law), destruction of all materials carrying the infringed mark, and the costs of the litigation, possibly even including the attorney’s fees. As noted earlier, a common misconception about trademarks is that the first party to apply for or receive rights to a mark automatically wins the right to use it commercially. As some of the cases in this book will attest, this is not always the case. It is commonly known that not all trademarks are equal. Ones considered “strong” are protected over related products and services on the basis of whether or not they sound alike, mean alike, or look alike. Ones considered “weak” are given more narrow protection. As McCarthy points out: The rationale is that the more distinctive, unique and well known the mark, the deeper is the impression it creates upon the public’s


consciousness and the greater the scope of protection to which it is entitled. (1984: 504) But the question is not so much whether a mark is “strong” or “weak” as it is about whether or not the senior user’s mark is sufficiently distinctive that the junior user’s mark will create customer confusion when used in a different format and with a different appearance, when it is used on a different line of goods, or when it is used in a different territorial area. When assessing the strength of a mark, it is necessary to weigh two things: 1. the placement of the mark on the spectrum of marks and assessment of the marketplace value of the mark, focusing on the inherent potential of the term at the time of its first use; and 2. the marketplace recognition value of the mark, which evaluates the actual customer recognition at the time the mark is asserted in litigation to prevent another’s use. One problem is that the concepts “strong” and “weak” are not discrete slots into which all trademarks can be placed. Strength and weakness are seen only in relationship to each other, a continuum of differences in degree rather than categorical. The difference is the effect the mark will have on consumers. Will they be likely to remember it? Will they confuse it with another product? Will they think that it is associated with the maker of a different product? Customer perception is the key. The more distinctive and well-known the mark is, the greater the impression it will have on the consciousness of the consumer. The generic-descriptive-suggestive-arbitrary/fanciful categorization is one of the measures to conclude the strength or weakness of a given mark. But not all descriptive marks are weak. Nor have all suggestive marks been declared strong. Descriptive marks that have achieved secondary meaning and strong consumer recognition have been given protection by the courts (see chapter 8, for example). Strength of a mark, therefore, cannot be determined by determining its place on the descriptive-suggestive continuum alone. Secondary meaning must be added to the picture, making it a two-prong measure: 1. where the mark falls on the spectrum of marks; 2. marketplace recognition of the mark.


Linguistic analysis fits primarily into the first prong, with market survey analysis fitting in the second. It is common for trademark cases to include experts from both fields. The “Five Golden Rules” about how owners should use their trademarks (Blackett 1998) also open the door to possible linguistic analysis. These golden rules are: 1. Always use the trademark as an adjective, not as a noun or verb. It should be in print as an adjective qualifying a noun or noun phrase (i.e., Goodyear tires). 2. Trademarks should not be used as plurals or possessives (i.e., not Toyota’s safety features). 3. Trademarks should be used with a generic term (i.e., Coors beer). 4. Indicate that it is trademarked through various print devices, such as quotation marks, all-capital letters, bold type, color, or with the word “trademarked.” 5. Use the word, “registered,” the symbol ®, or TM. In most countries, trademark battles can be started in two ways: with charges of trademark infringement and with charges of unfair competition. Some differences exist between British law (along with countries following British law traditions) and US (along with countries sharing US law traditions). In the US, decisions about infringement are based on whether or not the alleged offending party’s usage of the mark will lead to confusion, mistake or deception. In contrast, the UK Trade Marks Act of 1994 considers infringement to be based on identical marks used on identical goods or services, on identical marks used on similar goods or services resulting in a likelihood of consumer confusion or association with that trademark, and on identical or similar marks used on non-similar goods and services when their use would take unfair advantage of or be detrimental to the character or reputation of the mark. In the UK the term “passing off” is used to describe the action of protecting the reputation and goodwill of a trademark that comes from its use. It involves a mark’s reputation gained through use, any misrepresentation of it, confusion growing out of its use, and real damages. The US and most countries of continental Europe make use of a similar concept called “unfair competition.” Passing off leads to a focus on issues of actual financial damage while “unfair competition,” as its name implies, focuses more on issues of intention to benefit unfairly from the trademark owner’s goodwill and reputation, especially in the long term.


In many legal cases, both infringement and unfair competition/passing off are charged. Trademark attorneys tend to use accountants and other experts to deal with representations of actual damage and linguists to address the issues of linguistic similarities and the ways that language use can give clues to intentions. The following chapters are case histories of how linguists have been, and can be, used in such cases.

4 Generic vs. Secondary Meaning

In the fall of 1992 a Dallas-based corporation called Registry, also doing business as American Hospitality Management, brought a series of lawsuits filed in several venues against companies who used the term “hospitality management” to describe their products or services. Registry owned and operated five hotels under the name Registry, including one in Bloomington, Minnesota, and nine more under the assumed name, Hospitality Management Corporation. Once Registry became aware that the words “hospitality management” were being used by other companies, they began to send notices to those companies, demanding them to cease and desist using of the term.

Registry Hotel Corporation d/b/a Hospitality Management Corporation v. Hospitality Management Company One of Registry’s lawsuits was filed on December 23, 1992 and was directed against American Hospitality Management Company, a threeperson company managing two hotels. Registry and AHMC had competed and had direct dealings with each other since AHMC opened in 1988. The suit was filed in the US District Court of Minnesota, Third Division (3-92-852), claiming trademark infringement, false designation of origin, and unfair competition. At the time of this lawsuit, at least 14 companies were found to include the term “hospitality management” in their company names. Other companies did not use the term in their firm names, but indicated specifically that they offered services that they called “hospitality management.” The attorney for defendant American Hospitality Management 46

Generic vs. Secondary Meaning 47

Company, Bruce Little, called me to assist him in this case in March 1993. One obvious question was whether the term “hospitality management” is generic, descriptive, suggestive, or arbitrary. Intuition, if not common sense, would tell us that it is not arbitrary or suggestive in this context. Apparently reaching the same conclusion, in its initial Complaint, plaintiff Registry claimed that its name had already reached the level of secondary meaning: Because of the distinctiveness of the Hospitality Management service marks, these marks have acquired a secondary significance in the hotel, motel and restaurant industry, indicating a single source, namely HMC, for services offered by HMC or under its surveillance and authority. From the original complaint, the battleground was fairly clear. The issue would not be one of descriptiveness, for it seemed patently obvious that “hospitality management” pretty clearly describes the product or service. But was the term also generic? And could plaintiff’s claim that it had developed secondary meaning hold up? The linguist’s task was to discover how “hospitality management” was used by the industry and, for that matter, by the English-speaking world as a whole. I suggested that Little’s law firm conduct a computerized database search for the occurrence of the words “hospitality management” in major newspapers and magazines beginning in 1980 to determine whether or not the expression gave evidence of being used generically. I was also given newsclips from the journals Hotel and Resort Industry and Hotel Business News, both from January, 1993 issues. In addition, I was supplied with the 1992 Directory of Hotel and Motel Companies, a rich source of already existing names and services in the industry. Findings from the database From these sources I concluded that “hospitality management” had four categories of usage: company names, academic or training program names, names of books and magazines, and other common noun uses of the expression. 1. Company names Some 47 service companies include the words “hospitality management” in their titles, including International Hospitality Management, Inc., Sulcus Hospitality Management Systems, Landcom Hospitality


Management Company, Hospitality Management Behavior, Inc., all described as “hospitality management companies.” 2. Academic or training programs The academic or training programs cited in the database included some 44 universities, colleges, programs, and schools that used the words “hospitality management” in their program names, including: • The Florida International University School of Hospitality Management; • Penn State University Hospitality Management Program; • Bethune-Cookman School of Hospitality Management; • Leeds Poly Hospitality Management; • The Washington State University School of Hospitality Management. Other academic programs used “Hospitality” in their titles and also included courses in “hospitality management” in their curricula. These included the Swiss Hospitality Institute and the Oregon State Hospitality Program. Some institutions did not have “Hospitality Management” in their program titles but offered courses by that name, including courses at Harper Community College, McLaren School of Business, and the Community College of Denver. 3. Books and magazine titles The titles of published books and magazines found in the database also gave evidence of the common occurrence of “hospitality management.” One example is a book by Andrew Schwarz and David C. Dorf titled Glossary of Hospitality Management Terms. An example of a magazine is a specialized journal, called simply Hospitality Management. 4. General, common noun uses of the expression The general, common noun phrase uses of “hospitality management” and its referent, “hospitality,” included over 300 citations in the database, the majority of which used the expression either as a modifier of a following noun or as an independent noun construction. One example of such general usage was found in The Cook Political Report of October 29, 1992, which described a certain Congressman Lepinske’s background as follows: Lepinske then broke into food service and hospitality management and eventually became president and CEO of a small but diverse


corporation dealing in seafood brokerage, contract food service management and retail restaurant facilities. In addition to his work in the field, he had taught hospitality management at several colleges. Another interesting reference was found in Nation’s Restaurant News of October 5, 1992, which reported: There are too many lawyers because there are too many laws, and the challenge we have is to recognize the problems and find solutions, said panelist Fletcher Hudson, a Memphis attorney who represents hospitality management in labor law cases. There was ample evidence in these four categories of usage of the common noun expression “hospitality management,” for anyone to see that it is widely used in a number of different contexts. Forty-seven companies use the expression in their titles; 44 universities teach hospitality programs. There are published books and journals with that name. In all, I found over 300 citations of “hospitality management” in the database of media articles. It would appear from this that we were on solid ground that the expression had a generic quality which could cast doubt on any claim that Hospitality Management was guilty, as plaintiff claimed, of infringing on Registry’s trademark, suggesting false origin, or giving unfair competition. Linguistic analysis of the database Having determined these categories in which “hospitality management” was used, the next step was to discover how the meaning was conveyed. We have many examples of how to do this, beginning with the Oxford English Dictionary, in which thousands of citations were found in literary use to create the definitions found in that dictionary. Other more modern dictionaries, including Webster’s Collegiate Dictionary and the Thorndike-Barnhart Dictionary, also compile large citation files upon which to ground the definitions in their entries. Such practice is good, scientific lexicographic procedure. A dictionary’s main task is to describe the usage found in public representation of the words in naturally occurring contexts. This principle is not widely recognized by the lay public, many of whom believe that dictionary makers somehow determine the meanings of words by a mysterious individual intellectual superiority and skill. In contrast, as Bright (1992) points out, The evidence used by lexicographers comes from introspection, experiment, and observation. This view is well borne out by the lexicographer Sidney


I. Landau, whose Dictionaries: The Art and Craft of Lexicography (1984) should be considered by linguists and attorneys alike as they consider the quality and use of dictionaries in the knotty problems presented by trademark cases. Two characteristics of the quality of being generic include a word’s flexibility and recency. Flexibility means that it has shown the ability to move easily across grammatical functions: it does not remain fixed as noun, verb, or adjective, but may shift back and forth between these grammatical distinctions. Recency means that by being relatively new, it has not yet had the time to fix or freeze in meaning or in function: it is still an active, dynamic word. If, by examining the database, we could find evidence of flexibility and recency, this would assist in determining that Registry’s claim to sole use of “hospitality management” was weak. Flexibility The linguistic principles of combining and contracting usually give evidence of either an expression’s dynamic or fixed status. One of the signs of the generic nature of an expression is its ability to flexibly move across grammatical categories. Some expressions become fixed or frozen in meaning or grammatical form. Others are dynamic, moving from the noun to verb to adjective categories with relative ease. Trademarked words or expressions that are flexible and dynamic are more difficult to nail down and protect from use by others. If “hospitality management” has been in the lexicon for many years, and if it no longer shows ability to combine with other words or contract from its base, a reasonably good case could be made that its use is most common with the originator. The example of “Xerox” demonstrates that this term has become flexible to the extent that it is now used by the public in both its adjective and verb forms, despite efforts by its trademark owner to prevent such flexibility. The principle of lexical combining involves juxtaposing a noun, such as “air,” with another noun, such as “line,” in order to create a new compound noun, “airline.” In this case, “air” shifts its function from that of a noun to that of an adjective modifier, indicating a type of “line,” the base form of the construction. Likewise, in the current litigation, the noun “hospitality” combined with another noun, “management,” creates a compound noun, “hospitality management.” Within this compound construction, “hospitality” shifts its function from a noun to a modifying adjective, indicating a specific type of man-


agement. The principle of combining can be extended even further. When the noun compound “hospitality management” is used before another noun, such as “hospitality management program,” or “hospitality management service,” the entire “hospitality management” construction itself becomes an adjectival modifier of a specific “program” or “service.” The principle of lexical contracting is also evident in this case. When the noun “management” of “hospitality management” is deleted before another noun, such as “program” or “services,” this leaves only “hospitality” as the modifier, which shifts its form from an adjective phrase to a single adjective. In the database, “hospitality” sometimes also becomes the sole explicit reference to “hospitality management,” cycling back to its original noun function as a result of the contraction process. When the principles of combining and contracting are present, the construction shows lexical flexibility and is considered to be an active, dynamic (rather than fixed) form. Active language constructions give evidence of common, generic use, unfettered by whatever special definitions and functions that may be claimed for them. This is why owners of successful arbitrary or fanciful trademarks strive to prevent this from happening to their marks. It is why the owners of the marks Xerox, QTips, and Kleenex strongly oppose the public’s extension of their marks into apparent generic uses. When the processes of combining and contracting are operating, it is very difficult, if not impossible, for such constructions to be contained. They are passing over their former limits as a nouns, adjectives, or compounds to more dynamic, functionally adaptable, flexible, commonly used structures. Recency Another indication of the commonness of an expression is found when there is a dramatic sudden increase in its usage. The articles cited in the database showed that “hospitality management,” along with its noun referent, “hospitality,” were in use at least 13 years from the time of the litigation, but also showed a dramatic increase during the most recent period. In the four years before the litigation (1989–92), the 137 occurrences of “hospitality management” in the database contrast with only 37 occurrences in the preceding eight years (1981–88). “Hospitality,” as a referent for “hospitality management,” appeared 117 times between 1989 and 1992, but only seven times before 1989. It was clear, then, that the use of “hospitality management,” along with its derivatives,


both as nominals and noun phrases, was fairly recent. This quantitative increase of usage since 1989 was another indicator of common, accepted usage. The geographically widespread locations of this usage showed that it was not limited to any particular region. It was found in many different contexts as well, including education, law, computers, real estate, consulting, hotels, and restaurants. The term appeared to have made its way into the general vocabulary of English. Definitions How words and expressions are defined is not limited to dictionaries. Evidence of what “hospitality management” meant at the time of the litigation was also found explicitly and implicitly in the media writings themselves. Many such direct definitions occurred in the database, including the following from Crain’s Chicago Business on April 10, 1989: As recently as a decade ago, there were only 20 colleges offering fouryear degrees in hotel and restaurant management – or hospitality management, as it’s come to be called. Historical lexicography also supports the common and generic use of “hospitality” and “management.” The word “hospitality” derived from the French “hospitalité,” according to The Oxford English Dictionary first cited in the year 1375, is defined as: the practice of being hospitable; the reception and entertainment of guests, visitors and strangers with liberality and good will. This meaning has been maintained in English to the present day, indicating no new, unusual, or unique meanings of the expression. The word “management,” also derived from French “ménagement,” was first cited in 1598 (The Oxford English Dictionary), and is defined as “the action or manner of managing.” This meaning has also been maintained in English to the present day, indicating no new, unusual or unique definitions of the term. But no dictionaries, past or current, included the compound, “hospitality management” in their definitions or entries, indicating once again the obvious nature of the meaning of the expression. The consuming public can determine the meaning of “hospitality management” easily and obviously from its constituent parts, requiring no more further definition than would the expressions, “baseball management” or “office management.”


Non-distinctive (generic) uses The non-distinctive or generic uses of the expression “hospitality management” can also be seen in the lack of precision made by writers when they refer to it in company titles or academic programs. Many such imprecise references appear in the database, such as the one in an article about Sulcus Hospitality Management Systems, which is referred to in the same article as “Sulcus Hospitality Group.” Likewise, an article about International Hospitality Management, Inc. is referred to in the same article as “Hospitality Management.” Academic programs evidence the same referencing flexibility, as when the California Polytechnic Institute’s Hospitality Management School is referred to as Cal Poly’s “hospitality school.” When the writers in the public domain treat titles with such flexibility, it is a clear indication of the non-distinctiveness of the referent. Conclusion of the case At attorney Bruce Little’s request, I prepared and submitted an affidavit that included the above information in March, 1993. Plaintiff had scheduled a Motion of Preliminary Injunction at the end of the following month. Making liberal use of my affidavit, defense attorney Little submitted a Motion for Summary Judgment at the beginning of April. On May 27, 1993 I received word from Little stating that the lawsuit had been dismissed and that the matter had been settled with terms quite favorable to his client. As is common in such matters, a confidentiality provision in the settlement prohibited my learning any further details. The issue of claiming the advantage of secondary meaning was blocked by the fact that the Court had apparently accepted the argument that “hospitality management” was generic, and did not even rise to the level of being descriptive. If the term had been considered descriptive by the Court, the battle over secondary meaning could have taken place. This would have required the attorneys on both sides to show the extent to which consumers recognize one or the other litigant clients’ marks to have the most recognition by the public. In such cases, consumer survey experts are conventionally called upon to determine such facts. This is a difficult and expensive task, one that most trademark lawyers would prefer to avoid. Thus, if linguistic evidence is used effectively, such surveys can be avoided. This case showed that it can be financially helpful for a trademark attorney to make early use of linguistic expertise. It is difficult to know how this case might have gone without


linguistic help. It is possible that attorney Little could have shown the differences in capitalization between “Hospitality Management” and “hospitality management.” He could have cited third party uses of the term as well. What his case would have lacked, however, is the authority of linguistic analysis in showing how genericness works, where it comes from linguistically, and evidence of its growth. On the other hand, case history could also have worked for Little’s defense. There are well-documented instances of the transformation of a trademark into a generic word. For example, for the first 15 years of production and sales, the manufacturer of a vacuum bottle called Thermos used “thermos” as a generic term. For the following 30 years, the manufacturer, probably sensing his mistake, referred to it as a “Thermos vacuum bottle.” But by then the public had been taught by the very manufacturer that Thermos was a synonym for a vacuum bottle. Between 1953 and 1962, the company tried to stop the very generic use that it had produced, but it was too late, the Court ruled that, by its own hand, Thermos had created the generic term. The trademark infringement case brought by Thermos failed. Hospitality Management’s case was probably not even as strong as that of Thermos, since it alleged that its trademarked name was not generic from the very beginning.

A postscript on the use of dictionaries Lexicographers and the courts are often in conflict over what is generic and what is not. When a trademark is used by the public as though it were a generic word rather than as the brand itself, some dictionaries choose to record this usage. Most trademark owners are not at all happy about such reporting, even though the dictionary entry for this item may be produced with an initial lower case letter and identified as a trademark as well. The owner of the trademark believes, under the authority of law, that the generic use is illegal and that the dictionary is reporting this generic use by the public in contempt of the owner’s proprietary status, yet another instance of an infringement of the protected mark, this time by the dictionary. Thus we find an interesting conflict between the duties of competent lexicographers and the laws of trademarks. The dictionary maker’s duty is to record how a word is used by the public. If the public commonly asks for a “kleenex” instead of a “paper handkerchief,” that’s exactly what the dictionary is supposed to report. That mark has been used as a generic term and the dictionary maker is obliged by the standards of


the lexicograhical profession to indicate this. Trademark owners, vigilant for infringements, may then ask the dictionary makers to omit such generic citations from their dictionaries, since they believe the words are trademarked and not generic. Lexicographers may wonder what to do in such instances. Their citation files show numerous citations of the generic use of “kleenex,” “xerox,” and “band aid” (in various noun, verb, or adjective forms). These words are part of the language, like any other terms. If lexicographers omit such words, they are not following their duty to present the language in a way that users of the dictionaries who are unfamiliar with those meanings can discover them. By lexicographic standards, if the facts warrant, any word can be generic and any generic word should be included in the dictionary. Can trademark owners dictate lexicographic facts? Is this not a form of language policy and planning that violates natural language usage? Does the authority of law override the authority of the public? It should be pointed out that in such cases, the dictionary makers have no intention of depriving trademark owners of their exclusive rights to use their marks in the world of commerce. Lexicographers simply find that their primary obligation, the reason for their very existence, is to dictionary users and that no special interest groups should override such an obligation. Yet override it they have. If one examines the most recent editions of Merriam Webster’s Collegiate Dictionary or The American Heritage Dictionary, for example, one will find the words, “Kleenex,” “Band-Aid,” and “Xerox” defined as trademarked products. There is no accompanying citation for the generic use of these words. In these instances, at least, the trademark attorneys have won. They have used the threat of law to quash the obligation of lexicography to report the actual usage by the public. One might hope for a different outcome of George Orwell’s predictions about the way the government controls language in his futuristic novel, 1984. And, in this case, one also wonders what would have happened if Registry’s attorneys had managed to convince lexicographers that they should include “Hospitality Management” as a trademarked entry in their dictionaries.

5 Teaching a Jury about Meaning

It is always interesting when a “David”-sized company takes on a “Goliath.” Warren Distribution was a small manufacturer of antifreeze products, primarily for use in trucks, based in Omaha, Nebraska. It had developed a product called LongLife and was doing rather well in its relatively small-scale operation. The name must have been a good one, since a much larger company operating on the national level, Prestone Products, also began using LongLife as the major part of the name of one of its antifreeze lines, called LongLife 460.

Warren Distribution, Inc. v. Prestone Products Corporation In March, 1993, Warren Distribution brought a lawsuit against Prestone Products Corporation, located in Danbury, Connecticut, claiming that Warren’s trademarked LongLife antifreeze product was being infringed upon by Prestone’s product, LongLife 460 antifreeze. Warren claimed that Prestone’s action would result in confusion or misunderstanding of the source, with purchasers likely to believe that the antifreeze sold by Prestone was actually produced by Warren, or that the two companies were connected with each other in some way. Prestone’s answer and counterclaim, filed two months later, argued that the words “LongLife,” in connection with the sale and advertisement of an antifreeze/coolant, is no more than a common descriptive name of a type or class of antifreeze which is believed to extend the life and efficiency in an automotive radiator. Prestone further argued that the American Society for the Testing of Materials (ASTM) had formed a subcommittee to establish standard testing materials for evaluating “long-life engine coolants” and that major automobile manufacturers had decided that their products should contain such long-life coolants. 56

Teaching a Jury about Meaning 57

This being the case, defendant reasoned, the industry has come to refer to “long-life coolants” as a class of antifreeze and that consumers would treat that name merely as a descriptive, even generic, term. If this were the case, it would make Warren’s LongLife incapable of exclusive appropriation as a trademark. It was at this point that Warren’s attorney, Bartholomew McLeay, called me to help with his case. Prestone’s defense arguments As is common in such cases, both parties filed briefs for summary judgment. Prestone argued not only that the term was generic, but also that it was descriptive, and that even though Warren’s application for trademark was filed four months earlier than Prestone’s, Warren had not developed secondary meaning by the time Prestone had begun to use its only slightly different name, LongLife 460. Warren started to sell its product in 1995, whereas Prestone claimed to have begun its sales in 1993. This argument is traditionally dealt with by experts in accounting and sales and is outside the scope of linguistic analysis but, in any case, it never emerged as a salient issue in the subsequent trial. In trademark cases, language data are often central. Attorneys conventionally bring in dictionary definitions, advertisements, corporate memoranda, product labels, press releases, and media references about the product in question. I was asked by attorney McLeay to use my linguistic expertise to address these accumulated materials. I did so, first in my expert report, submitted in January, 1996, and in my following deposition on February 22 of the same year. Dictionary definitions In an effort to prove the genericness of Warren’s LongLife mark, Prestone had used separate dictionary definitions of “long” and “life,” citing The Compact Edition of the Oxford English Dictionary (1971). Prestone called attention to the following definition of “long”: having a great extent in duration. Prestone then cited that dictionary’s separate definition of “life”: the term or duration of an inanimate thing; the time that a manufactured object lasts. Having used these separate dictionary definitions of “long” and “life,” Prestone was then unable to locate a dictionary entry for the combined


expression, “long life,” so it cited what it considered the next best thing, using the Random House Dictionary of the English Language (2nd edn. 1987) definition of “long-lived”: having a long life, existence, or duration. In deposition and at trial I pointed out that “long-lived” does not equate with “long life,” since the two expressions are grammatically non-equivalent. Warren’s trademark, LongLife, is an adjective–noun combination forming a proper noun, while Prestone’s definition, “longlived,” combines an adjective and a verb to form another adjective. In addition, I noted that “long-lived” appears to be becoming archaic as an English expression. Two things are important about defendant Prestone’s dictionary evidence. First, Prestone attempted to define the words, “long” and “life” independently, not as a compound noun structure found in the mark, “LongLife.” Second, curiously enough, as they used the dictionary senses, they stressed only the long life of the engine that uses the product and overlooked or ignored a second meaning used by Warren, the advertised long life of the antifreeze itself. Thus two linguistic issues became central to the case: the alleged generic quality of the compound construction versus its component parts, and the referential meaning of “longlife” as Warren employed the term. One of the first things apparent from the Prestone’s use of the dictionary citations is that these were highly selective. In my affidavit I first pointed out that Prestone had cited separate and selective definitions of “long” and “life” as though these separate and selective definitions are the only senses found under those entries. The 1989 edition of the Oxford English Dictionary lists 14 different senses of “long,” from which Prestone cited only one. The overwhelming number of definitions, not cited by Prestone, include: 1. 2. 3. 4. 5. 6. 7. 8. 9.

spatial measure; extension from end to end; reference to shape; reference to liquor; reference to serial of spatial extension; implication of excessive duration; excessive expectation; continuity; distant or remote point in time;


10. phonetic usage; 11. continuity; 12. special terms of commerce. Prestone’s selective use of a dictionary with regard to “long” was not outdone by its equally selective use of the sense of “life.” The Oxford English Dictionary (1971) lists 18 senses of “life,” of which Prestone’s corresponding reference was to only the OED’s number 8c: the term of duration of an inanimate thing. All other senses in this dictionary relate to animate existence. The single inanimate sense selected by Prestone, amidst all other animate senses, overlooks the fact that this single inanimate sense is hyperbolic. Readers of this hyperbolic sense must use their imagination to associate the common use of the word “life” with inanimate objects such as antifreeze/coolants used in engines. In order to confirm that there was no recent change in this association, I checked the 1989 edition of the Oxford English Dictionary, which also includes only one inanimate definition among its 18 senses of “life.” I continued my dictionary hunt. Of the 21 senses of “life” listed by Webster’s Third International Dictionary (1971), in only two subparts of sense 14 (14b and 14f), is “life” associated with inanimate objects. By far the preponderance of senses of the word “life,” in both these dictionaries, is to animate existence, a fact conveniently ignored by Prestone as it apparently selected definitions to suit its own specific purpose. As noted earlier, Prestone reported dictionary definitions of “long” and “life,” then claimed that these separate definitions are consistent with “long life” as the industry’s alleged term for a type, kind, or class of coolant/antifreezes. Contrary to Prestone’s suggestion, we cannot always assume that when we put individual words together to make a new compound expression we preserve the meanings of the independent components from which they were derived. For example, if we take the adjective “short” and the noun “stop,” and combined them to form “shortstop,” the resulting word does not mean that someone or something short is some kind of stop. Likewise, if we combine the nouns, “dead” and “pan” into “deadpan,” we do not easily understand this to be a pan that once was living but is now not. To understand how such combining works, we also need to understand how semantic change works. Semantic change, whether it relates to individual vocabulary or words combined to make a new word, comes in four types of processes:


specialization, generalization, pejoration, and amelioration. “Shortstop” and “airline,” used above, are examples of the process of specialization. The two words combined create a new word with a specific meaning that is different from the original two parts. The process of generalization is illustrated by words like “frock,” which at one time referred to a specific type of coarse gown worn by monks, but by modern times has generalized to any kind of gown or dress. The process of pejoration is illustrated by the word “villain,” which once referred to a farm laborer, but now has taken on a pejorative reference. The process of amelioration is illustrated by the word “marshal,” which once referred to a humble groom at a stable, but now means a leadership rank in government or armed services, among other things. Prestone claimed that “long” and “life,” when combined as one word, continued to refer only to the long life of the automotive object of its use, whereas Warren had generalized its use to also include the long life of the agent itself, the antifreeze. Prestone claimed that only one of the many meanings of “long” and “life” was maintained in each of the derived words that made up “long life,” and overlooked or ignored the fact that the single meaning that it attributed to “life” is the hyperbolic one that requires a reader’s imagination to apply it to an inanimate product. For example, Life cereal requires imagination of the reader to apply an animate meaning of “life” to this inanimate breakfast product. There is a degree of inherent incongruity in a reference to the “life” of an inanimate object or process, one that suggests that the imagination test might be in order here. The trademark LongLife is an adjective–noun combination forming a proper noun. It is not possible for people unfamiliar with the product, upon seeing or hearing the name alone, to determine the qualities, kind, condition, or functions of the product going under that name. Nor is it possible to determine the type or kind of product that it is. Instead, LongLife suggests the potential effect of using that product, whether for the user, for the object on which the user uses it, for the larger product itself, or for all or some combination of all three. For example, a hypothetical name, Long-life Sox, conveys to the consumer nothing about the function or qualities of the sox, such as the material used in them or the quality of sewing that put them together. Nor does this name indicate a type or kind of sox (men’s, women’s, knee-length, ankle-length, argyle, or athletic). Instead, it suggests that the purchase and use of these sox would be economical because they allegedly would last a long time. In other words, the consumer must exercise thought or imagination to conclude that these sox are claimed to have a long life, possibly longer than sox without such a name. The


reader’s exercise of thought or imagination goes beyond any direct representation or recognition of the product’s qualities, functions, types, or kind. Obviously, this analysis was intended to show that Warren’s LongLife mark was suggestive, not descriptive or generic. I was actually giving evidence required by the imagination test, described in chapter 2. The more imagination that is required by the consumer to get some direct description of the product from the mark, the more likely the term is to be suggestive. The exercise of thought and imagination required of the consumer who confronts the product LongLife is carried one step further than recognizing that LongLife refers to endurance of the product. Warren’s LongLife, on its package label, indicates that the long life suggested by the mark is not only the long life of the product but also the long life of the vehicle in which the product is used. This package label reads: “Protects Aluminum and All Engine Parts.” Again, a hypothetical name can be illustrative. If there were a mark called Long-life Vitamins, it, like that of Long-life Sox, would suggest a potential effect of using the product. But, unlike Long-life Sox, the potential effect of using Long-life Vitamins is not on the product itself, for there would be little reason to want vitamins to have a long life, but on the user of the vitamins, for whom a long life is obviously more salient and desirable. Warren’s name, LongLife, suggests the long life of the product but also implies that there is also a long life for the vehicle in which this antifreeze is used, not just on the product itself. Prestone, in its claim that “long life” has but one generic meaning, missed this semantic point and apparently overlooked the imagination test. But even Prestone’s own package label directly indicates two different meanings of long life: 1) the product will last longer: Guaranteed cooling system protection for a minimum of 4 years or 60,000 miles. and 2) the engine will last longer since it is protected against rust and corrosion of metals: Exceeds all ASTM and SAE standards for corrosion protection. Protects all metals including aluminum. Maximum freeze-up/boil-over protection.


Prestone’s advertisement in Motor Trend (December, 1993) suggests the same thing: These coolants include new formulations of corrosion inhibitors that protect the cooling system for dramatically longer periods. Prestone’s advertisement in Non-Foods Merchandising (October, 1993) says the same: . . . a product specially formulated to provide guaranteed cooling system protection for 4 years or 60,000 miles. LongLife 460 also offers protection against rust and corrosion to all cooling system metals, including aluminum. Warren LongLife’s own press release (September, 1993) also supports that company’s dual attributes of the product. One attribute is the convenience to customers who “do not maintain their vehicles on a regular basis” and presumably would be convenienced by a product that does not have to be changed often. The other attribute is that the product provides “unsurpassed freeze-up and boilover protection” and provides extended protection against rust and corrosion to all cooling system metals, including aluminum, showing that Prestone could have understood that its alleged genericness of “long life” is compromised. Warren’s LongLife, therefore, suggests two separate meanings of long life, one for the product and the other for the product’s effect on the vehicle, calling into question any individualized descriptiveness of the term, LongLife, as Prestone had claimed. Defendant Prestone also cited advertisements of other companies that used “long life” to refer to a class of coolant/antifreeze that lasts longer than the conventional product, some going back as far as the 1960s. One such advertisement was used by Sierra Antifreeze, a wholly-owned subsidiary of Warren, in which Prestone claimed that Sierra used “long life” in the generic sense. The text of that advertisement was cited in Prestone’s brief, as follows: How does SIERRA Antifreeze compare to the so-called “long life” antifreezes? It is not clear why Prestone chose to use this reference, since the expression “so-called” clearly calls into question any claim that “long life” is an incontestable generic reference to a type or class of antifreeze/coolant product.


In my affidavit and trial testimony I also pointed out how the multiple meanings of the expression “long life” could be illustrated by a recent television advertisement for Energizer batteries. In this commercial, a man finds a gold lamp and rubs it. A genie comes out and offers him three wishes. The man first asks for wealth, and several bags of gold appear. He then asks to be adored by women, and he gets this wish also, as several beautiful women dance around him. The genie then reminds him that he has only one wish left and that he should be very careful how he uses it. The man pauses briefly then asks for long life. The famous Energizer bunny immediately appears beating his drum, and presumably going on forever. In this advertisement, “long life” clearly refers to the Energizer battery’s durability. The humor of the commercial lies in the confusion of the man’s request for his own long life and the genie’s grant of a battery with a long life instead, illustrating effectively the fact that “long life” can have multiple meanings which require imagination to decode. This was, in fact, the very point I was making when I argued that “long life” was more than generic or descriptive, especially when associated with an inanimate product. Using a recent, well-known television advertisement also added credence to the recency or modernity of the public’s perception of the multiple meanings of “long life.” Finally, Prestone had conducted its own NEXIS search of media references to the use of the words, “long” and “life” in association with coolants/antifreezes. This search claimed to have revealed 71 mentions of the words, “long” and “life” in association with coolants/antifreezes. My examination of Prestone’s own search, however, revealed that they had made certain overclaims. For example, I found that 24 of the citations were to the names of the competing products (most of which were Prestone’s own), reducing Prestone’s claim of genericness here to 47 remaining mentions. But that wasn’t all. An additional 17 mentions were of automotive products other than antifreeze, including spark plugs and transmission fluids, leaving a total of 30 rather than 71 mentions of “long life” coolants/antifreezes. Two of the alleged mentions were of “extended life” rather than “long life” and one mention was to “everlasting,” with no mention of “long life” at all. This reduced the total to 27 advertisements between 1987 to 1995. The fact that Prestone’s NEXIS search of “long life” revealed competing terms, “extended life,” and “everlasting,” for what was claimed to be the generic name of a class or type of antifreeze/coolant, opened the door for me to request Warren to conduct another NEXIS search, this time for “extended life” in association with coolants/antifreezes. This


search revealed 30 mentions of “extended life” in direct association with coolant/antifreezes and two mentions of the same expression in connection with brand names, suggesting strongly that Prestone’s claim that “long life” was the sole generic term for the class or type of antifreeze/coolant had considerable competition from other expressions. These two NEXIS searches show pretty clearly that these two terms, along with others, had been used at the industry level to describe that type or class of coolant product. Even Prestone’s own advertisements used “extended-life” to describe the industry’s term for this classification of the product. In fact, even its own catalog used the term “extended life” to refer to this particular type of antifreeze/coolant product. The advertisement submitted by Prestone reads: Prestone LongLife 460 is the only choice as a super premium extended-life antifreeze/coolant. This usage is also repeated in another Prestone advertisement: When you choose an extended life antifreeze, go with . . . Prestone LongLife 460. Other companies producing comparable products also used the term, “extended life” to refer to the class or type of antifreeze. Texaco Lubricants Company, in the minutes of its 1995 Western Distributor Advisory Meeting (August 15, 1995) said: It was also stated that the new Havoline extended life antifreeze will be aligned with current Prestone pricing. Likewise, a General Motors information bulletin, dated June, 1995, began: “Subject: New Extended Life Engine Coolant Known as DEXCOOL.” The first paragraph of that bulletin stated, “A new extended life engine coolant known as ‘DEX-COOL’ will be used in all General Motors vehicles.” The term “long life” is not the term used by Prestone, Texaco, or General Motors in any of their attempts to categorize the type or class of antifreeze/coolant. Teaching the jury The issues discussed so far were addressed largely in my report and deposition. Since it had been decided that the trial would be before a jury, attorney McLeay advised me that one of my tasks as an expert witness


in this case was not only to present my linguistic analysis but also to teach the jury how to understand trademark distinctions such as “generic,” “merely descriptive,” “suggestive,” “arbitrary,” and “fanciful.” In my experience, jury instruction about aspects of the law was not normally done by the linguistic expert witness. But the attorney thought it best that I do this task, so I did. For this, I used a chart, as follows:

Generic

Merely descriptive

Categories: class only of describes goods: characteristic or ingredient of goods Examples: CAR TASTY BREAD Referent products: car bread

Suggestive

Arbitrary

Fanciful

Suggests a characteristic or ingredient of goods

an arbitrary word as applied to goods

a made-up word

STRONGHOLD

APPLE

KODAK

nails

computers

photographic supplies

I pointed out that everyone knows immediately what the generic words mean. For merely descriptive words, we make an instantaneous leap to the meaning. Suggestive words make another cognitive leap beyond the descriptive category, the leap of imagination required by the imagination test described in chapter 3. In the ConAgra v. Hormell case (see chapter 6), the judge advised that if that mental leap is not instantaneous, then the word is suggestive. Arbitrary refers to common words that do not conjure up any mental reaction connected with the products. Any such juxtaposition is purely arbitrary. Finally, fanciful marks are not common words and have no apparent connection at all with the product. We decided that the best way to explain this further would be to give examples of real or imagined products that would illustrate each category of trademark protection. In my testimony, therefore, I used the following chart which I presented and discussed:


Generic

Merely Descriptive

Suggestive

hair gel car wash slacks

Styling Gel Brushless Car Wash Sports Slacks

Style Touchless Action Slacks

I testified that we all know what the products are in the “Generic” category. The names in the “merely descriptive” category contain adjectives plus nouns. The adjectives are, by definition, descriptive adjectives of well-known products or functions. The “Suggestive” category, on the other hand, contains words that could be descriptive, such as “Style,” but are not merely descriptive. They suggest something more. “Touchless,” for example, is somewhat like “Brushless,” but a bit more. The “Action” of “Action Slacks” suggests the ability of the wearer to be comfortable even during strenuous activity. My trial testimony began with Warren’s attorney asking me if it is possible for two separate words to be put together, making a new noun which has a different meaning from its component parts. I responded that this was quite possible and used, as examples, the words, “blackball,” and “whitecap,” which have meanings quite different from the adjectives and nouns that compose them as a compound. To teach the jury the important distinction between animate and inanimate uses of “life,” and to show that there are steps in the inferencing process, I presented a chart containing five illustrative drawings: 1. a baby, representing life; 2. an older man, representing a long life; 3. an automobile engine with the words, “life of engine” written on top; 4. a radiator with antifreeze being poured into it with the words, “extending engine life” on top of it; 5. a cutaway drawing of a car engine, showing the radiator with the words “engine longevity” next to it. I asked the jury to imagine how a small baby might indicate the meaning of life, then how and older man might be associated with a long life. Next I asked them to imagine how they could transfer the concept of animate life to an inanimate object, such as antifreeze, and how using that product in an automobile engine could give longer life


to that vehicle’s engine. I pointed out how these inferences came in small steps, from human beings to inanimate things. The use of these charts was preliminary to the presentation of my linguistic findings. Following the arguments noted above, I testified that “long life” refers to animate existence, except in its metaphorical uses, which, being metaphorical, could in no way be considered descriptive. Then I pointed out that “extended life” is not commonly used in reference to animate existence. For example, one extends credit, benefits, offers, authority, payment, time, or services. When the term “extended” is used in relationship to “life,” the inanimate quality of the word “extended” signals the metaphorical use of the following word, “life.” The fact that some elements of the antifreeze/coolant industry misused the term “long life” in place of more appropriate terms, such as “extended life,” “durable,” long-lasting,” “everlasting,” “enduring,” “permanent,” “protracted,” or “prolonged,” is evident from extant divided usage. Even the occasional use by manufacturers of “long life” in place of one or more of these terms does not render “long life” descriptive or generic. It remains metaphoric. Central to trademark cases is the question of whether or not consumers will be confused by similar or identical marks. Consumer survey specialists are conventionally called upon to conduct appropriate research with actual or potential consumers to find quantitative evidence to address this issue. Linguists frequently have something to say about how close the names are to each other in sounds, letters, words, morphemes, syntax, and semantics. But such discussion is not common when the issue is one in which the battleground is whether or not the name is generic or descriptive. That issue trumps confusion in most cases. I was also prepared to show how similar were the names LongLife and LongLife 460. Both parties used a capital “L” before both “Long” and “Life.” In both cases, the elements in the compound are alliterative. Both companies conjoin the two words into a single word, leaving no space between them. Both words are compound nouns derived from an adjective and a noun. In any case, Prestone’s own Answer and Counterclaim had already admitted that the term LongLife “is likely to cause confusion, mistake or deception as to the affiliation, connection or association with Prestone.” If Warren’s LongLife was confusing to Prestone’s customers, then Prestone’s LongLife must have been equally confusing to Warren’s customers as well. The issue of confusion never arose in the trial.


Conclusion of the case It is not often the case that trademark litigation occurs before a jury. The issues of trademark law are complex (see chapter 3) and difficult for lay persons to absorb. The issues of linguistic analysis are equally complex, of course, and there were certain risks in calling on me to testify. This explains my concerted effort to begin with the jury where it was, a crucial principle of pedagogy. My use of the simple charts and testimony outlined above was as basic as we could get it. Needless to say, much of my affidavit and deposition findings were simplified greatly at trial. Mostly I needed to get the jury to understand the imagination test. Since the issues of confusion and secondary meaning were not central, the problem of communicating these issues effectively was reduced somewhat. The trial was held in April, 1996 and after two weeks the jury recommended that Warren get $832,000 in actual damages and $839,360 in deterrent damages. After this judgment, Warren’s attorney, Bartholomew McLeay, next planned to request an injunction against Prestone to stop using the LongLife 460 name, stop advertising it, and withdraw the product from store shelves. He said that he also expected Prestone to appeal. I could find no published record that showed that an appeal was ever made.

6 Sounding Alike and Meaning Alike

During the last quarter of the twentieth century, microwaving became a stable part of the cooking done by busy families. Food producers rushed to take advantage of this technological advance and to develop new easy-to-prepare individual entrées, deserts, side-dishes, whole meals, and even lunches. One such microwave lunch was packaged in 213 gram, single-serving cups that could provide a quick meal in about a minute and a half. Two different food producers began to market such meals, leading to a trademark battle over their competing microcup lunch meals.

ConAgra, Inc. v. Geo. A. Hormell & Company Early in 1991, ConAgra, Inc. of Omaha, Nebraska brought a complaint against George A. Hormell & Company of Austin, Minnesota, claiming that Hormell’s use of “Health Selections” as the name of one of its new shelf-stable food product was a conscious imitation of ConAgra’s “Healthy Choice” trademark, and that this action would confuse and deceive the public as to the source and sponsorship of ConAgra’s products. Both companies are diversified food companies, producing many similar types of food products. Both parties consented to having the case tried before a US Magistrate Judge in Omaha. In the fall of 1988 ConAgra filed an application to register the mark Healthy Choice for frozen dinners made of meat, poultry, seafood, vegetables, and fruit. In March, 1990 ConAgra filed a similar application for shelf-stable products. At about that same time, John Alex, the attorney representing Hormell, anticipated a battle and called me to bring whatever assistance linguistics might have to bear on the case. The salient facts are these. In March, 1989, Hormell sought to regis69


ter “Health Selections” as one of its many shelf-stable products, having marketed a line of such products earlier under its trademark “Top Shelf.” On March 3, 1990 the US Patent and Trademark Office issued certificates of registration to Hormell for “Health Selections,” and no opposition was made by ConAgra. Prior to ConAgra’s introduction of its “Healthy Choice” frozen products, Hormell had manufactured and sold shelf-stable foods using the name, “Health Selections.” ConAgra made much of its even prior use of “Healthy Choice,” albeit for frozen food dinners, but the issue of who got there first with a trademark registration, as is often the case, turned out to be less salient to this litigation than other issues. Central to ConAgra’s lawsuit was the strength of its “Healthy Choice mark.” ConAgra contended that its “Healthy Choice” was suggestive, not arbitrary or fanciful. In contrast, Hormell argued that “Healthy Choice” was merely descriptive, indicating that the term informs the buyer of the purported quality of the product and that the use of this term would not help the consumer distinguish between products of different sellers. As noted earlier, one of the tests used to distinguish descriptive from suggestive marks is the “imagination test.” Once again, I recalled McCarthy’s observation: If the mental leap between the word and the product’s attributes is not almost instantaneous, this indicates suggestiveness, not direct descriptiveness. (1984: p. 492) The linguistic issues were thus framed. Was “Healthy Choice” merely descriptive or was it suggestive? If the Court were to determine that it is suggestive, ConAgra could prevail and there would be no need to examine other areas of language involved. If the Court were to rule that “Healthy Choice” is descriptive, the battle would be fought over whether or not the two marks, “Healthy Choice” and “Health Selections,” were so similar that consumers would be confused about them and deceived by the source and sponsorship of the two brands. Both ConAgra and Hormell retained linguistic experts to help address these issues. The standard procedure is for the Plaintiff’s expert to write the first report, which would then be rebutted by defendant Hormell’s linguistics expert. This provided an opportunity for different linguistic analyses and opinions, if any, on the same issue, to be brought forth.

Sounding Alike and Meaning Alike 71

It should be pointed out that the two linguists involved here had known each other for a long time and had supported each other in various ways in the past. I was Hormell’s linguist. For the sake of confidentiality, ConAgra’s linguist will remain anonymous here. ConAgra’s linguist produced a 48-page report, noting that there are two types of claims: (1) whether or not the two marks are linguistically similar, which he labeled a linguistic claim; and (2) whether people came to believe or acted as if they believe that the marks are the same, which he labeled a cognitive claim. This difference in linguistic vs. cognitive claims was held to be very important by ConAgra, since their linguist presented himself as a “cognitive linguist,” as opposed to Hormell’s conventional (presumably non-cognitive) linguist. This distinction was promoted throughout the litigation, possibly in an effort to impeach the testimony of Hormell’s linguist for allegedly not being “cognitive” and possibly to add the appearance (or perhaps reality) of additional academic strength to the case. On July 27, 1991, the linguists for both litigants were deposed, Hormell’s in the morning and ConAgra’s in the afternoon. Each was present at some or all of the deposition of the other. The same was true at trial on August 5, 1991, with both linguists present throughout. This was the only case in which I had ever had the experience of being faceto-face with the opposing linguist in this way. The depositions felt very odd, especially since we knew each other so well. Perhaps this odd feeling was a result of our being placed across the table from each other alongside the respective opposing attorneys, who were charged with being advocates for their clients. Although the linguist’s role is not that of an advocate, the proximity to the advocacy situation in the deposition event can easily confuse any expert’s need to be an objective outsider. At trial, where the rules of decorum are more strictly enforced, our mutual presence seemed less troublesome. The opposing positions of the two linguists in this case can be summarized in the four issues that they addressed: 1. 2. 3. 4.

Do Do Do Do

the the the the

two two two two

marks marks marks marks

sound alike? have the same grammatical structure? look alike? mean the same thing?

Following is a comparison of the testimony of both linguists on these four issues.


ConAgra’s linguist 1. Do the two marks sound alike? The report of ConAgra’s linguist noted that Health Selections consists of a one-syllable word followed by a three-syllable word whereas Healthy Choice consists of a two-syllable word followed by a onesyllable word, making them phonologically different in this respect. He also argued that initial segments of words and phrases are of more importance than final segments in reader recognition of those words and phrases, encouraging confusion of the product lines. This led him to point out that the two marks have quite different second words, a place in which the names were less apt to lead to confusion. To support the importance of initial elements of names, he explained that nicknames are normally formed from the beginnings of full names, citing the examples of Mike for Michael, Sam for Samantha, Rob or Bob for Robert, Bert for Bertrand, Di for Dianne, and Carol for Carolyn. He explained further that dialect variation tends to occur in consonants at word endings, as in the loss of “-r” in words like “car” and “cart” but not in words beginning with “r,” as in “red.” He also noted that in pronouncing word combinations, such as “first choice,” we never delete the first consonant of the second word, which is much more crucial to word recognition than is the last consonant in the first word, which can be deleted easily. He acknowledged that there is a quantitative difference in phonetic sounds between Healthy Choice and Health Selections but he believed it more appropriate to place special weight on the initial segments of the marks. 2. Do the two names have the same grammatical structure? The ConAgra linguist pointed out that both names employ the free morpheme “health” and that Hormell’s first word, “healthy,” is an adjective derived from ConAgra’s term, “health.” He noted also that both marks are proper nouns and that they function in sentences as full noun phrase units. He agreed that Health Selections contains a plural noun while Healthy Choice does not, but he opined that this difference was marginal and had no more syntactic importance than the adjectival nature of Healthy, an argument that supported the notion that the words had the same essential grammatical structure. He further stated that although the “sight, sound, meaning” trilogy makes no official room for syntax, syntactic differences ordinarily accompany differences in meaning and are therefore of interest here. Both names, Health Selections and Healthy Choice, function as proper


nouns and they function in sentences as full noun phrases. Even though one is a plural proper noun phrase and the other consists of an adjective and a singular noun, the fact that they are both proper nouns makes this difference marginal in importance. They both can appear in the same syntactic distribution in sentences. Therefore, nothing in the external syntax of these phrases reflects the fact that “health” can be used as a noun and “healthy” can be used as an adjective in other constructions or that “selections” is plural in the one mark while “choice” is singular. We cannot say, for example, “Health Selections are a good product.” As proper nouns, Health Selections and Healthy Choice are frozen structures and do not allow internal modifiers. This means that they have no internal syntax of consequence. We can say, “You have made a very healthy choice” but we cannot say, “Very Healthy Choice is a fine product.” In his deposition, however, when asked to analyze another well known-mark, Blue Ribbon (beer), he stressed the syntactic importance of the mark’s morphological origins. 3. Do the two marks look alike? The ConAgra linguist stated that one must distinguish how marks look when presented in advertising copy from how they might appear on product packages. He noted that on its package, “Healthy Choice,” unlike “Health Selections,” depicts a small figure of a jogger between the two words in its name. He claimed, however, that the visual difference in the packaging has very low perceptual salience, since he did not even notice it until it was called to his attention. The image of the jogger does not appear in advertising copy or even in the headlines of the advertising. He stated that it is not comprehensible linguistically. He also noted that there was a color difference between the two packages (ConAgra’s has green with white lettering while Hormell’s is white with green lettering), but concluded that these differences were neither memorable nor important. 4. Do the two marks mean the same thing? ConAgra’s linguist, distinguishing between literal meaning and significance meaning, argued that proper nouns do not have meaning in the way common nouns do. Literal meaning is relatively invariant among speakers of a language, while significance meaning can be expected to vary from speaker to speaker. Proper noun constructions have significance to consumers that is not related to the words that make up that construction. Speakers tend to remember the gist of what they hear and see, more than the actual sentence uttered or its literal meaning. To him,


the gists of Health Selections and Healthy Choice were one and the same, a significance meaning that is identical at the level of meaning most critical to memory. This single gist of both names is: items the selection of which can be expected to result in some sort of health benefit. The ConAgra linguist’s report summarized that the two names employ different words, that both use the morpheme “health” and that they are pronounced differently, but consumers will retain only the gist of the meaning of the two names, which he considered to be the same. An additional two-thirds of his report consisted of what the ConAgra linguist called “A New Approach to the Descriptive-Suggestive Distinction.” Although this was a creative and linguistically interesting set of suggestions about how to reconsider these categories of trademark law, it seems to have had no ultimate bearing on the trial itself. Nevertheless, it pointed out how linguistics might assist trademark law in ways that are explored further in chapter 17. Hormell’s linguist My testimony was grounded in a database consisting of media usage, commercial brochures, advertisements, and a list of previous trademark uses. I testified that the phrase “healthy choice” is used commonly in the English language. I based this in my Nexis search in the public media, where I found 57 articles in which the phrase was used in connection with nutritional or healthy food products, restaurants, labeling, nutrition programs, and so forth, and one article that used it in connection with role models for children on television. I noted, however, that even this one article dealt with health, in this case, mental health. I then discussed a number of commercial brochures and advertising materials, including one put out by the Washington Apple Commission about its program called “Healthy Choices,” a fold-out wall chart for schoolchildren advising them to “Make a Healthy Choice.” The Washington State apple growers’ “Healthy Choices for Kids” program was written up in a Washington Post article (June 14, 1991) and indicated that about three million schoolchildren throughout the county received this free educational kit. Even one of ConAgra’s witnesses had admitted that he had been aware for about two years of a General Foods nutritional program that used the slogan or phrase, “Make a Healthy Choice,” a registered trademark of General Foods. Evidence such as this about how the public is likely to perceive the mark is customarily received in trademark cases.


I also testified about my search of the database for instances of the words, “health” or “healthy,” and “choice.” I found 78 such marks, 30 containing “health” or “healthy” and six containing both “health” and “choice.” I concluded from this that the words, “health/y” and “choice” are commonly used in reference to products and trademarks for food products. 1. Do the two names sound alike? In earlier trademark litigation involving the cupcake product names “Little Dolly” and “Little Debbie,” another linguist, Jerrold Sadock, had very successfully used distinctive feature analysis to show that the sounds of “Little Dolly” were only 13 per cent different from those used in the name of his client, “Little Debbie.” By breaking down the structural, phonetic components of the consonants and vowels in the two names, Sadock noted whether or not the two words contained the same or distinctive features of consonantality, vocality, their relative fronting or backing, whether they were articulated at a high or low point in the mouth, whether their quality was coronal or grave, whether or not they were voiced, and whether or not they were strident. Following the conventional distinctive feature analysis, Sadock produced a complex chart that showed a plus or a minus for each of these features. In preparation for this trial, I followed Sadock’s lead and carried out a distinctive feature analysis of the sounds of “Healthy Choice” and “Health Selections.” I found that the two names produced 228 phonological features, of which 89 per cent were different. Thus, just as the procedure worked wonderfully in Sadock’s case to show that the two names were very similar, I believed that it had worked the same way in the current case to show that they were very different. The attorneys I worked with were impressed with this finding, but very skeptical that it could be effectively used in this trial. There is a fine line between being overly technical and being convincingly clear. They feared that the judge would be neither impressed nor convinced. The fact that distinctive feature analysis had done well in Sadock’s case did not sway their opinion here. So distinctive feature analysis was abandoned for other ways to show that the two words did not sound alike. Using a much simpler approach, I pointed out that of the 13 phonetic sounds in Health Selections, eight are totally different from Healthy Choice, and four are the same. One of these four is the “s” in “Selections” and the “s” sound represented by the spelling, “-ce” in “Choice.” This “s” sound occurred in very different places in those words. I concluded that for all intents and purposes, nine of the 13


sounds were different. I have found that representing difference with numbers or percentages is often a simple and effective way to convey such information to judges and juries. This 69 per cent difference conclusion was not as dramatic as the 89 per cent difference found using distinctive features, but it had to be sufficient in this case. In response to the ConAgra linguist’s use of nicknames and dialect reduction as evidence of the overarching significance of word beginnings, I responded that this is not always true (linguists are good at finding counterexamples). Since he used nicknames as examples, I noted that many nicknames are also taken from the endings of such names, such as Rick for Frederick, Tina for Christina, Gene for Eugene, Trish for Patricia, Shelly for Michelle, and Beth for Elizabeth. As for his claim that dialects, especially non-standard ones, habitually delete ends of words, I presented counterexamples that delete the beginnings of words, such as ’member for remember, ’cause for because, ’taters for potatoes, and ’lectric for electric. It is also the case that some dialect forms are not content with preserving the beginnings of words, as the ConAgra linguist claimed, and that they sometimes precede such words with still other sounds, as in “h” which precedes “ain’t,” yielding “hain’t” and “hit” for “it,” and the “-a” that precedes some verbs in certain dialects, as when it is added to “coming” and “running” to yield “acoming” and “arunning.” 2. Do the two marks have the same structure? I agreed, of course, with the ConAgra linguist that the two marks are both proper nouns which take the singular verb grammatically. But I did not agree with his effort to trivialize the importance of the fact that Health Selections differs from Healthy Choice in that one is a noun–noun compound and the other an adjective–noun compound. I argued that even though Healthy Choice and Health Selections are both proper noun constructions, their structural origins are not obliterated from the consciousness or memory of readers and listeners once the proper noun is formed. As for his statement that the “-s” grammatical inflection in Selections is made meaningless by the fact that it was part of a proper noun, I argued that the “-s” plural inflection still means something. Although it is not grammatically functional in the proper noun, it is etymologically present in that form. No readers can see the word Selections and still believe that there is only one item to be picked, even though they may well understand that there is only one brand name. As examples I cited other marks, such as Health Watchers, Con-


sumer Testing Laboratories, Pudding Treats, Yellow Pages, and Handy Wipes. Can readers believe that there is only one of these? I stressed that the only similarity between Healthy Choice and Health Selections is found in the morpheme, “health,” which they share. After that, everything is different, including the fact that “Selections” is plural. 3. Do the two marks look alike? I had determined that the way the two products looked, in this case package design and color, was very marginal to the type of linguistic testimony that I could offer. If there were issues of document design, print size or type, or other more orthographic concerns, a linguist’s testimony might have been more appropriate. For this reason, I avoided dealing with this question. 4. Do the two marks mean the same thing? ConAgra’s linguist claimed that there was no meaning difference between the two marks, that the gist of the meaning was the same. This could be taken to signify that the two marks operate as synonyms. Synonymy is a difficult area for a linguist to defend, since a very good case can be made that no words are completely synonymous. To address this issue, I decided to apply the conditions for synonymy to the facts of this case. The conditions for synonymy are that the pairs must: 1. have identical meaning; 2. be synonymous in all contexts; 3. be identical in all dimensions of meaning. It is not difficult to find examples of alleged synonyms which fail to meet at least one of these three conditions. “Peace” and “tranquillity,” for example, carry somewhat different meanings when occurring in the context of the word, “conference.” I pointed out that the adjective “Healthy” carries the meaning of “quality of being healthful” while the noun “Health” signifies an internal characteristic of the other word in the compound “Selections” and could be paraphrased as “selections made for health purposes.” I also noted that both “Choice” and “Selections” involve some type of picking of something, by somebody or something. With Choice, an animated subject/agent/benefactor makes the pick while with Selections, the animated subject/agent can be totally outside that semantic frame. I used the following examples:


“This choice is yours.”

The addressee can make it.

“This selection is yours.”

The selection was made by someone else, not the addressee.

“The gallery has a choice of paintings by Monet.”

The choice of what to look at can be made by the addressee.

“The gallery has a selection of paintings by Monet.”

The selection was made by the gallery, not by the addressee

Healthy Choice

The addressee can make it.

Health Selections

Someone else made the selections, not the addressee

I also pointed out that another meaning difference is found in the array of things to be chosen between or selected from among. Evidence of the way we use English shows that “selections” has a wider array of things to select from among than “choice” gives us to choose between. In fact, acceptable English usage recommends that we “choose between” two (or at least a small quantity of) things but “select from among” a larger quantity or array of items. I cited as an example the words to a familiar love song: “I find a broken heart among my souvenirs.” versus “I find a broken heart between my souvenirs.” “Between” in this citation indicates a more specific physical location, such as between the scrap book and the prom program. “Among,” in contrast, conveys a much broader array of possibilities, such as somewhere in the trunk of memorabilia or in the attic. I also used other sample sentences in which an English preference for “choice” or “selection” is clearly evident. “They don’t give you much choice (not selection) in the matter.” “It’s your choice (not selection): clean your room or be grounded.” “Young couples face difficult choices (not selections) these days.” “A woman should have a choice (not selection) when it comes to deciding what to do with her body.”


The narrow array offered by the word “choice” is also well illustrated in other contexts, such as: “The choice between good and evil” (a moral alternative) “Fielder’s choice” (baseball player’s option to throw the ball to one base rather than another) “Sophie’s Choice” (movie about a woman who has to choose between her two children) I concluded that there are two major meaning differences that exist between “choice” and “select”: 1. the person choosing is intimately connected as animated agent to the word “choice,” while “selections” exist outside the meaning of the beneficiary and are made by someone else; 2. the array of options is much larger for “selections” (and even for the singular form “selection”) than it is for “choice.” To be sure, some speakers violate these differences, using the words interchangeably, but even the remnants of changing usage offer insights into residual meaning. Unfortunately, during my testimony, I inadvertently violated the point I was trying to make when I said, “the word ‘selections’ indicates an even broader array of choices available.” I immediately caught myself and then tried to correct my slip, but the judge, in his decision, called attention to it with a certain degree of humor at my expense. Since the major argument of the ConAgra linguist about meaning was that the gists of Healthy Choice and Health Selections are the same, the Hormell attorney asked me if I agreed that the same gist existed here. I answered, “No, I would describe the gist of Healthy Choice as ‘you can make a healthy choice’ and the gist of Health Selections as ‘a number of products have been selected for you’.” I agreed that there is similarity of gist in “health” and “healthy” but not in the context of the second words in the two products’ names. Conclusion of the case On January 7, 1992, US Magistrate Judge Richard G. Kopf issued his Memorandum Opinion in this case. Among other things, he noted that linguists for both parties agreed that “Healthy Choice” was a laudatory mark. This helped settle the issue of whether or not this mark was


merely descriptive, since laudatory marks are generally considered just that. In any case, ConAgra, in previous litigation against Comsource Independent Food Service, had claimed that their opponent’s product name, “HealthChoice,” was a weak, descriptive mark. If ConAgra claimed that “HealthChoice” was merely descriptive, it would be very difficult to argue that “Healthy Choice” was not. The judge found in favor of Hormell on ConAgra’s claim of trademark infringement, ruling that the “Healthy Choice” mark is relatively weak because it is more descriptive than suggestive. His words, echoing the imagination test and supporting this conclusion were: In this case I conclude that the “mental leap” between the words and the product’s attribute’ is almost instantaneous. He added that “Healthy Choice” provides a message of “the healthy characteristics of the goods simply by looking at the mark.” The judge also found that the marks sound significantly different, supporting my testimony, but that they mean essentially the same thing, agreeing with the ConAgra linguist’s opinion rather than my own. My argument may have been too complex or too arcane for the judge’s taste. I hate to think it was that wrong, but maybe I should have been less detailed and minute in the way I presented my position. The judge cited other cases in which word marks were not held to be confusingly similar even when they convey the same or a similar meaning. He also ruled that the trade dress and other visual images of the two products are not confusingly similar and that the possibility of consumer confusion is fairly small and not caused by trademark relevant reasons. In addition to the linguistic testimony, both sides presented consumer surveys. The judge decided that the surveys on both sides were technically flawed and therefore he discounted them as a factor in the case. This made the linguistic testimony even more central to his decision. ConAgra appealed this decision to the US Court of Appeals for the Eighth District on November 9, 1992. On March 31, 1993 the appellate court denied the appeal and affirmed the district court’s opinion, calling it “well reasoned.” Hormell, though victorious, eventually discontinued its “Health Selections” line, but preserved the microwave cup lunch idea under another popular Hormell line called “Dinty Moore”.

7 Descriptiveness: Nouns and Modifiers

One might think that it’s an easy call to decide whether or not something describes a product or suggests something quite different. But trademark lawyers fight endless battles over such matters. Clearly “Raisin Bran” is a descriptive mark, since the name describes the major ingredients of the cereal product, raisins and bran. Other noun–noun base trademarks that have been held to be merely descriptive include “Beer Nuts,” “Chap Stick,” “Litter Basket” (trash receptacles), “Pudding Treats,” and “Vision Center.” But when the name conjoins a noun base with a modifier, such as “Culture Ripened” (coffee), “Continuous Progress” (educational materials), and “Easyload” (tape recorders), does the resulting mark have the same descriptive quality? A trademark case involving the expressions, “woodroast” and “wood roasted” tested this point.

Woodroast Systems, Inc. v. Restaurants Unlimited, Inc. This issue was an important part the battle fought when a company providing restaurant services, Woodroast Systems, Inc. (also known as Shelly’s Woodroast) in Minnesota, filed a complaint against Restaurants Unlimited, Inc., doing business as Palomino Euro-Metro Bistro, which was also providing restaurant services in Minnesota. Plaintiff Shelly’s Woodroast brought a complaint on January 22, 1992, claiming service mark infringement, federal unfair competition, and deceptive trade practices. Defendant Restaurants Unlimited then filed for a summary judgment and on May 19, 1992, Judge J. Doty held that there were unresolved material fact issues concerning whether or not “Woodroast” was generic and about the possibility of confusion by consumers, precluding a summary judgment about the origin of the claims. He also ruled 81


that Woodroast was entitled to preliminary injunction relief. In short, the case would have to be settled by a trial. Shelly’s Woodroast had been using the marks “Woodroast,” “The Original Shelly’s Woodroast and Design,” and “Original Woodroast Cooking and Design” in its restaurant services business, claiming “strong secondary meaning among the public.” The company further claimed that the use of these marks by others is likely to cause confusion, mistake, and deception. In particular, Shelly’s Woodroast objected to Restaurants Unlimited’s use of “Wood Roasted,” on its menu and advertising, claiming that this company had intentionally infringed Woodroast’s mark. The original complaint included a menu from the Palomino Euro-Metro Bistro which contained a salad item called “Wood Roasted Vegetables,” and three selections of “Wood Roasted Seafood,” one of which was “Wood Roasted Oven Prawns.” From this complaint it was clear that the attorneys for Shelly’s Woodroast believed that their oneword, “Woodroast,” was equivalent to the Palomino Bistro’s two-word “Wood Roasted,” on its menu. It was equally clear that there would be a controversy over the significance of nouns versus modifiers here. In his Answer and Counterclaim filed on March 2, 1992, Restaurant Unlimited’s attorney, Bruce Little, claimed that “wood roasted” is a generic phrase and not subject to trademark or service mark protection when used in connection with the business of providing restaurant services. He also noted that his client’s use of the phrase, “wood roasted” is descriptive of defendant’s goods and services. He described it as a method of cooking in which meats are slowly roasted in wood-burning ovens, noting also that there are many other restaurants in the country that use the expression “wood roasted” on their menus. Woodroast’s answer to defendant’s counterclaims, on March 18, 1992 admitted that its Woodroast mark identifies its unique method of preparing foods in its patented wood-burning oven and expressed no knowledge of other restaurants that use the expression, “wood roasted” in their menus. Over a year after the suit was filed, the attorney for Restaurants Unlimited called me to assist him on any language issues that might be involved in his defense. Subsequently, he sent me the complaint, affidavits of Shelly Jacobs (the president of Shelly’s Woodroast), and the affidavits of several other company officials. I was at first unclear about what possible relevance such affidavits might have for my eventual linguistic analysis of the terms in question. But I soon learned that what the participants say about the issue also constitutes language data that can have linguistic significance to the case.

Descriptiveness: Nouns and Modifiers 83

My initial response In June, 1993 I sent attorney Little a letter outlining some of my ideas related to my future analysis. As is common in such cases, the months following were occupied with many depositions, briefs, and much preparation. Five months passed with no communications from Little. Then, in November, 1993, he requested that I refine my letter into an affidavit, number my paragraphs, and create a “less academic” tone. He also urged me to continue searching for the terms, “woodroasted” and “wood roasted” in any sources I could think of. Attorney Little’s request that I take a “less academic” tone troubled me a bit at first. Academic expert witnesses think about new data and problems from the perspective of their fields. They take out their tool bag of theory, research, and analytical procedures and look for the most appropriate tools to deal with the issue presented to them. This is how academics usually think. The product of this thinking, not surprisingly, looks academic. The attorney who calls them does so because they are specialists in their field, so it would seem that to present an academiclooking result would be predictable. But I had faced this issue of being “too technical” in the past and I was prepared to try to deal with it here as well. The attorneys’ perspective is a different one from that of the academic. They see farther down the road. They have a pretty good sense of what will work in a trial and what will not. They may have appeared previously in front of certain judges and know what impresses them and what does not. They also know about the comprehension level of juries. So the specialist’s first job is only the beginning. It is not wrong to make the first reportings “too academic,” but it is prudent to listen to the attorneys and modify the resulting reports, affidavits, and testimony to make them palatable and understandable to the people who will ultimately receive and try to understand them. When linguists follow the instruction to be “less academic,” they sometimes have to pay the consequences, especially when another linguist is used on the other side of the case. This opens the door to the other linguist’s possible criticism that one’s analysis is not linguistically sophisticated enough. This is the common problem of applied linguistics in any context, whether the application is made to fields such as law, education, medicine, diplomacy, or advertising. Since such audiences are not linguistically sophisticated, one must begin with where they are, not where the linguist is. This means that the most abstract concepts must be made as concrete as possible, that the latest developments and controversies in the field may have to be avoided, and that the more complex aspects may have to be simplified, even to the extent


of omitting things that one would say to a different audience of peers in one’s own field of study. Adjusting to one’s audience is a central principle of sociolinguistics and effective communication. Expanding the database If a linguist is to determine how a word is used, what it means, and how it varies, it can be useful to expand the database beyond the individual words disputed in the lawsuit. This can provide additional evidence on which the expert’s opinion can be based. Following good linguistic practice, linguists can search the broad database for how the item is employed in real-life contexts and ground their conclusions on how the term is actually used. Therefore, in response to Little’s request for me to continue searching for the term, I asked for copies of a list of restaurant names and menus using the expression, “wood roasted” as well as an electronic search of any mentions of “wood roasted,” “wood roasting,” and “woodroast” available in the print media during the preceding five years. After plaintiffs were deposed, I also requested copies of their depositions. Meanwhile, I examined dictionary entries for various permutations of “woodroast,” and “wood roasted.” Data from restaurant names In addition of Shelly’s Woodroast, five other unrelated restaurant names in the database included the phrase “wood roasted” in their titles: Cluckers Wood Roasted Chicken, Henpecker’s Wood Roasted Chicken, Rollo Pollo Wood Roasted Chicken, Kenny Rogers’ Roasters Wood Roasted Chicken, and Red Hot Hen’s Wood Roasted Chicken. In contrast, only one restaurant could be found to use “woodroast” in its title: that of the plaintiff, Shelly’s Woodroast. Data from menus Twelve restaurant menus were located that contained 38 items in which the phrase “wood roasted” in some orthographic form, was used, including “woodroasted,” “wood roasted,” “WOOD ROASTED,” “WOODROASTED,” “Wood-roasted,” and “Wood-Roasted.” Related phrases included “wood-fired roasted,” “wood rotisserie,” “OVEN ROASTED,” “SPIT ROASTED,” “SPIT-ROASTED,” and “wood flamed.” In all, 38 different entrée items followed the word “wood roasted” in the names of their menu items. In contrast with Shelly’s Woodroast, all 38 menu items used the term in the past tense, “-ed” form. Also contrasting with Shelly’s Woodroast, all used “woodroasted” as an adjective describing the following seafood, meat, or vegetable items on their menus.


Data from the electronic search of press articles The way that the public media uses language plays an important role in measuring either the commonality or the uniqueness of a trade mark. Since electronic searches are available, I requested that one be made for the common nouns, “woodroast,” “woodroasting,” and “wood roasted.” The resulting search turned up 44 articles from national and local newspapers and business magazines in which common noun uses of variations of “wood roast” appeared. These citations were found in national newspapers including: Los Angeles Times, Boston Globe, Miami Herald, Chicago Tribune, Saint Louis Post Dispatch, Louisville CourierJournal, Wall Street Journal, and Seattle Times. Articles from business journals were found in Restaurant News, Restaurant Business, Business Journal of Milwaukee, Nation’s Restaurant News, Nation’s Business, Restaurateur, and Entrepreneur. Media publications more local to Minneapolis included the Star Tribune, St. Paul Pioneer Press Dispatch, Midwest Living, Minnesota Monthly, City Pages, Skyway News, and Where/Twin Cities. In addition, I was provided video and audio-taped franchise promotion by Kenny Rogers’ Roasters which used the terms as common nouns. Data from dictionaries I searched for “woodroasted” in every lexicographic source I could think of, including Websters second and third New International Dictionaries, the tenth edition of the Merriam Webster Collegiate Dictionary, Dictionary of Regional English, Dictionary of American English, Oxford English Dictionary, Random House Dictionary, and American Heritage Dictionary. “Woodroasted” appeared in none of the dictionaries either as a proper noun, common noun, adjective, or verb. I then searched for “-roasted” as an element in compound words and found “dry-roasted,” “halfroasted,” “ill-roasted,” “sweet-roasted,” “twice-roasted,” and “wellroasted,” but no “woodroasted” or “wood-roasted.” My search for the compound element “-roasting,” yielded even fewer words, with nothing relevant to the term in point. My search for “wood” as the first element of a compound in the Oxford English Dictionary produced “wood-walled” (1595) and “wood-fired” (1956). The latter suggests that from the associated terms, “gas-fired,” and “coal-fired,” cooking fuels might have generated the usage applicable to restaurant preparation. There were no references to “gas-roasted” or “coal-roasted” however. Data from affidavit and deposition testimony The way in which people involved in the case use the words in question is often interesting and helpful. In the affidavit and deposition


testimony of various officials of Shelly’s Woodroast, I found another 27 uses of the usage of “wood roasted” and “wood roasting.” My database had now become rather large. The small amount of data originally presented for analysis, “Woodroast” and “wood roasted,” had now expanded from the two expressions in dispute to 117 uses in various contexts, as well as variations in dictionaries. Linguistic analysis and affidavit presentation The meaning of “wood roasted” Dictionaries offered no entries for “woodroast,” “woodroasting,” or “woodroasted.” Definitions of the terms would have to be found in the meanings attributed to these words by those in this specialized field and, of course, by the context of how they were used. “Woodroast cooking” was defined in Shelly’s Woodroast menus as “a new style of food preparation.” In contrast, Entrepreneur magazine (August, 1989) calls wood roasting merely a “form of cooking,” omitting the word, “new.” Nation’s Restaurant Magazine (May 2, 1988) calls it “a method of slowly roasting meats, poultry and fish in special wood burning ovens.” Midwest Living (June, 1989), in an article specifically about Shelly’s Woodroast, refers to “woodroasting à la Shelly,” indicating a special type of wood roasting food preparation. Skyway News (February 1988) calls wood roasting “a style of cooking,” which includes marinating and slowly roasting. The Seattle Times (September 7, 1991) states, “wood roasting was the beginning of all cooking.” An article in Nation’s Restaurant News (May 13, 1991) describes roasting in general, with a focus on the recent trend of restaurants to offer wood roasting. The Kenny Rogers’ Roasters audio tape, which was produced to promote new business franchises, says that this is the way chicken was cooked 200 years ago. Citations from the restaurant business articles make it clear that wood roasting is a method of cooking and that it is not, as Shelly’s Woodroast menu claimed, a “new style” of food preparation. One more confusing source of information about whether or not wood roasted defines the manner in which the entrée is cooked came from Sheldon (“Shelly”) Jacobs, himself. In his deposition of April 13, 1993, Jacobs testified that he did not believe that “wood roasted” is an adjective when that expression precedes the food that it describes. On the other hand, Jacobs offered the observation that if the words “wood roasted” were placed after “chicken”, as in “chicken wood roasted,” the words, “wood roasted” would indeed work adjectively. Later, in the same deposition, Jacobs continued, “I do not object to someone using


the word ‘wood roasted’ as descriptive of a style of food that they’re cooking or an item of food that they’re cooking.” He was then asked, “Do you have an objection to anybody using the phrase, ‘wood roasted,’ if they use it preceding the name of the food item?” Jacobs replied, “I do not object.” It is difficult to follow Jacob’s reasoning here. He said that he had no objection to the adjectival use of “wood roasted.” At the same time he denied that “wood roasted,” when it appears before “chicken,” is adjectival. At another point he agreed that he would not object to “wood roasted” being used before “chicken.” This was probably a nightmare for his lawyer but clearly advantageous to the defendant. One principle for discovering, describing, and defining the meaning of words comes from the distinction between proper and common nouns. In print form, proper nouns are capitalized; common nouns are not. The distinguishable capitalized “W” form of “wood roasted” occurs only in connection with Shelly’s Woodroast in this database, always as a proper noun with no space between the two elements. It became clear that Shelly’s Woodroast used “Woodroast” as a proper noun while all others use the word as an adjective describing a particular food item. Syntax of the menus If, as the words on the Shelly’s Woodroast menu indicate, “Woodroast” cooking is a unique style or method of cooking, one might also expect this to be incorporated into the way the menu items are presented. In every menu collected, except for those of Shelly’s Woodroast, the word “wood” is combined, using a hyphen or a word separation, with “roasted.” A central principle of effective communication is to be as brief and orderly as possible (Grice 1975). The more syntactic slots that are filled in a menu item, the less crisp and communicative the message will be. This principle argues for the use of a minimal number of optional syntactic slots available in the menu item, namely, the food item itself plus the most descriptive and salient other slot or slots. Examination of the syntactic slots in many menu items revealed that there are five such slots: (1) self-congratulations, (2) method of cooking, (3) style of food, (4) the food item, and (5) serving modifications. A hypothetical example of a menu item that fills all five slots is the following: 1 Our best

2 grilled

3 Italian

4 cheese

5 sandwich

Analysis of restaurant menus indicates that these slots are fixed in the order just given, and slot reversals can produce ludicrous-sounding menu items in the same way that reversals in word order can produce


ludicrous-sounding English sentences. It is also the case that not all of the menu slots are required. Just like English syntax, some slots are optional, but at least one slot is obligatory. In this case the obligatory slot is number four, the food item itself. It is obvious that the self-congratulatory slot (1) has the least salience and descriptive power in a menu. Of the remaining slots, style of food (3) and serving modification (5) are useful but still optional for effective communication. In the case of entrées, however, no serving modification (5), such as “sandwich,” is absolutely necessary. A menu’s representation that this is an entrée makes this the unmarked form. The method of cooking (2) is the most central optional slot to be filled. Too little information is equally problematic in communication. In the case of restaurant menus, describing the food item alone gives the reader too little to go on. In a fine restaurant, a menu item labeled “chicken,” for example, does not communicate a well-cooked, careful meal. The menu items, then, had five slots for information. Slots 1, 2, 3, and 5 are optional (indicated below with a +/- sign). That is, putting information into those lots is optional. In all cases, however, slot 4, the food item itself, is obligatory (indicated with a + sign). Although all menus of restaurants other than Shelly’s followed this formula, only four are used here as illustrations.

Slot 1 (+/-) selfcongratulation

Slot 2 (+/-) method of cooking

Slot 3 (+/-) style of food

Slot 4 (+) food item

Slot 5 (+/-) serving modification

Items on Palomino menu: 1. (none) iron grill 2. (none) grilled 3. (none) wood roasted

Italian (none) (none)

ham chicken stromboli

sandwich sandwich (none)

Items on Biba menu: 1. our wood famous roasted

(none)

planked salmon

(none)

Items on Cluckers menu: 1. our wood famous roasted

(none)

chicken

(none)

Items on Tuttaposto menu: 1. (none) wood roasted

Sicilian

seafood sausage

(none)


The Shelley’s Woodroast menu fits into this syntax sequence in a slightly different way: Slot 1 (+/-) selfcongratulation

Slot 2 (+/-) method of cooking

Items on Shelly’s Woodroast menu: (none) (none)

Slot 3 (+/-) style of food

Slot 4 (+) food item

Slot 5 (+/-) serving modification

Woodroast

chicken

(none)

The point of this slot and filler analysis was to emphasize that the most common practice in English is for descriptive adjectives to precede the nouns they modify. Other variations from the above syntactic formula could be unclear, if not ludicrous. For example, “Iron grill sandwich ham” would be confusing, if not meaningless on a menu. Although the English rule of descriptive adjectives preceding the nouns they modify is not invariant, the act of placing the adjective after the food type in the more rigid menu formula invites the question of “wood roasted what?” By using “Woodroast” as a proper noun, indicating ownership and style of food, Shelly’s Woodroast effectively switched the proper noun from the slot of method of cooking to the slot of style of food, the slot where “Italian” is found in the Palomino menu and where “Sicilian” is found in the Tuttaposto menu. At issue here, of course, was to demonstrate that the use of “wood roasted” in menus other than the Shelly’s Woodroast menu is an adjective describing the obligatory slot of method of cooking (slot 2) as opposed to Shelly’s Woodroast, which left this method of cooking slot empty and placed Woodroast in the style of food slot (slot 3). The menu syntax of the plaintiff and defendant was simply not the same. Syntax of the restaurant names The five restaurants with names containing “wood roasted” follow a slightly different syntactic formula: Slot 1 business name

Slot 2 method of cooking

Slot 3 food specialty

Cluckers Henpecker’s Rollo Pollo Kenny Rogers Red Hot Hen’s

Wood Wood Wood Wood Wood

Chicken Chicken Chicken Chicken Chicken

Roasted Roasted Roasted Roasted Roasted


This syntactic formula is not shared with Shelly’s Woodroast, however: Slot 1 business name

Slot 2 method of cooking

Slot 3 food specialty

Shelly’s Woodroast

(none)

(none)

For Shelly’s Woodroast, the company business name says it all, with no slot for method of cooking and none for food specialty. By its own definition found on its own menu, the name “Shelly’s Woodroast” conveys this style of food preparation for an unspecified number of different meat, poultry, and seafood items, and not the method of cooking it. Syntax of the use of wood roasted by the media As noted earlier, none of the 44 articles from national and local newspapers and business magazines used the common noun “woodroast” at all, and they also did not use it as an adjective. All used “wood roasted” or “wood roasting” as ways to describe the method of cooking a specific food. The recency of “wood roasted” and its variants One of the claims in a trademark infringement case is that one party makes use of an expression that was used first by another party. In the case of “wood roasted,” therefore, it is useful to determine just how old this expression is. If it is recent and other parties appropriate it as their own, a case for Shelly’s Woodroast might be made. In this case, the concept of roasting with wood is a very old, unmarked form, since in former days there was virtually no other way to roast. It became less used when newer different fuels began to be common, causing those who want or need to refer to the older method of roasting to have to mark that form with the word, “wood.” It is this newer process of marking food preparation by means of roasting with wood that became the focus of this case. One way to determine a word’s relative age is to examine any occurrences of it over time. Since there is no record of “woodroast” in the lexicographic records, I turned to its component parts. “roast” and “wood,” for clues. According to the Oxford English Dictionary, the verb “roast” can be found in the English language since at least 1297. Other citations are


given spanning the centuries before the invention of modern household stoves and ovens. Obviously before modern days, roasting was accomplished over wood or coal (or possibly peat) fires. For this reason there would have been no reason to mark the only method of roasting in the early citations. Only in recent years, when this method of cooking needed to be marked as “wood” roasting because of the availability and preponderant use of alternative fuels, would there be any need for the term, “wood roasting,” which was now created to distinguish this method from the current, everyday, unmarked electronic or gas-oven cooking. Marking theory is commonly used in linguistic analysis. Linguistic features that are statistically dominant are often referred to as “unmarked.” Areas of language that differ from the commonly used and accepted forms are considered “marked forms” because of such variation. For example, in 99 per cent of the languages whose word order has been studied, grammatical subjects precede their objects, causing language typologists to consider syntax with the object appearing first to be the “marked” form. I felt that this approach might help explain the situation in this case as I examined the dictionary, menu, and media representations of the relevant terms. Modern dictionaries continue to define “roasting” in ways that are consistent with the pre-modern period. Webster’s New Collegiate Dictionary (ninth edition, 1983) defines the verb “roast” as follows: 1. To cook by exposing to dry heat (as in an oven or before a fire) or by surrounding with hot embers, sand or stones. Webster’s New World Dictionary (1960) includes the following in its definition of the verb “roast”: 1. Originally, to cook (meat, etc.) over an open fire or in hot ashes, etc. From these definitions one can conclude that despite the influence of modern kitchen stoves and ovens, roasting still includes open fire methods, as any backyard barbecuer knows. The New World Dictionary definition further supports the conclusion that the original, unmarked form of roasting was over an open fire, and none of these definitions excludes roasting with wood as a defining characteristic. The fact remains, however, that modern references to this return to a method of cooking with wood require marking it as such. Otherwise there would


be no distinctiveness from modern, conventional roasting with electric or gas fuels. Punctuation marks provide still another clue to the emerging status of “wood roasting.” New words formed by combining two regularly used terms have three punctuation possibilities: 1. The two words are used separately and contiguously. 2. The two words are combined by hyphenation. 3. The two words are combined into one word. It is common in the creation of new words combined from old ones, for the process to follow the above sequence, from 1 to 3. When commercial air service was started, companies like United and American first called themselves Air Lines, then later, Air-lines and, finally, Airlines. Until the last stage is reached, it is common to experience divided usage between these three punctuation practices. All three of these punctuation possibilities are found in the database, with the hyphenated “wood-roasted” being the most frequently used form (44 of 55 articles). Punctuation analysis of this term, therefore, leads to the conclusion that “wood roasted” is still in divided usage in terms of punctuation, supporting its modernity as a marked adjectival describing this return to the ancient method of food preparation that involves roasting foods with fire. Other possible terms to fill the lexical gap If the claim of Shelly’s Woodroast required other parties to find a different expression for wood roasting, the possibilities are very slim at best. The absence of a word at a specific structural place in a language is called a lexical gap. For example, in Swedish there is no single term for “grandfather” or “grandmother.” English has such gaps in meaning that some other languages easily fill, and vice versa. For example, English kinship terms do not include specific terms for the relationship of persons to their father’s cousins. It has no widely understood or used names for the collection of dust under a bed, the grass strip between the sidewalk and the curb in cities, or the chunk of accumulated snow and slush that falls from under a car onto the roadway, thus becoming a traffic hazard. However trivial, these may be considered lexical gaps. We have to use circumlocutions to describe them, since no single word does so. Creating or inserting a particular word at a particular place in language structure is often carried out by means of the processes of lexical


substitution and lexical transformation. Some languages lexicalize a contrast, creating new words for the new concept. Other languages grammaticalize the contrasts, as in the comparative (-er) and superlative (-est) suffixes used in English. One can create a new word to fill the lexical gap created by elements of this new trend in roasting, which sometimes is also referred to as “urban rustic cooking.” Alternatively, one can call upon the already available word inventory to fill that lexical gap. For example, a certain Amazonian tribe called upon its existing word inventory to create the term “fire canoes” to refer to the first airplanes that they saw. If one is to find alternative words for “wood,” one might try searching in Roget’s Thesaurus. But Roget’s list is short and the possibilities all create new meaning problems. Roget lists “kindling,” for example, as a possible synonym for “wood,” but kindling refers to small pieces of wood used to start fires, not nearly the same meaning as wood used for roasting food. “Log” suggests large pieces of untrimmed wood of any description, more suitable for lumber mills than for cooking. The only truly descriptive term available is “wood.” As for alternatives to “roasted,” Roget’s help is not much better. “Heated” is too broad and suggests warming rather than cooking. “Singed” suggests being burned or badly cooked. “Burned” is usually a negative concept in food preparation. “Blazed,” “flamed,” and “fired” focus on the fire, not the cooking. “Grilled,” “broiled,” “fried,” “baked,” “stewed,” “toasted,” “smoked,” and “barbecued” are all substantially different from the method of cooking called “roasting.” The only truly descriptive term available is “roasted.” Since there is a finite number of fuels with which to cook (gas, electricity, wood, coal, microwave, etc.), the inventory of choices to re-mark the formerly unmarked form of wood roasting is very limited. If one wishes to identify the food preparation process at issue here, one is limited to roasting, which best describes the cooking method, and wood, which best describes the fuel. Summary From the preceding analysis, it became clear that Shelly’s Woodroast used “Woodroast” as a style of food, not a method of cooking, whereas other parties, including the defendant, Restaurants Unlimited, used “wood roasted” as a method of cooking. Shelly’s used “Woodroast” as a proper noun, in contrast with the adjectival employed by all of those whom Shelly claimed were infringing his mark. Wood roasting is ancient in origin, an unmarked term that, when replaced by other fuels


in modern times became the marked form, now revived under the label of “wood roasted” cooking methods. In the old days when wood was the only available fuel, there was no need to mark “wood,” but today there are other fuels with which to roast, creating the lexical need to mark cooking with wood as “wood roasted.” The recency of this revival of wood roasting can be seen in the current divided usage about how to punctuate “wood” plus “roasted.” Complicating matters even more for Shelly’s, there appear to be no acceptable alternatives in the English lexicon for saying “wood roasting.” Conclusion of the case The case was settled before trial. Although the terms of the settlement were sealed, Restaurant Limited’s attorney hinted to me that his client was very pleased with the results. He told me, however, that my affidavit played an important role in the settlement negotiations. One of the things that a linguist must learn about work in this area is that the end results are often not totally clear. One might want to learn a lot more than I was able to learn about the interactions that took place during settlement, but confidentiality issues often make this impossible. It would have been helpful to learn which of the linguistic arguments presented carried the most weight, so that they could be emulated in later cases. But, unfortunately, this was not possible here. There are undoubtedly many good reasons why settlements are sealed in this way. Such secretiveness may have implications for other future or current lawsuits, based on the precedents reached in this one. It is difficult, however, to imagine the prohibition of information in most other areas of any open society. One can think of military strategy, undercover law enforcement, the inner workings of high-level government decision-making, and perhaps a few other instances where it is common and acceptable to withhold information from the public. Commerce and the laws regulating it, on the other hand, may hardly seem to fit the pattern of required sworn secrecy. Should omission of full disclosure favor the province of business and not the rest of a society? It would appear that the law’s desire to control and own the expressions of language, as in the trademarks themselves, is sometimes extended to controlling and owning the language of the negotiated results of litigation – another area of battle between trademark law, language, and society.

8 The Meaning of a Patronymic Prefix

The courts have ruled that adding a suffix to a descriptive word will probably not render that word non-descriptive. McCarthy cites the addition of the suffix, “-ize” to a base form, as in Flexitized (collar stays) and Nylonized (nylon-treated fabrics) as examples of a suffix that does not deprive the term of a descriptive connotation (1984: 460). McCarthy does not mention prefixes, but there is no reason to think that prefixes and suffixes do not operate the same way in terms of trademark law. Linguists lump both prefixes and suffixes into the same category, called affixes (which would include infixes, although English does not make use of these). The major question in the following case is whether or not conjoining a prefix found in another company’s proper noun name with a different common noun constitutes a trademark infringement. Of particular interest here is that the prefix in question is, historically at least, a patronym commonly found at the beginning of some Irish and Scottish names.

McDonald’s Corporation v. Quality Inns, International, Inc. In the fall of 1987, a large hotel chain, Quality Inns International, made public its plan to create a new chain of inexpensive hotels to complement its other market brands. The name of this new hotel was to be McSleep Inns and they planned to open some 200 McSleep franchises within three years. Three days after this initial announcement, the McDonald’s Corporation, the famous fast-food marketer, sent a letter to Quality Inns alleging trademark infringement and demanding that Quality Inns not use the proposed McSleep name. Quality Inns responded by seeking a declaratory judgment that a McSleep mark would not so infringe, that it did not allege false description of origin, 95


and that it did not infringe upon or violate any common law rights that McDonald’s might have. Preliminary maneuverings lasted for almost a year, culminating in a seven-day trial in Baltimore during July, 1988. Just a few years before this case, McDonald’s had successfully sued a small restaurant in New York called McBagel’s. A judge ruled that using “Mc-” in combination with a generic food noun did indeed constitute infringement. Although it might seem that this settled the issue, there was one big difference here. Quality Inns proposed to use “Mc-” in combination with a non-food product, for a hotel chain that would not even contain a restaurant. In fact, McSleep Inns were planned to be built near fast-food restaurants so that patrons would have easy access to them. One of the less profitable aspects of the hotel business is their restaurants, and Quality Inns wanted to reduce such a liability. McDonald’s countered that even without an attached restaurant, the name of a hotel, McSleep Inn, would likely cause confusion and that Quality Inns had deliberately selected the name “McSleep” to trade on McDonald’s reputation and goodwill. McDonald’s cited evidence of their own survey report, which said that 100 per cent of American children aged two to eight know and recognize their icon, Ronald McDonald, a statistic comparable only to their recognition of Santa Claus. Precedents about “confusion” are many. The marks of Notre Dame University and Notre Dame cheese have been permitted to coexist, as have the marks for Bulova watches and Bulova shoes, and Alligator raincoats and Alligator shoes. Because these marks occur in widely different fields of business, the public presumably was judged not to be able make a connection between these two quite different commercial entities. An important legal trademark principle known as the Aunt Jemima Doctrine concerns related but non-competitive markets. Black’s Law Dictionary defines the Aunt Jemima Doctrine as follows: The principle that a trademark is protectable not only from an act of copying, but also from the use of any similar mark that would likely make the buyer think that the item bearing the similar mark comes from the same source as the trademarked item. (p. 127) In the 1917 case from which this doctrine got its name, Aunt Jemima Mills Pancake Batter held the trademark against which Aunt Jemima Syrup was found to infringe, since syrup and flour were found to be food products that are commonly used together. The question in McDonald’s v. Quality Inns was not one of products in competing areas

The Meaning of a Patronymic Prefix 97

but rather, whether or not there is some connection between the food and lodging industries that, like the Aunt Jemima case, hold them together. Also important in this case was the issue of the genericness of the prefix, “Mc-.” When a trademark becomes so strong that it becomes the generic label for goods or services associated with another corporation, the owners of the trademark may still protect the mark from use by others, but they have very little control over its use in everyday language. The public sometimes loses sight of the original corporate sources of certain terms, such as “escalator,” “aspirin,” and “cellophane,” especially when the owners of those marks surrender their rights to them. On other occasions there are words that become generic while the owners are still protecting them, such as “xerox” for a photocopying process, “kleenex” for a facial tissue, “q-tips” for a cotton swab, “cuissinart” for a food blender, and “thermos” for a container that keeps food hot or chilled. The prefix, “Mc-,” shows a mixture of both popular and corporate use. It is most common for trademark cases to focus on the use of a protected word or phrase, not on a prefix. “Mc-” is neither a word nor a phrase. It is a bound morpheme that does not occur in isolation. It functions as a derivational prefix, and one that is very active and productive as well. At trial, McDonald’s pointed out how in one of their marketing campaigns, their advertising icon, Ronald McDonald, had traveled around the country and actually taught children how to add the “Mc-” prefix before many different words, such as “McFries,” “McShakes,” and “McBest.” McDonald’s vice-president for advertising testified that the purpose of this campaign and its resulting commercials was to create a “McLanguage” that was specifically associated with McDonald’s. The relevance of such a program as a form of language planning will not escape the attention of linguists and other social scientists. To counter McDonald’s claim that they own complete rights to the formative, “Mc-” when it is used before a generic word, Quality Inns presented four arguments. First, relying on the Aunt Jemima Doctrine, it argued that there is no likelihood of confusion between McSleep and McDonald’s. Second, it maintained that the uses were not in competition: that McDonald’s marks, developed for the fast-food business, do not preclude the use of McSleep in the lodging business. These two defenses were debated as legal issues. For the third and fourth defenses, Quality Inns called on me, as a linguist, to help them.


Quality Inns’ linguist Quality Inns’ third argument concerned common third-party uses of the prefix, “Mc-,” when used before common nouns. Quality Inns argued that it should not be denied the right to use McSleep since there was already in common use evidence of this coming from a proliferation of words containing “Mc-” combined with generic words. Quality Inns’ attorney, Lawrence Hefter, helped me search telephone directories, credit reports, and media usage. The attorneys made phone calls and visited businesses for the purpose of verifying these uses of “Mc-.” In addition to what the attorneys gleaned from their initial list of telephone directories and credit reports, we made use of a computer search system called Nexis and a national clipping service to find a large corpus of words containing the “Mc-” prefix. This produced 150 articles in newspapers and magazines, all written recently between March and July, 1988. All were citations in which the writers used the prefix “Mc-” with a descriptive or generic term. Sources included magazines such as Forbes and Time, major newspapers such as Washington Post and Los Angeles Times, local newspapers such as Maple Heights Press, and technical publications such as Rubber and Plastics News. The majority of these uses, all unrelated to McDonald’s or to the fast food industry, included the following: McArt McBaby Factory McBook McCaviar McChekov McCinema McDigest McDome McDrive-Thru McDuck’s McEverything McFashion McFood McHairpiece McHealthcare McHistory McTelecast McTelevangelism McZippy’s

McHospital McJobs McJournaled McLaw McLife McLifestyle McLube McLunch McMagazine McMail McMarketing McMedia McMedicine McMiz McMovie(s) McNews McTelevision McWidgets McEconomics

McOffice Supply McOil Change McPaper McParticles McPost Office McPrisons McProgram McPulitzer McRather McReads McRobot McShopping McStory McSurgery McSweater McTax McVideo McYear


In categorizing these and other third-party citations by their meaning types, I separated them into seven divisions (see Lentine and Shuy 1990). 1. Proper names: McGruff (the crime dog), MacThrift (a budget office supply store with a Scottish motif). 2. Alliterative patternings based on a proper name: Jim McMyth (football player Jim McMahon) and McVeto McKernan (a name given by protesting union workers to Maine governor John McKernan). 3. Acronyms: McDap (Mason County Drug Abuse Program) and McRIDES (Morris County Rides, a ride-sharing program). 4. McDonald’s product names, which were not analyzed here for obvious reasons. I never disputed the fact that the “Mc-” morpheme has a strong association with McDonald’s. Instead, I tracked its use into other quite different domains. Of the 150 articles gathered, I excluded 56 of them that were directly about McDonald’s restaurants and/or about specific McDonald’s products. 5. Macintosh Computer Products or related businesses, such as McTek (a computer discount store specializing in Macintosh products) and McToy (computer accessory for accelerating computer processing time). McDonald’s and Macintosh had worked out their differences over these names in earlier agreements, making references to them unuseful in this context. 6. Parodies of a fast-food product or service: McChow Mein (a hypothetical name for a Chinese fast food restaurant) and McMania (a hypothetical drive-through therapy clinic). 7. Other words that carried the meaning of basic, convenient, inexpensive, and standardized. This turned out to be a very long list that included such terms as McArt, McBook, McCinema, McEconomics, McJobs, McLube, McMail, McMedicine, McNewspaper, McPrisons, McTelevangelism, McPaper, and McYear. Quality Inns’ attorneys used my categorization in their fourth argument, for which my testimony at trial was most central. Quality Inns maintained that the prefix “Mc-” had by now become generic, and had entered the lexicon of English with a recognized meaning of its own. That meaning is most notable in category 7 above: “basic, convenient, inexpensive, and standardized.” If anything identifies McDonald’s process of making, selling, and advertising its hamburgers, it is these four characteristics. Fast food is basic (not gourmet). You can find McDonald’s stores virtually anywhere in the world (convenient). The


product is relatively inexpensive (at least in most places at the time of the litigation). If you’ve had one McDonald’s hamburger, you know that the next one will be just like it (standardized). There was little debate about the first six categories but it was necessary to go through the process of categorizing all “Mc-” references in order to arrive at the crucial part of my testimony, category 7. In my trial testimony I described my rationale for using this database as follows: I wanted a large body of data because I did not want to rely only upon my own intuitions. I wanted to have an empirical base from which to do the analysis. In an analysis of any kind one can sit at one’s desk and think about it and say “here’s what I know to be true because I am a native speaker of this language and here is how it works, and I know.” Or, you can do experiments by, say, bringing people into a laboratory, sitting them down and asking them questions. Or, one can do empirical studies with actually existing natural events, in this case, articles in magazines. . . . Here we have a body of citations and evidence from all over the country and in which the people who wrote them didn’t know that they were going to be looked at with this case in mind. The range of meanings found in the 94 citations at first seemed very broad but was later narrowed down to these four basic meanings. It was also clear that this bound morpheme was active and productive. Writers attached “Mc-” to words referring to everything from automobile tune-ups to major surgery. I also found examples in which one use would spawn a subsequent use, such as McPaper (referring to USA Today) which then gave birth to McTelecast (related to the USA Today newspaper). It was clear that this bound morpheme, originally a Scottish and Irish patronym, had undergone, over time and with the nudging of McDonald’s massive marketing, the processes of lexical shift and lexical generalization. Whatever “Mc-” once meant about family ancestry, it now also meant speed, efficiency, consistency, and basic standardization, for this is how contemporary writers were using the morpheme to convey the meanings in their articles. Not all the coined words carried all four meanings. For example, the budget office supply store MacThrift conveyed the meaning of inexpensive, but not necessarily that of efficiency; while “McLifestyle” was used by the press primarily to refer to the qualities of speed and efficiency through easy access, but not


necessarily inexpensive. At least one and usually more of these four meanings fit each of the 94 citations. The important aspect about the 94 citations was that not one of them was about McDonald’s or even about hamburgers or fast food. In reaching my decision about whether a “Mc-” citation meant speedy, efficient, consistent, or standardized, I grouped together key words and phrases in the articles that fit these designations. For example, an article in Forbes magazine was titled “McArt.” It spoke of the “mass marketing” of art stores that were open seven days a week and “chains” that indicated how convenient it is to purchase these standard, mass market art products. The expressions used in the articles usually defined the writers’ own meanings of their uses of “Mc-.” The title of an article in The California Law Review, called “McLaw: Lawyering for the Masses,” describes the easily accessible and inexpensive basic legal services (consultation fees of $20–$25) that were then cropping up around the country. Expressions such as “near-omnipresence,” “franchise legal clinics,” “serving the masses,” and “drive-in windows,” occur frequently in the article, with no mention of McDonald’s, fast food restaurants, or hamburgers. One sentence in this article that suggests basic services is: “McLaw attorneys lack the prestige, comforts, and salary that come with other types of private practice.” If one were to remove “Mc-” from this sentence, the rest of it would be nonsense, for attorneys certainly do not lack prestige, comforts, and good salaries. The presence of a single bound morpheme creates a contrast in meaning from that which we might otherwise expect. Similarly, syndicated columnist Erma Bombeck used “Mc-” in the sense of mass-marketing when she referred to a “McStory on the paper’s front page” in her syndicated column about the preoccupation of the media with the wardrobes of Nancy Reagan and Raisa Gorbachev. Bombeck related that she read such a McStory, never mentioning fast food or McDonald’s. Nor did Bombeck bother to put quotes around the expression, suggesting that she did not feel obliged to alert readers to any unusual use of it. By pointing out the importance of context, I hoped to make clear that McDonald’s claim that “Mc-” always means McDonald’s is not the case. At trial I went though these articles point by point, observing that the prefix “Mc-” acts like any ambiguous lexical item, in that a reader or hearer must rely on context to determine which of the possible senses the author intends. I emphasized that meaning is flexible and determined by context. I used the word “green” as an example in the sen-


tence, “Give me the green!” If this sentence is spoken by a speaker holding a bank teller at gun point, it means something quite different from what it would mean when an artist purchases oil paint at an art supply store. In the sentence “The new mechanic was still a little green” we understand that “green” refers to inexperience, primarily from clues given by two other words in that sentence, “new” and “still.” I also pointed out that the context of the sentence alone does not always disambiguate the meaning of “green.” In the sentence “The young sailor was green” we cannot know for sure whether he was seasick or inexperienced. In such cases we have to look for larger context clues found in the surrounding text. I also made the point that there were several factors about the prefix “Mc-” that suggest that anyone reading the media articles I had cited could easily dismiss the possible association with McDonald’s. By placing the prefix in the context of an actual usage situation unrelated to food, in this case, hotel marketing, it is unlikely that one would associate it with McDonald’s restaurants. I argued that context clues provided by Quality Inn’s print marketing materials provided the same type of disambiguating clues that were cited in the illustrations of the use of “green.” These marketing materials included the Quality Inns fourbrand, stylized logo that contextualized it as relating to hotels, not to hamburgers. Each hotel franchisee was required to have the words, Quality Inn, placed beneath each McSleep Inn logo on its exterior signage. I pointed out that the combination of these elements acted in much the same way that it does in the prototypical sense of “green” in the sentence: “The girl went green at the thought of drinking sour milk.” Even though it may be true that the primary sense of “green” is a color, its use gives rise to the metaphorical extension of it to the sense of being nauseated. That is, we quickly move from the literal level and seek the metaphorical meaning when it becomes apparent that the literal meaning is not relevant. This metaphorical process is what I found overwhelmingly in the media citations using “Mc-.” It would appear, in fact, that the imagination test was appropriate here. A imaginative leap must be made to achieve the intended meaning. It is not possible for a linguist to say exactly what consumers understood by “Mc-,” nor did I even try here. The data that a linguist works with are outside of the sender’s intentions or the receiver’s understandings. What the linguist uses is the language itself, what it could mean, and what is possible for consumers to understand. As in other trademark cases, surveys were taken to determine what consumers said about what they thought. McDonald’s hired an advertising firm to


survey the perception of the public to the “Mc-” formative. Surprisingly, the survey was conducted at a McDonald’s office, suggesting a marked contextual bias in itself. Some of the results of the consumer survey carried out at the McDonald’s office that were reported at trial included the following: A kid’s product . . . a product children will like better because it’s associated with McDonald’s. Reliable, at a good price. Prepackaged, consistent, fast, and easy. Processed, simplified, has the punch taken out of it. Junk food, processed, not real, pre-made, uniform, cheap, bland, a gimmick, etcetera. These consumer comments agreed, in general, with the senses determined by my analysis of the media usage of “Mc-.” Despite the occasional negativity of the consumers (20 percent of those surveyed had very little good to say about the product), one of McDonald’s own documents took a more positive view: By virtue of extensive advertising and sales effort and expense and maintenance of the highest standards of quality and service by McDonald’s, “McDonald’s” and its “Mc” formative marks have come to be so distinctive and well recognized that the vast majority of consumers upon seeing the marks identify them with McDonald’s. It is the sheer market presence of McDonald’s that keeps its name and all it stands for in the consciousness of consumers. In this, the above quotation from McDonald’s is quite accurate. Secondary meaning is achieved through the massive expenditures on advertising and marketing, and secondary meaning often trumps everything else in trademark cases. In contrast, the media writers that I found had used “Mc-” in such a way as to suggest that the prefix had come develop its own meaning. An exclusive connection to McDonald’s was fading about as fast as its original patronymic meaning had slipped out of awareness. In fact, its use in these articles was often more pejorative than positive. It appeared that the prefix had broken loose from association with McDonald’s and had taken on a life of its own, with the original literal connotation becoming considerably less strong. Language has a way of doing this. Different ways of pronouncing “Mc-” were never a part of the lin-


guists’ presentations in this case. Attorney Hefter advised me, before deposition and trial, that the Quality Inns people had been pronouncing it “muck,” while the McDonald’s attorneys and witnesses had tried to stick with a pronunciation that carried a less negative connotation, “mack.” Early in the trial it became clear that the judge, perhaps unconsciously, had begun to follow the Quality Inns pattern, himself calling it “muck” regularly. I would assume that this was all a kind of subliminality game that had little to do with the facts of the case, but I followed Hefter’s advice and called it “muck” myself. This seemed to be the proper pronunciation of “Mc-” anyway, as far as I could tell, since “mack” differed by being the obvious way to pronounce the prefix, “Mac-,” not “Mc-.” Following the usual procedures in trademark cases, I was deposed by opposing counsel, who shortly afterward called their own linguist to rebut my proposed testimony. This brought about a very interesting set of events, including that linguist’s deposition and the testimony of both of us at trial. McDonald’s linguist In his deposition, the opposing linguist identified himself as a theoretical linguist, in contrast with other types of linguists, such as neurolinguists, psycholinguists, and sociolinguists. He repeatedly pointed out that theoretical linguistics has no relevance when determining the meaning of the prefix “Mc-” and that any opinions he might have are from his stance as a native speaker of English, not as a professional linguist. Later in his deposition he also claimed that theories of semantics and pragmatics were also irrelevant to the goals of the case. The only way to discover the meaning of the “Mc-” prefix, he noted, was to ask people what they thought it meant. He agreed that sociolinguists deal with the meaning of language in various social contexts but that linguistic theory deals with the meaning and use of words outside such contexts. Since he was a theoretical linguist, for him context apparently played no role in the matter. When confronted with the uses of “Mc-” by the media, the opposing linguist agreed that the context of the articles led to the reader’s understanding of the meanings of “Mc-” as basic, convenient, inexpensive, and standardized. He held firm, however, that such meanings were “merely” contextually defined and that “Mc-” had no meaning in itself. He claimed that context is not a determiner of the meaning of an individual word, apparently disagreeing with the findings of specialists in


pragmatics, whose work shows how inferential and indirect meaning is determined by context clues. Words in English, he claimed, have an innate core meaning independent of context, and even words with multiple meanings still have an innate core meaning. When asked about the meaning of “McLaw,” he testified, again as a layman but not as a linguist, that he knew that it had something to do with law but did not know what meaning the prefix “Mc-” was intended to convey. He agreed that “McArt” meant that it was mass-produced and standardized and that “McThrift” could be understood to mean inexpensive. Perhaps not wishing to seem to agree with the findings of my analysis, his objection then became that most English nouns have no more than three or four meanings and that he had been shown copies of my early draft notes in which I had indicated some 27 different meanings of “Mc-.” His information came from the discovery of notes that I had made when I was just beginning my analysis. Later, as I delved further into the matter, I had synthesized, relabeled, and narrowed down those 27 terms used to classify these meanings to only four: basic, convenient, inexpensive, and standardized. At the request of McDonald’s attorney, I had made my notes available and the other linguist used them in a way for which they never had been intended, as my final product rather than my tentative and developmental one. From this I learned something very important about the danger of preserving my early thinking and notes. The odd thing about the opposing linguist’s disagreement with me was that my findings inadvertently supported his thesis that words have no more than three or four meanings. In this case these four meanings were basic, convenient, inexpensive, and standardized, the same ones to which he agreed in his deposition cross-examination. The perhaps unintended gist of his testimony was that he was actually in general agreement with my analysis of the four meanings of “Mc-” as used by writers in the media. He agreed that context determined the meaning of “Mc-” in each case. His major disagreement was not with my findings, but that a theoretical linguist could reach them. In his view, linguistic theory could not get him there. During the trial, the judge summarized the differences between the two opposing linguists as follows: I think both Dr. Shuy and Dr. [name withheld here] performed the very same functions and performed the very same analysis. Dr. Shuy said it was within his discipline and Dr. [name withheld] said it


doesn’t fall within the discipline of linguistics . . . a difference of opinion. In this conclusion, the judge overlooked an important distinction that the opposing linguist had made. He did not say “it doesn’t fall within the discipline of linguistics.” He made it very clear that his opinion was that such analysis does not fall within his own expertise as a theoretical linguist. But even this is debatable. It is true that many of today’s theoretical linguists tend not to use inferences from corpora or experimentation to reach their conclusions about how language works (even though at trial it was revealed that the opposing linguist appeared to have actually done so in his earlier doctoral dissertation research). It would seem difficult indeed to claim that the application of linguistic knowledge about semantics, pragmatics, speech acts, and grammar is irrelevant to a trademark case for, in doing so, one would have to say that there can be no application of linguistic knowledge to anything. If this is the case, some might wonder why anyone would study this field at all. It is true that linguistics is, as sciences go, a relative newcomer on the academic scene. It is also true that linguistic theory has undergone rapid changes since the 1960s and that early attempts to apply whatever version of theory had been in vogue during the last few decades has sometimes resulted in failure, especially in the area of language teaching (Newmeyer 1983). But, as Newmeyer and other theoretical linguists continue to conclude, this does not mean that theoretical linguistics has nothing to say about how it can help resolve problems or lead to improved understanding of language interactions in educational, medical, or law settings. In direct contrast with the opposing linguist’s claim in this case, respected theoretical linguist Newmeyer says, in fact: Improved understanding of the interaction of grammatical and conversational principles in discourse might aid the already extensive work in progress that attempts to apply linguistics to questions that arise in the law. (1983: 157) The opposing linguist’s claim that meaning can only be found by asking people what they think something means also deserves comment. As noted earlier, there are at least three avenues open for discovering meaning. The senders of the message may report that they have an idea of what they mean. The receivers of the message may


report what they believe to be the meaning of a message. Both can be asked, of course, but even then we have only their self-reports of what the sender meant or what the reader or listener comprehended. Selfreported data have been questioned by many academic disciplines. We can never get into a person’s head to find out for sure exactly what they meant or understood. We have only their reports of it, whether or not these are accurate. It is the third perspective, that of the text itself, that is the subject of the linguist’s attention. The linguist examines the message to determine all the reasonable possibilities that language offers the reader or listener about what the text could have meant and how it could have been understood. There is no leaping inside anyone’s mind in this kind of work. Nor is it possible to make any absolute claim about the actual intention of the sender or the actual comprehension of the receiver. The text alone is the evidence of what alternative language use is possible to have been meant or understood. Thus our respective testimonies left the judge with no dispute about the correctness of both linguists’ conclusions but, rather, about the differences in our academic specializations. Trying to impeach the testimony of witnesses on the other side is an old and common technique used by attorneys. I have faced it on other occasions, as in the litigation between ConAgra and Hormell (chapter 6), when the opposing attorney seemed to imply that the qualifications of his linguist as a “cognitive linguist” trumped my own as a plain old linguist. It is unfortunate when experts occasionally are sometimes placed into the trap, perhaps unconsciously, of helping become advocates for the side that they represent. This is a role that only the attorneys should play. If the linguist for one side disagrees with the analysis done by the opposing linguist, those differences should be pointed out clearly. Differences between competing definitions of what a “real” linguist is may be appropriate for discussion at academic meetings, but seem to have less place in the courtroom. The quality of the analysis and conclusions are the most salient issues. In fairness to the opposing linguist in this case, however, it should be pointed out that he did not say that “linguistics” could not lead to the same conclusion that I made. His exact words were that he, as a theoretical linguist, could not reach it. It is left unclear whether this was because theoretical linguists are not accustomed to doing applied work, because they do not find such work very interesting, or because theory has no bearing on actual language use. If it is the latter, it should be pointed out that even Noam Chomsky, clearly the leading linguistic


theoretician of our time, has made a case for the relevance of linguistic theory to its application in language teaching: Surely the teacher of language would do well to keep informed of the progress and discussion in these fields, and the efforts of linguists and psychologists to approach the problems of language teaching from a principled point of view are extremely worthwhile, from an intellectual as well as a social point of view. (1966: 52) If linguists would do well to keep informed of theoretical matters as they apply it to language teaching, it is not a great leap of imagination to understand that such application can also be made to other areas of life, such as trademark disputes. Conclusion of the case This was a bench trial and the judge delivered his written decision in favor of McDonald’s some weeks later, on September 16, 1988. The surveys used by both sides appeared to be somewhat inconclusive about the issue of confusion. Quality Inns’ argument that the hotel business was not in competition with the restaurant business did not prevail. Nor was the proliferation of the use of “Mc-” in the public media apparently enough to sway the judge’s opinion. The judge seemed somewhat baffled by the competing testimony of the two linguists, despite their agreement about the current meanings of the “Mc-” prefix. In his ruling, the judge held that: 1. 2. 3. 4.

the marks using the prefix “Mc-” had not become generic; the marks were not fanciful and entitled to protection; there was a showing of likelihood of confusion; the holder of the mark with the prefix “Mc-” was entitled to enforce its mark in the related area of lodging; 5. the injunction against the use of the term “McSleep” would be issued without qualification as efforts to reduce confusion had been shown to be unsuccessful; 6. but the case was not an exceptional one warranting the award of attorney fees. The underlying theme of the judge’s decision was that McDonald’s had spent millions of dollars on advertising and marketing their name, prefix included, and secondary meaning won the day. A recent book, Fast Food Nation, notes that McDonald’s is the largest employer in


America. One out of eight Americans has worked at McDonald’s at one time or another. It is also among the largest buyers of beef, potatoes, pork, and poultry. The judge had a valid point about secondary meaning. As noted earlier, however, one can still wonder why it is that the expenditure of money can determine who can have ownership of a word, much less a prefix. So did the prefix “Mc-” actually become generic? The words of Ladas seem appropriate here: When a trademark has ceased to indicate origin, that is, when the public has come to use the trademark as a generic term indicating the natural class of a product, the public is no longer interested in the origin of such a product and it is not deceived when a competitor offers the same product originally coming from the owner of the mark. On the contrary, to continue protecting the owner in the exclusive use of such a word that has become a generic term is to grant him a commercial monopoly in a word not in the public domain, and it is to prohibit free competition by other suppliers of the same product. (1975: 1168) Since this trial, McDonald’s has gone on to apply for legal trademarks for many other words, including McSpace Station, McTime, McFamily, McMom, McSmile, McTravel, McMemories, McBaby, McProduct, and McBunny. It appears that this huge company has a monopoly on the prefix “Mc-,” and that the public domain will have to get along the best we can without using it in its generic meaning. Meanwhile, Quality Inns International, now operating as Choice Hotels International, renamed its proposed McSleep Inn hotel chain Sleep Inn, and has been highly successful with it. Interestingly enough, this chain is generally thought to be basic, convenient, inexpensive, and standardized, even without using the “Mc-” prefix. Many people believe that Sleep Inn was probably a better name anyway.

9 Sounding, Looking, and Meaning Different

The birth control pill has helped revolutionize family planning. Not only does it prevent unwanted pregnancy, but one of its active ingredients, progestin, is said to help prevent cancer of the ovaries and uterus. By the 1980s, however, the other active ingredient, estrogen, began to be suspected of causing breast cancer. This issue was researched by a Dutch oncologist, Dr. Michael Cohen, who, in conjunction with AMR Pharm Holland (a wholly-owned subsidiary of Applied Medical Research, Ltd. in the US), developed an estrogen-free contraceptive, which he called B-OVal. In it, estrogen was replaced with the hormone melatonin, secreted from the pineal gland of the brain, which plays an important role in the reproductive cycle of females. Melatonin inhibits the monthly cycle of lactation. Dr. Cohen’s theory was that when pregnancy does not occur, a residue of cells sometimes remains in the breast and can lead to cancer. His idea was that by substituting melatonin for estrogen, cancer development could be prevented.

AMR Pharm Holland, B.V. and Applied Medical Research, Ltd. v. American Home Products Corporation In October, 1988, AMR Pharm Holland filed its application with the US Patent and Trademark Office to register B-OVal. In May, 1990 American Home Products Corporation filed opposition to this application, stating that American Home was the owner of the registered trademarks, OVRAL and LO/OVRAL, which had been in use as the mark of their oral contraceptives earlier than AMR Pharm Holland’s use of its own mark, B-OVal. American Home further alleged that AMR Pharm’s marks were used for the same pharmaceutical products in the same trade channels for the same users as its own OVRAL and LO/OVRAL, that consumers 110

Sounding, Looking, and Meaning Different 111

would be confused by the two products, and that one party’s marks were likely to be thought to be endorsed by the other party. AMR Pharm Holland denied American Home’s claims, pointing out that its B-OVal is estrogen-free and has no known side-effects whereas OVRAL and LO/OVRAL contain estrogen, which is alleged to cause serious adverse reactions, including pulmonary embolisms, coronary thromboses, and cerebral hemorrhages. AMR Pharm also responded that its B-OVal mark is not likely to cause confusion, mistake, or deception because its consumers can get its product only through prescriptions written by trained physicians who are the ones who make up the actual market for the product. Physicians can readily distinguish such differences and are even required by law to know the material differences. In any case, AMR Pharm added, the trademarks of the two parties are sufficiently distinctive to justify their peaceful coexistence within the same market. The usual answers, counterclaims, and motions followed. By early 1993, the trademark case was on hold, pending negotiations involving the licensing of AMR Pharm’s B-OVal product. In the spring of 1996, I was notified by counsel that the Court was soon to hear a breach of contract issue related to this case, after which the trademark case would be tried. I was scheduled to be deposed, but this event was canceled. I was never informed what happened after this, but I was led to understand that, as far as I was concerned, my involvement in this case was then over. I point this out for two reasons. One is to make clear to linguists that we are sometimes ill-informed about what is going on in cases in which we are involved. Many times, after my testimony in a civil or criminal case, I have had to call the attorney with whom I had worked to learn the final outcome of the case. Experts who have spent hours on a project like to have closure and learn how things eventually ended up. To attorneys I commonly point this out and advise them that it is only considerate to provide this information to those who worked hard on their cases. To this day I do not know what the breach of contract part of the case was all about, whether it prevented the trademark part of the case from ever happening, whether some kind of out-of-court settlement was reached, or whether the attorneys decided that my continued involvement would not further their case. It would have been nice to know these things. As it turned out, the totality of my involvement was my affidavit testimony. If I had been deposed and had testified at trial, I would have presented the following analysis which undergirded my affidavit.


The linguistic dissimilarities between AMR Pharm’s product, B-OVal, and the product names of American Home, OVRAL and LO-OVRAL, stem from the differences in the way these names are pronounced, the differences in the way these names look, and the differences in the meanings conveyed by the names of these different products. Dissimilarity of sounds In the 1990 edition of Physicians Desk Reference (PDR), we find that the manufacturer of OVRAL and LO/OVRAL indicates that its products are to be pronounced as follows: OVRAL: LO/OVRAL:

öh’vral löh – öh’vral

From this we learn that stress is placed on the second syllable of OVRAL, that its first vowel, “o,” is to rhyme with “slow” and “go,” and that the last vowel, “a,” is to be pronounced the same as it is in the words, “all,” and “fall.” Since B-OVal had obviously not yet made it into PDR, representatives of AMR Pharm explained that it was to be pronounced as follows: B-OVal

biy-’ohvuhl or biy-’ohvahl

The first syllable was to be pronounced to rhyme with the word, “bee.” Stress was to be placed on the first syllable of “-OVAL,” the vowel pronounced to rhyme with “go,” and “slow.” They were undecided on the pronunciation of the last syllable, suggesting that it consumers might say it to rhyme either with “full,” or with “fall.” As in the ConAgra v. Hormell case (chapter 6), I suggested that a distinctive feature analysis and comparison might be made. Again, the attorneys expressed doubt that this would be understood by the judge. There is a complex line between presenting one’s linguistic expertise in a way that other linguists might approve or making a simpler case that does not seem to be so technical to non-linguists. The expert witness’s quandary seems always to require both proving one’s expertise and, at the same time, being clear to the trier of the fact. This is, in fact, the quandary of all applied linguists, no matter in what area they try to communicate their findings. The attorneys were cautious about my appearing to be overly technical, so I agreed to take a more simplified and unit-by-unit phonetic approach, avoiding even the technical symbols of phonetics as much as possible, as follows.


1. The “vr” consonant cluster The PDR pronunciation guide places the stress on the second syllable of OVRAL and indicates that the vowel of “o” is pronounced to rhyme with the words, “bone,” “know,” and “beau.” So far so good. The difficult part about this pronunciation guide, however, is the idea that English speakers can combine the “vr” cluster at the beginning of the second syllable, “-vral,” without inserting an epinthetic schwa vowel, “uh,” between the consonants “v” and “r.” English does have morphemes containing the reverse, “rv” cluster, which is limited to the ends of words, as in “swerve” and “observe” (note that since the English spelling system has a prohibition for ending words in “v,” a final “e” is added after the cluster). But morphemes that might otherwise have a “vr” cluster insert a schwa vowel, “uh,” between the two consonants, as in “overall.” The word, “over,” is a very productive prefix in English, with 175 such prefixed words listed in the American Heritage Dictionary, but there are no English words with a “vr” consonant cluster in syllable initial position that would help to suggest how the “vr” cluster might be pronounced in OVRAL. It is most likely that consumers who speak English natively would pronounce OVRAL with an “uh” vowel between the “V” and the “R.” AMR Pharm’s product, B-OVal has no such problem, since there is no “r” after the “V” in its name. The sound (and therefore the spelling) of “v” occurs there between vowels. 2. The vowel represented by the letter, “a” The PDR pronunciation guide for OVRAL and LO/OVRAL apparently makes no phonetic symbol distinction between the various pronunciations of the letter “a” (such as the differences between “map,” “day,” and “father”). Since the second syllable of OVRAL is stressed, however, we are led to believe that the “A” of OVRAL is intended to be pronounced to rhyme with “all” and “ball.” The pronunciation of AMR Pharm’s product, B-OVal, exists alongside the common English pronunciation of the word, “oval,” in its “a” sound. It is likely that many consumers would see the familiar word, “oval,” here and pronounce BOVal in the same way that they pronounce this common word meaning egg-shaped. The fact that the “a” in B-OVal is in lower case might also lead readers to pronounce it as they would other unstressed vowels, with the “uh” sound. In any case, the company itself suggested that there were two possible pronunciations of this vowel, either “OVuhl,” or “OVahl.”


3. Beginning consonant and following vowel The first two sounds of B-OVal are a voiced bilabial stop consonant, /b/, followed by a high front vowel, /iy/, the conventional way we pronounce the letter “b” (the same as the words, “be,” and “bee”). Neither OVRAL nor LO/OVRAL has a “b” sound at the beginning or anywhere else. LO/OVRAL obviously begins with a voiced lateral continuant “l”, quite different from the bilabial voiced stop “b” sound that begins B-OVal. 4. The “r” sound B-OVal does not contain an “r” sound. Both OVRAL and LO/OVRAL do. 5. Word stress The stress or accent is on the last syllable in OVRAL, indicated as [öh’vral] in PDR, while, as the contrastive use of capitalization suggests, stress is on the “OV” syllable of B-OVal. 6. Final syllable Since the final syllables of OVRAL and LO/OVRAL are stressed, they are pronounced differently from the final syllable of the English word, “oval,” /’ow vuhl/. They parallel the unstressed vowel pronunciations of other English words ending with the morpheme “-al,” such as “capital,” “topical,” “legal,” “mural,” “acquittal,” and “signal,” and even with words ending the morpheme, “-il,” such as “civil,” “fossil,” and “utensil” and words ending with the morpheme “-le,” as in “gentle” and “cuddle.” The sounds of OVRAL and LO/OVRAL vs. B-OVal, therefore, contain important pronunciation differences: /owv ‘rahl/ and /lohowv’ rahl/ vs. /biy’owvuhl/ or, possibly, /biy’owvahl/. Dissimilarity of looks The most obvious print feature differences are found in the capitalization and punctuation practices used in the two marks. OVRAL and LO/OVRAL are in all caps, whereas B-OVal ends with two lower case letters (-al). Between the modifier and the proper noun, LO/OVRAL contains a slash, whereas B-OVal makes use of a dash. Perhaps most different about the looks of the two marks is that BOVal begins with an entirely different letter, “B.” Readers attend to the beginnings of words more than anywhere else, as specialists in child and adult literacy regularly agree.


Dissimilarity of meaning Both the OVRAL and B-OVal marks make use of the association with the common root or base form of “ovum,” the Latin word for egg. In itself, this association cannot be considered distinctive. Many English words, including “ovulate,” derive from the singular form, “ovum” or its plural form, “ova,” as in “ovary.” By adding an “l” to “ova,” however, AMR Pharm also suggests the oval shape, not unlike that of an egg, differentiating it from the OVRAL mark, which, at best, uses only the first two letters of the Latin derivation, “ovum” in its name. Perhaps the most significant meaning feature in the trade mark, BOVal, however, derives from the contextual association of its initial “B” with breast or breast cancer. This meaning is necessarily an acquired meaning, but because of the public concern about breast cancer, especially as it had become associated with contraceptive pills, it could be readily foreseeable that the initial “B” in B-OVal could easily acquire such meaning. Conclusion of the case As noted earlier, I will probably never know for sure what happened in this case. I understand that two of the lawyers I was working with left the firm handling the case sometime during my involvement. My calls were not returned after a call on my telephone answering machine briefly explained that my services were no longer required. Subsequently I have searched for any reports of this case in legal references and have found none, suggesting that it may have settled and never gone to trial. Like the ConAgra v. Hormell case (chapter 6), where my goal was to show that the two brand names had different meanings, this case followed the same general principles of analysis but showed that the brand names were similar – same tools, different conclusions. My point is that linguists select from among the same tools in their toolbag of linguistic analytical routines no matter what the data, and let the conclusions be what they are. This is not advocacy. It is linguistic analysis. I would have made the same analysis for either side, whether or not they liked it. If they didn’t like it, they could thank me and say good-bye, which might have been the case with AMR Pharm for all I know.

10 Differences in the Ingredients, Qualities, and Characteristics of the Products

At the Toronto Toy Fair of 1990 a new children’s play product was introduced by its creator, Toni McKee. She got the idea from her university class in which she studied different types of chemical suspensions. One such suspension had molecules that would bind to a certain extent yet come apart easily, melting into a liquid when it was handled. “I got the idea of making it into a children’s toy,” McKee stated (BC Woman, October, 1990, p. 13). McKee soon found business partners, consulted with chemists, and came up with a toy which she named GUK. Later McKee assigned her rights for GUK to Pyewacket Enterprises, Inc. of Canada, which registered it with the US Patent and Trademark Office on February 5, 1991 as a powdered mixture, which, when mixed with water, becomes a tactile substance for entertaining and educating children.

Pyewacket Enterprises, Inc. v. Mattel, Inc. On March 24, 1994, Pyewacket brought a complaint against Mattel, Inc., a large US toy maker, alleging trademark infringement, unfair competition, and trademark dilution arising from Mattel’s advertising and sale of a novelty item that it called GAK, claiming that Mattel’s “gooey, tactile substance” was similar to that which Pywacket had sold and marketed as GUK and would cause market confusion to consumers, who might believe that GUK and GAK emanated from the same source, or that Pyewacket and Mattel were associated with each other. Mattel, in its Answer of May 24, 1990, denied these allegations, among other things noting that its own use of GAK preceded any use of the mark GUK by plaintiff and that there was no possibility of confusion by consumers since the two names are not similar in pronunciation, 116

Differences in the Ingredients, Qualities, and Characteristics of the Products 117

meaning, or appearance. Mattel further noted that Pyewacket’s name, GUK, along with its phonetic equivalent, “guck,” is a common descriptive term that describes the nature and quality of Pyewacket’s product. As such, Mattel claimed, GUK is merely descriptive and had acquired no secondary meaning. Therefore, it could not function as a trademark that signifies or identifies to the public that the goods came from only one source of origin. Therefore, Mattel claimed, Pyewacket did not own any valid rights to the exclusive use of GUK as a trademark, nor could Pyewacket assert any rights against Mattel’s use of GAK. In addition, Mattel stated that there have been third parties using the GAK mark, noting Nickelodeon Gak, owned by MTV Networks, as an example of an indispensable party that failed to join Pyewacket in this suit. The Nickelodeon channel had broadcast a program called “Double Dare” as a nationally shown quiz program in which two teams compete against each other by answering questions or by taking physical challenges. Since 1986, that show had featured GAK in stunts and physical challenge events. In those programs, the name GAK had appeared on stage plainly printed on large vats and was mentioned regularly on the show. In 1988, Nickelodeon also aired a program called “Family Double Dare” and in 1989 it aired “Super Sloppy Double Dare.” Both of these programs also featured GAK. By 1990, Nickelodeon had opened up the “GAK kitchen” attraction at its studios at Universal Studios in Florida, showing customers how GAK was made and inviting visitors to interact with the product. Mattel’s Answer to Pyewacket’s Complaint pointed out that its own GAK mark had been used extensively at least two years before Pyewacket’s registration or use and that the GAK mark had become strongly associated with Nickelodeon. In 1991, MTVN began negotiations with Mattel concerning the licensing of various Nickelodeon marks for use on activity toys derived from Nickelodeon’s programming. In February 1992, Mattel offered for sale to the trade the product Nickelodeon GAK, a pre-made, synthetic substance that could be played with, used to make noises, stretched, thrown, or inflated. Packaging had made it clear that Mattel’s GAK was a spin-off from the Nickelodeon television programs and was a copy of the GAK product used on those programs. When it went into production, Mattel’s GAK product ensured this connection by displaying the Nickelodeon name prominently and by using the Nickelodeon color scheme and burst design. Mattel also claimed that Pyewacket’s GUK differed from GAK in that GUK was a powdered substance that had to be mixed with water to create a usable substance that liquefies and drips out of the user’s hand. GAK,


in contrast, was a solid substance that holds its shape when in the user’s hand, was capable of making a noise when used, and could be inflated. Linguistic analysis In the context of these legal arguments, the attorneys for Mattel called me to analyze the names, GUK and GAK. Were these names similar in how they sounded, how they looked, and what the names meant? Is there anything linguistics can say that would contribute to an understanding about whether or not consumers would be confused that the products might be one and the same, or that the two companies were one and the same? Perhaps more importantly, does the name of Pyewacket’s product, GUK, fall into the category of being merely descriptive? To frame a basis for my analysis, Mattel’s attorneys sent me Pyewacket’s Complaint, Mattel’s Answer, photocopies of Mattel’s packaging for GAK and Pyewacket’s (along with its distributors’) packaging for GUK, various newspaper articles about Pyewacket that discussed GUK, documents that showed how GUK was promoted by Pyewacket and its distributors, Pyewacket’s marketing plan, and a videotape advertisement for GUK entitled, “GUK, What Is It?” The path from the linguistic analysis to the expert report is often winding. What I report next is the process of my thinking about the issues, even though much of this thinking did not reach the ultimate expert report. I began with the sounds of the two product names. Both GUK and GAK contain three phonetic sounds, of which 33 per cent are clearly different and distinct. If a difference in only one-third of the sounds did not seem enough, I noted that it is commonplace for English words to have what linguists call minimal pairs, words that differ by only one sound but denote very different meanings. Words pairs that differ minimally, such as “big” and “pig,” or “top” and “tap,” are not commonly confused. Even one phonetic difference in a word can create an entirely different meaning. The vowel of GUK is produced in the mid-central portion of the mouth (rhymes with “yuck”) while the vowel of GAK is produced at the low-front portion of the mouth (rhymes with “tack”). The GUK spelling could also yield an alternative pronunciation using the high-back vowel (rhymes with “book”). The mid-central vowel of GUK occurs very commonly in English. For most speakers, in fact, it appears in virtually all unstressed syllables. For many English speakers, especially those from the Midlands dialect areas


of the US, the first and last vowels in the word, “tomato,” are unstressed, and therefore are produced with the “uh” vowel even though their spelling indicates an “o.” In contrast, words spelled with the letter, “a,” such as “dad,” “glad,” and “static,” usually preserve the same sound that is found in GAK, whether they are in the stressed syllable (as in the word “tactic”) or in the unstressed syllable ( as in the noun, “contact”). My preliminary examination of the sounds in GUK and GAK led me to the conclusion that they differ in much the same as do other distinctive minimal word pairs found in the English language. In my initial thinking about the issue, I also tried to determine the meanings of the words GUK and GAK. Recent American dictionaries contain neither GUK nor GAK, but they do contain the reasonably spelled variation “guck,” Webster’s Collegiate Dictionary showing it to be in use since 1949. This dictionary defines “guck” as “oozy, sloppy, dirt or debris” and gives as synonyms the words, “goo” and “gunk.” “Goo” is cited since 1911 and is defined as “a viscid or sticky substance” and “guck,” in use since 1943, as “filthy, sticky or greasy matter.” The American Heritage Dictionary defines “guck” as “a messy substance such as sludge,” and suggests that the origin of “guck” is possibly the combination of “goo” and “muck.” I was able to find separate entries for both “guck” and “guk” in the Macquarie Dictionary, both of which cite “gunk” as a synonym. In contrast, I was not able to find any mention of GAK, or its possible and reasonable spelling, “gack,” in any dictionary. It was clear that “guck” has an already existing descriptive meaning associated with a mucky, messy, oozy, gooey, icky, mud-like substance. All dictionaries were silent, however, about GAK or “gack.” I also discovered that GUK has a robust ability to extend beyond its noun form. In the media and advertising materials sent to me I found instances of the adjectival form in a compound, “Guk bucket,” of the adjectival suffix, “-y,” “gucky,” and of the verb form, “Guking.” One of the characteristics of a word’s descriptiveness is its ability to extend itself to other grammatical forms. There was no evidence, however, that GAK had in any way begun to spread itself across grammatical categories (from noun to adjective or verb). It was my opinion, in summary, that GUK and GAK do not share the same grammatical status and do not sound alike. At this point there was not enough semantic evidence to determine whether or not they mean the same things. Even from this, however, it is possible to conclude that the two marks are not likely to be susceptible to confusion by consumers. Although I had drafted the above opinions for use by Mattel’s attor-


neys, the procedure Mattel chose to present them differed in this case from most of my past experiences in civil cases. Instead of presenting an opinion written by me, the attorneys produced a document responding to Local Rule 26(a)(2)(B), called Statement Re Opinions and Testimony of Roger Shuy, presented to the United States District Court for the Western District of Washington at Seattle. This document organized and presented my proposed trial testimony in the form of two questions, to which my proposed testimony answers were then summarized. These two issues/questions, along with the gist of my proposed responses, follows here: ISSUE A: Does the mark “GUK” describe the ingredients, qualities, or characteristics of the product provided under that mark? 1. Dictionary definitions The attorneys stressed my finding that “GUK” and “guck” are equivalent, as evidenced by the Macquarie Dictionary entry that defined them the same way. It was also noted that the most important source for assessing how the consuming public perceives or understands a term is its dictionary definitions, since dictionary definitions are based on how the public uses the term and that it is clear that “guk/guck” has an existing meaning in the English language and was not created or coined by plaintiff. 2. Synonyms Once it becomes clear that a term has a commonly understood meaning, one must determine whether or not this meaning describes the ingredients, qualities, or characteristics of the object to which it is attached in order to discover whether the term is descriptive of that object. If a word is descriptive of an object, the synonyms attached should also be descriptive. It is clear that “guck’s” dictionary definitions and purported synonyms (cited above) describe the ingredients, qualities, or characteristics of the product advertised and promoted under the “GUK” trademark. Virtually every single one of the definitions and synonyms of “guck” found in various dictionaries is used in Pyewacket’s own descriptions of the attributes or qualities of the “GUK” product. Based on this, it is clear that “guk” describes the product in connection with which the term is used. 3. Advertising, packaging, and promotion One of the packagers of GUK called the product “Mutant Muck” and told purchasers to “get ready for hours and hours of mad, mushy, mutant, muckity-muck fun with GUK.” This product, the packaging


went on to explain, is “an oozy, squishy, icky, gooey, slimy lump of laughs.” The 15-second television commercial for GUK ended with the tag line, “these four boxes made all this gucky stuff.” Mattel’s document concerning my proposed testimony included many similar citations from the media as well as from the plaintiff’s own business plans. Comments found in the plaintiff’s own documents and words clearly identified GUK’s characteristics, uses, and qualities in a way that was consistent with the way that dictionaries used descriptors and synonyms to describe the word, “guck.” 4. The media It is clear from the media articles about GUK that it is possible for the public to understand it to mean a mucky, messy, oozy, gooey, icky, mud-like substance. For example, one newspaper article, entitled “GUK ‘somewhat messy’ but lots of fun,” began with the sentence: “It squishes, drips, oozes, splatters and solidifies.” According to that article, GUK has been called “the designer mudpie.” Other articles agreed that GUK could best be described as “an upscale mudpie.” Another article described GUK as “strangely fascinating pink mud” and commented that “playing with GUK is like playing with pink mud . . . you just mess around mindlessly.” Still another explained, “it’s more than GUK’s gooey nature that makes it fun for kids. . . . GUK is a synonym for messy.” Such references are direct assertions of the meaning of the GUK product’s name. Like the plaintiff’s own descriptions of its GUK product, these media descriptions of the qualities and attributes are consistent with the dictionary definitions and synonyms of the word, “guck.” The media references support and validate the dictionary descriptions of the characteristics and qualities of “guck” and indicate that there is no variation between dictionary definitions and the use of “guck” by the media. Both describe the same characteristics, qualities and attributes of the product. ISSUE B: Is the mark GUK similar in sound and meaning to the mark GAK? The question of whether or not GUK and GAK sound alike is addressed by phonetic comparison. While there is not necessarily only one way to pronounce a word, here, no matter what English dialect one uses to pronounce “guk” and “gak,” there is no way that the two words will sound alike. Vowels typically are the nucleus of syllables. Since English stress (loudness and pitch) is carried by vowels, vowels are central to intona-


tion. Although bordering consonants are also important, the stressed vowel is at the heart of words. Thus, it is quite common for English words to have similar consonants but, because of a single important vowel difference, not be confusing to listeners. Differences in the stressed vowels of otherwise similar words will result in entirely different pronunciations that convey different meanings. Phonetically, GUK is likely to be pronounced in a way that rhymes either with “yuck” or with “book.” Phonetically, GAK is likely to be pronounced in a way that rhymes with “pack.” Both of these words contain three phonetic sounds. The vowels, making up a third of the sounds in each, are simply not the same. The vowel of GUK is a mid-central vowel while the vowel of GAK is a low front vowel. This difference between GUK and GAK makes them different in a linguistically significant way. For example, the words, “buck” and “back,” contain the same vowel differences as GUK and GAK, yet “buck” and “back” are significantly different phonetically and denote very different meanings. Nor are GUK and GAK similar in meaning and, because of this, they will be perceived and understood very differently by the consuming public. GUK has an existing dictionary meaning and the plaintiff and the media are consistent with that meaning when they refer to it. Based on what is known about its sound and meaning, the public will recognize the word GUK, and immediately think of something gooey, mucky, oozy, or sticky. By contrast, GAK has no dictionary meaning in the English language. It has no definitions, associations or synonyms, apart from its association with Mattel and Nickelodeon, and, therefore, will not call to mind exactly the same image as does the GUK mark. More should be said here about the fact that Mattel’s attorneys produced the document that told the Court what I was “expected to say” at trial. Much of Mattel’s summary was accurate, but at least some parts of it went beyond such an expectation. In the preceding paragraph, for example, Mattel indicates that since GUK and GAK are not similar in meaning, what they mean “will be perceived and understood very differently by the consuming public.” I would have said something more like, “are likely to be perceived and understood differently by the consuming public.” Nor could I have been reasonably expected to say that GAK “will not call to mind the same image as does the GUK mark.” I would have said “is not likely” again. I was not prepared to go as far as the attorney would have me go, since linguistics cannot say exactly what was actually intended or was actually understood. My findings were that GUK had a clearly defined meaning, but that GAK had not yet achieved that status, although it is possible that it might in the future.


There is a lurking ethical issue involved here. In most law cases, the attorney (and the Court as well) search for quantitatively indicated degrees of confidence. In the past, I’ve been asked, for example, whether the certainty of my conclusions is 100 per cent, 90 per cent, 75 per cent, etc. This type of analysis simply doesn’t lead to the kind of answer that may be desired. The range of certainty in matters of how any person may speak or understand something is just not that tight. Based on what linguists know about the alternative possibilities of language, the best one can come up with is a range that includes markers such as “likely” or “unlikely.” When pressed by an attorney to broaden this range more to include “very likely,” “extremely likely,” “very unlikely,” or “extremely unlikely,” the linguist begins to fidget. In this lawsuit, the attorneys for Mattel pushed too far. They upped my conclusion far beyond what I had presented them to the absolute category represented by “will” rather than “likely,” an ethical matter of their own making. Perhaps more effective communication between the linguist and attorney would have clarified these points. I can’t help thinking, however, that it would have been better if the draft expert report that I had prepared had been used instead of the attorney’s attempt to represent what I was expected to say at trial. My draft said clearly that “guck” has an existing meaning but that “gack” does not, as follows: Since GUK derives from the word, “guck,” which has anestablished meaning (as revealed by dictionary definitions, media articles, its own advertising plan, its own business plan, its own advertising, and from the testimony of plaintiff’s president), this meaning is well established and clear. GAK (or its graphemic equivalent, “gack”) on the other hand, has no dictionary definitions and no discernible descriptive meanings, associations, or synonyms. I don’t know what would have happened if I had been deposed, but I suspect that the attorneys I was working with would not have been pleased. This instance appears to be another case of the authority of attorneys taking precedence over the authority of others. They did not consult with me about how they would word my “expected testimony at trial.” I was shown it only after it was submitted, too late for any input or corrections from me. Conclusion of the case Just as in the case of AMR Pharm Holland v American Home Products (chapter 8), the conclusion of the Pyewacket v Mattel case remains some-


what unclear to me. Once again, there were changes in the attorneys in the law firm representing Mattel. The attorney I began with was no longer there when I called to find out when my deposition might be held. I was unable to get anyone else at that firm to return my calls, leaving me in limbo once again. Perhaps Mattel’s attorneys realized the difference between my draft report and their own summary of what I would say about the issue of differences or similarities of meaning and decided not to take a chance on this coming back to haunt them. I was simply unable to find out. Nor was I able to find any legal references that would enlighten me about the conclusion of this case. In desperation, I finally called one of the attorneys on Pyewacket’s side. He informed me that the case was settled in February 1996 and that “there was a payment to Pyewacket” but the specific terms were confidential. There was no way to verify the accuracy of what he told me. In any case, this type of “settlement” does not always indicate that the paying party conceded defeat. Assuming that what the Pywacket attorney told me was accurate, it may have been the case that Mattel simply wanted to avoid the expense of a trial and felt that it would be easier to pay Pyewacket something and make it all go away. Or it may be that Mattel didn’t think it had a good case. I’ll never know for sure. In any case, this points out the futility of trying to keep score of “victories or losses” in civil cases. Without a trial, it’s difficult to know who actually won or lost.

11 Going beyond Competing Company and Product Names

Most trademark cases involve claims of infringement of product names, such as the ones discussed in the previous chapters of this book. But litigation also can be brought for the alleged infringement of expressions, including slogans, that are closely associated with product names, such as the General Mills expression “Breakfast of Champions,” which accompanies the name of its breakfast cereal, Wheaties. It is also the case that the names of procedures and devices used by different companies as part of their products or services can also achieve trademark protection, as long as such a designation is promoted, used, and advertised in such a way that buyers link it only with only one seller.

AutoNation, Inc. v. Acme Commercial Corporation et al. d/b/a CarMax the Auto Superstore In a 1996 trademark dispute between AutoNation, Inc. and Acme Commercial Corporation, d/b/a CarMax the Auto Superstore and Circuit City Stores West Coast, Inc., the major issues were not over the names of the two companies, but over 1) the association of AutoNation’s name with CarMax’s computerized inventory system, called AutoMation, and 2) the two companies respective slogans. CarMax’s slogan was “The New Way to Buy Used Cars,” while AutoNation’s was “The Better Way to Buy a Car.” Slogans are protectable in the US if they are distinctive or capable of distinguishing. France has used a somewhat different standard, protecting a slogan that is technically an explanatory phrase containing euphony, brevity, and impact. A bit of background should be given first. It is no secret that one of the most troublesome events in American culture is thought to be a customer’s effort to purchase a used car. Linguists view this as one of many 125


speech events, occasions in which the topic and context dictate how language should be used. In a negotiation, for example, certain levels of formality, seqencing rules, specialized vocabulary, and politeness routines are expected. These and other language characteristics govern discrete speech events, such as business meetings, classroom talk, and counseling. In some speech events, the exchange takes place symmetrically in terms of social status, knowledge of the participants, and other indicators. The used car sales event, however, is an asymmetrical speech event, that is, an event in which one party has language skills and technical knowledge superior to, or unknown by, that of the other party, making the exchange seem (or be) unfair to the customer. There are other such asymmetrical speech events in life, of course, such as the appearance of a layperson on the witness stand at a trial. Laypersons do not have knowledge of the specialized language and culture needed at a trial and they often founder or become sitting-duck victims of the lawyers who question them. Another example of an asymmetrical speech event is commonly found in the communication between physicians and their patients. Physicians have specialized knowledge and language that, when used with a person lacking that specialized knowledge and language, sometimes has been found to lead to severe miscommunications. The high status attributed by laypersons to doctors and the courts often silences any possible complaints by laypersons. But although used car salespersons do not share the social status of physicians or lawyers, they still have technical information and language that, if misused, can lead to customer discomfort. Thus, one of the most commonly disliked speech events grows out of the asymmetry that develops when a consumer attempts to buy a used car. One effort to make this generally disliked asymmetrical speech event more palatable to American customers began in 1993. CarMax, whose parent company was Acme (whose parent company, in turn, was Circuit City), put forth a new, consumer-friendly approach to used car buying with its car superstore concept, using the trademarked expression “The New Way to Buy Used Cars.” People interested in buying a used car could come to the CarMax location and be greeted by a host instead of a salesperson. The host would then usher the customer to a bank of easy-to-operate computers containing CarMax’s inventory browsing system, which was called by its trademarked name, “AutoMation.” On this computer customers could then punch in the brand, year, and model that they were looking for and the computer would provide a color photo, price, and other details of the vehicle selected, also giving

Going beyond Competing Company and Product Names 127

them a map of where to find the vehicle on the car lot. Above each computer stand was a large sign saying “AutoMation.” The price listed was the final price, no haggling, no negotiating, no slick sales talk. No CarMax employee was paid a commission. Everyone, including the host, was either on a set salary or paid by the hour. Among other things, CarMax believed that this simple, friendly system would remove the common stigma of used car sales experience. By 1996 this used car sales approach had captured the attention of the national press, with articles appearing in Forbes, Fortune Magazine, Business Week, Economist, Automotive News, Wall Street Journal, Los Angeles Times, and Washington Post. Meanwhile, or shortly after, AutoNation was developing similar plans. Wayne Huizenga, chairman of Republic Industries, had been trying to sell his most recent acquisition, Blockbuster Entertainment, and was approached about heading a new operation to buy new car dealerships. In the process, his business associations led him toward the used car business and eventually he teamed up with Jim Moran, a Toyota distributor in the Southeast, to form the AutoNation USA enterprise. Before it opened its first store, Huizenga made an offer to buy the older, more established CarMax chain. His offer was rejected. AutoNation was subsequently sold to Huizenga’s Republic Industries in January 1997. As is common in trademark cases, this lawsuit was initiated by the party that felt that its trademark had been infringed. On January 11, 1996, counsel for CarMax sent a letter to AutoNation accusing it of violating federal and state law and insisting that AutoNation cease using the marks “AutoNation” and “AutoNation USA” as well as its accompanying expression, “The Better Way to Buy a Car.” The salient parts of this letter are reproduced here, as an example of a how a trademark dispute begins: Your proposed name, trademark applications, and slogan are likely to be confused with our clients’ marks. We can only assume that you are attempting to trade on the valuable goodwill developed, at great expense, under our clients’ marks. Such actions violate state and federal law and infringe upon our clients’ intellectual and property rights. . . . It also appears from various articles . . . that AutoNation intends to follow in CarMax’s footsteps, using the concepts pioneered by CarMax. . . . We hereby demand that you withdraw your trademark applications, and abandon your plans to use your proposed name and slogan, any confusingly similar words or phrases, and any other features or designations confusingly similar to the


marks, signage, and trade dress of CarMax and Circuit City. . . . If I do not receive your written commitment to abandon all such use within ten days of your receipt of this letter, we will take appropriate action to protect our clients’ rights. The following month AutoNation filed a complaint before the US District Court, Southern District of Florida, Fort Lauderdale Division, requesting a declaratory judgment concerning claims arising under the Federal Trademark Act. This complaint pointed out that plaintiff had operated under the names “AutoNation” and “AutoNation USA” since late October 1995 and that these names were chosen without knowledge of CarMax’s use or registration of the mark, “AutoMation.” It pointed out that plaintiff had filed trademark and service mark applications for these names and services as well as for its slogan, “The Better Way to Buy a Car.” AutoNation’s complaint indicated that CarMax’s mark was merely the name of a computer browsing service provided in its showrooms and was not the name of any of its dealerships. CarMax, therefore, was considered to be the primary mark used by CarMax while the mark AutoMation was called “an ancillary mark,” used only to provide “a small portion of the services” offered at the CarMax dealerships. In short, AutoNation denied any infringement of intellectual property rights, denied that any federal or state laws had been violated, and stated that the marks in question were not confusingly similar to or in conflict with CarMax’s mark and slogan. Of course, CarMax responded to this complaint, and on October 11, 1996, filed a counterclaim and charging AutoNation with the same things introduced in the original letter of January 11 of that year, and demanded a trial by jury. This counterclaim included evidence of CarMax’s preceding federal and state registrations for the mark, “AutoMation,” as well as the expressions, “The New Way to Buy Used Cars” and “The New Way to Buy Cars.” Involving linguists In January 1997, I received a call from Louis Colombo, the attorney representing AutoNation. I had worked on several other cases with Colombo in the past, but this time he asked me for the name of a linguist who might be a bit closer to the site of the trial, which was to be held in West Palm Beach, Florida. He said that it was a trademark case and would require travel and expense that would probably preclude my own possible involvement, since I had recently retired from teaching at Georgetown University and had moved from Washington DC to


Missoula, Montana. I quickly suggested that he call my colleague and friend Ronald Butters, at Duke University, and I thought the matter was settled. Within a few weeks, however, I received a call from Gilbert Schill, the attorney for CarMax, about this same case. It appears that Schill had used Ron Butters in the past and expected to call him in this case but discovered that Butters had already been selected by AutoNation as its linguistics expert. Schill asked me to help him with his CarMax side of the case. Thus the two opposing linguists in this case were in the strange position of working on the opposite sides of the very attorneys with whom they had worked on other cases in the past. Was “AutoNation” similar in sound to “AutoMation”? Were the taglines “The New Way to Buy Used Cars” and “The Better Way to Buy a Car” similar? Or were the two tag lines so generic that no controversy was even possible? What was to be done with the fact that AutoNation insisted that any comparison should be made with the name they preferred and sometimes used, “AutoNation USA.” Perhaps most interesting of all was the fact that here we had a case in which the mark of the computerized inventory browsing system of one company was said to be infringed by the brand name of another company. This was a serious concern for CarMax. The following reports the sequence of developments by both linguists in this case, beginning with my preliminary thinking before Butters submitted his expert report. After summarizing his report, my own is summarized. Next follows a summary of Butters’ depositon testimony and a summary of my deposition testimony. This is the order in which these events happened and it does not imply strengths or weaknesses of the arguments of either side.

My preliminary thoughts on the case To address the issue of how to compare CarMax’s AutoMation with AutoNation USA, I first needed to determine whether or not the “USA” part of the name was salient for such a comparison. For this I needed data. I requested a Nexis search (beginning in May, 1996) and a Westlaw search (beginning in April, 1996) of media mentions of AutoNation and AutoNation USA, in order to find out how the public referred to the company. I was also given print, radio, and television data mentioning AutoNation and AutoNation USA in advertisements and news programs, along with the deposition testimony of AutoNation officials. From this data base I compiled the following preliminary information from the print media for the CarMax attorney.

130 Linguistic Battles in Trademark Disputes AutoNation and AutoNation USA in the Print Media: 84 articles In the headline of the article: AutoNation AutoNation USA All references within the article: AutoNation AutoNation USA

69 13

84% 16%

265 101

72% 28%

When I broke out the usage by individual newspapers and magazines, I found that only one of the magazines, Automotive News, had favored the use of AutoNation USA with any consistency. Other print media, including the Miami Daily Business News, Wall Street Journal, USA Today, Newsweek, and Fort Lauderdale Sun-Sentinel lopsidedly referred to the company as AutoNation, at least in most references to the company after it was initially cited as AutoNation USA. From this compilation, it seemed clear that the quantitatively preferred reference to the company was AutoNation. But further supporting this position seemed to be that one of AutoNation’s own executive officers, Thomas Gruber, was quoted in an article in USA Today saying: The logo is designed to resemble a highway sign, so that as you’re going down the road all the signs will remind you of AutoNation. In that same article another AutoNation executive was also quoted, saying: We are only buying new-car dealerships where we are running AutoNation stores. If AutoNation’s own executives called it AutoNation, it could seem fair to compare the uses of AutoNation with CarMax’s AutoMation. On the other hand, AutoNation’s own radio and television advertisements clearly favored the use of “AutoNation USA” while television news programs, not controlled by the company, favored AutoNation, as follows: Spoken Mentions of AutoNation USA and AutoNation in Electronic Advertisements and Newscasts

AutoNation AutoNation USA

TV ads

radio ads

TV news

3 5

0 5

10 5

46% 54%


The concept of this type of store was still very new, accounting for the lack of data noted above. What this comparison told me, of course, was that AutoNation was careful to include “USA” in any representation of its name when it had control over the actual text that would be used or when it actually wrote the script. But when it did not have such control, as in a local news programs, use of “USA” was not nearly as frequent. This fact seemed to support the frequency of the non-USA references found in the print media articles. From this I concluded that it seemed fair to attempt a comparison of the sounds and meaning of CarMax’s AutoMation with AutoNation rather than with AutoNation USA. My initial comparison of the sound similarities and differences between AutoMation and AutoNation follows. The two words contain eight speech sounds, seven of which are identical. The only difference is in the fourth speech sound in each word, the sound represented by the letter “m” in AutoMation and by the letter “n” in AutoNation. The sounds represented by “m” and “n” share many phonetic qualities and are as similar as any two separate speech sounds can be. Both are nasal sounds (produced by sending air through the nasal passage) and both are classified as continuants (sounds which can be continued as long as one has breath to do so), as opposed to “stops,” which are made by a stoppage of breath (as in the sounds represented by the letters “p,” “t,” “k,” “b,” “d,” and “g”) or “fricatives,” sounds made by producing friction within the mouth (as in the sounds represented by the letters “s,” “sh,” “z,” “zh,” and “j”). Both are voiced sounds (made by friction in the larynx) as opposed to voiceless sounds, which have greatly reduced friction in the larynx, as in the sounds represented by the letters “p,” “t,” or “k.” Voiced nasal continuants, being so similar to each other, are easily misperceived or confused by listeners. But even if it were decided that AutoMation should be compared with AutoNation USA, the similarities between the two names appeared to me to outweigh the differences. Although “USA” contains five additional phonetic sounds, these are at the end of the name, located in a position that readers are less likely to notice. Good readers process words primarily at their beginnings, not at their endings, as literacy materials abundantly illustrate. My initial examination also led me to conclude that AutoMation and AutoNation are graphemically similar in looks, with only the capitalized “M” and “N” letters being slightly different. These two letters differ by only one stroke and a small variation in angle, both of which are perceptually minor. The two letters are readily capable of visual


confusion. From my past experience in developing beginning reading materials, I was well aware that literacy specialists warn about the possibility of confusion between “m” and “n.”. As for the meaning of AutoMation and AutoNation, it was very clear that they refer to different things: a computerized inventory browsing system vs. the name of a company. Yet both names are not only similar in sounds and appearance, but also are both used in the same commercial context, the sale of used cars. As such, they seemed to leave open the potential that they could be confused with each other. Grammatically, AutoMation and AutoNation both contain two morphemes, one free and one bound. The fact that AutoMation has a capital letter “M” after “Auto” isolates the first morpheme and suggests that it associates with “auto,” or “car.” I also noted that the etymological source of “auto,” as we use it today, is from the Greek, “automata” (selfdirected or propelled). In the context of an auto store, the word, “AutoMation,” seemed to suggest a reference to cars. The fact that the bound morpheme, “-Mation,” follows “Auto-” in the context of a computer screen seemed to suggest creative wordplay about the connection between cars and the computer process that was the first integral step CarMax’s used car sales operation. I also found a similarity in the accompanying expressions used by both companies in the representation of their products. CarMax used the expression “The New Way To Buy Used Cars,” while AutoNation (with or without “USA”) used the expression “The Better Way To Buy A Car.” These two expressions share the words, “The,” “Way,” “To,” “Buy,” and “Car.” The similarity of the two expressions is noted in the words of the AutoNation executive, who stated in one news program, “This is the way to buy a used car,” uttering, perhaps unintentionally, words closer to those used in the CarMax expression than the slogan used by his own AutoNation. In addition, an advertisement for CarStop, an affiliate of AutoNation, read: “A Whole New Way To Buy a Car!”, echoing the use of “new” in CarMax’s advertising expression. Even at this preliminary stage, however, I was concerned that both expressions contained words that were very common, if not generic. I presented these initial thoughts to the attorney, along with the problems that the case might present. Even at this stage, it seemed that the linguistic battlefield might involve the possible generic quality of the tag lines, the fact that AutoNation USA usually appeared as the first reference to the company in the media, and, most difficult of all, that there would be considerable discussion about the fact that AutoNation USA was the name of the company while AutoMation referred to the


computerized inventory browsing system used by CarMax. But by then the die had been cast. Without becoming an advocate, expert linguists must make the best case they can for their clients. AutoNation’s linguistic expert’s report Butters and I began our analyses at approximately the same time, but the plaintiff’s linguist always has to make the first formal move. Butters prepared his expert report and submitted it on February 4, 1997. It was nine pages long, single-spaced and contained 36 paragraphs. Butters focused his report on five areas in which AutoNation USA and AutoMation were different: in sounds, minimal pairs, morphemes, pragmatics, and semiotics. He then dealt with the expressions, CarMax’s “The New Way To Buy Used Cars,” and AutoNation’s “The Better Way to Buy a Car.” AutoNation USA vs. AutoMation 1. Phonetic differences. Butters agreed with my initial observation that AutoNation USA contains 13 sounds (but without the “USA,” only eight) and that AutoMation contained eight sounds. Butters further argued that the stress on the third syllable of each name is stronger in AutoNation than it is in AutoMation. He pointed out that CarMax had produced only one electronic commercial containing a pronunciation of AutoMation giving him, as well as the rest of the public, only one opportunity to hear how CarMax might prefer to have it pronounced. For Butters, this one pronunciation confirmed his analysis, based on the rules of English, that the “o” vowel in Automation was pronounced with a schwa (“uh”) sound. This report also observed that the two names contain a difference in their vowel sounds represented by the “o” in Auto-. Basing his argument on English stress rules and confirming this with the only existing CarMax television advertisement in which the name AutoMation was uttered, Butters concluded that the “o” vowel is characteristically pronounced with a schwa (“uh”) sound (rhymes with the last sound in “sofa”) whereas the sound represented by the letter “o” in AutoNation is pronounced with the vowel of “go” or “so.” He was quite right that 100 per cent of the sample of CarMax television advertisements used the schwa vowel here. There had been only one CarMax television advertisement produced and aired at that time. 2. Minimal pair lexical differences. Butters argued that an “m” versus “n” phonetic difference in the two names is sufficient to create a


minimal pair, citing the words “moose” and “noose” as examples of where this minimal contrast obtains. Minimal pairs form the basic distinctions between many otherwise similar English words. 3. Morpheme differences. Butters maintained that “USA” added another morpheme to the two morphemes of AutoNation, making it one morpheme longer than AutoMation. He noted further that CarMax’s “-Mation” is a bound morpheme (one that does not occur freely by itself) whereas AutoNation’s “-Nation” is a free morpheme. This difference in the quality of these morphemes was also said to distinguish the names in another important way. He pointed out further that the “Auto-” morphemes in these two names are not the same morpheme, despite their ultimately clipped derivation from “automobile.” This historical relationship, Butters claimed, is no longer relevant to their contemporary meanings, even if used in a playful sense. According to him, the “Auto-” of AutoMation means “self,” as in the words, “autograph,” “autohypnosis,” and “autobiography.” This contrasts with AutoNation USA’s meaning of “Auto-,” which means automobile. Another difference pointed out by Butters was that AutoNation is a noun compound while AutoMation is not. The “-Mation” of AutoMation cannot be found in a dictionary. The addition of “USA,” Butters continued, “further particularizes the meaning of ‘Nation’ to the American context, a geographical and patriotic association that is confirmed by the extensive use made of the nationalist theme in the television advertising of AutoNation USA.” 4. Pragmatic meaning differences. From my perspective, the strongest linguistic evidence that Butters had available to him was that AutoNation and AutoMation are used to refer to different things. Butters argued that their contexts were really quite different: a used car store vs. a computerized inventory system employed within a used car store. He downplayed the importance of the computerized inventory system, calling it “a relatively minor aspect of the CarMax operation,” as noted by the fact that “it plays only a tiny role in CarMax’s advertisements” and that it appears “only sporadically in its print advertisements.” 5. Semiotic differences. This section of Butters’ report began with a definition of semiotics: a branch of linguistic pragmatics which examines the ways in which meanings are conveyed through “signs” which are in themselves not


strictly linguistic but which often interact with strictly linguistic features. Butters then pointed out how the use of color, type case, and size indicate important semiotic differences. AutoNation USA and CarMax were different in these respects, with AutoNation USA making extensive use of US interstate highway geographical information signs while AutoMation appeared only as a small part of CarMax ads. “The better way to buy a car” vs. “The new way to buy used cars” Butters agreed that the two slogans “bear some obvious similarities” but went on to describe their differences as compelling, noting that there is little likelihood of consumers confusing the two. He pointed out three areas of linguistic difference: phonological/morphological, semantic, and pragmatic. 1. Phonological and morphological differences. Butters pointed out that there are 17 segmental phonemes and six internal word boundaries or junctures in AutoNation’s slogan, “The better way to buy a car.” CarMax’s slogan has 19 segmental phonemes (12 per cent more than AutoNation USA’s) and the same number of word boundaries or junctures. The two slogans share only 12 phonemes (37 per cent), the rest differing. He also opined that CarMax’s slogan “The new way to buy used cars” does not scan metrically whereas AutoNation’s slogan, “The better way to buy a car,” since it is in iambic pentameter, resists the need to pause between the noun phrase, “the new way” and the verb phrase, “to buy used cars.” Butters also noted the different morphemes used in the two slogans, calling the difference between the adjective “new” and the adjective “better” very significant. Also important was the meaning contrast between “a car” and “used cars.” 2. Semantic differences. Butters pointed out that the uses of “new” and “better,” which contrast in the two slogans, are common claims of superiority and uniqueness found in advertising. Consumers, moreover, are used to seeing and hearing them and can easily differentiate something new from something better. He noted further that “a car” denotes the entire class of the product while “used cars” denotes only a subclass thereof. Finally, Butters pointed out that CarMax’s use of the words, “new” and “used” in the same slogan produced a kind of antonymic uniqueness not found in the AutoNation slogan, further distinguishing the two.


3. Pragmatic meaning differences. Here Butters argued that the two slogans are used in “such different environments that it is not at all likely that native speakers of English would confuse the two marks.” He also stated that the overlapping words, such as “way,” “buy,” and “car,” are so fundamental to the enterprise at hand that “it is hard to imagine a coherent slogan that didn’t make use of them.” Perhaps illustrating his position that pragmatics and semiotics are intertwined, Butters concluded his report with the observation that AutoNation USA’s red, white, blue, and green logo always appears with its slogan attached while CarMax tends to accompany its name with the words, “The Auto Superstore,” all in blue, white, and yellow colors. This observation led Butters to the conclusion that the two slogans are not semiotic equivalents. CarMax’s linguistic expert’s report It was now CarMax’s turn to submit an expert’s report. My first task was to review Butters’ report and provide a rebuttal, also obviously giving any ideas I could about his analysis to CarMax’s attorney to use in his deposition of Butters. CarMax’s attorney, Gilbert Schill, explained to me that the best expert report was one that was as short as possible. At his request I prepared a very short one and submitted it on April 19, 1997. It turned out to be five double-spaced pages of 13 paragraphs. It was decided that, even though I had read Butters’ report, I would make no mention of it in my own. When it came time for depositions, there would be enough time for this type of interaction. As usual in such reports, my first four paragraphs were devoted to brief statements about who I am, what linguistics is, what documents I was given, and what my defined task was in this case. In the remaining nine paragraphs I addressed the similarities between the two names, followed by the similarities between the two expressions used in their connection. Similarities between AutoNation and AutoMation 1. Phonetic similarities. My paragraph 5 pointed out the fact that in the television news, AutoNation was referenced alone, without USA, in ten of the 15 mentions. In three of the eight AutoNation television commercials the company refers to itself as AutoNation, with no USA attached. In the print media, over 70 per cent of the mentions were to AutoNation, again with no accompanying USA. I next gave a very simplified explanation of the phonetic similarities of AutoNation and AutoMation, employing no phonetic symbols and few technical terms, noting that the only difference was in the nasal


consonants, “m” and “n,” which are easily misperceived or confused by listeners. I pointed out that even when the company referred to itself using the “USA,” these sounds were at the ends of the word, where most readers are less likely to closely attend. The similarity of looks between the names AutoNation and AutoMation is patently obvious, but I pointed out that the difference comes from only one stroke and a minor variation in angle of the nasal consonants, both of which are perceptually small and easily capable of visual confusion. 2. Grammatical similarity. I pointed out that both AutoNation and AutoMation contain two morphemes. The fact that AutoMation has a capital “M” after Auto- splits the word in two, strongly suggesting that “Auto-” associates with cars. The fact that -Mation follows Auto- in the context of a computer screen suggests creative wordplay about their obvious connection: cars and the computer process. 3. Semantic similarity. I pointed out that “auto-” is derived from the Greek “automata,” meaning self-directed or propelled, the etymological source of the first element of both marks. The bound morpheme, -Matic, like most bound morphemes, is not devoid of meaning. In this case the pragmatic meaning found in the context of its use makes clear that it relates to the computerized inventory browser. The fact that the two names are used in the same context of the commercial enterprise of selling used cars is another reason why they are capable of being mistaken for each other. Similarities between the tag expressions For reasons that were never clear to me (but may be perfectly clear to trademark attorneys), I was told to refer to the tag phrases at issue as “expressions,” and not as “slogans.” Butters had called them slogans but I followed instructions not to. I first pointed out the identical words used: “the,” “way,” “to,” “buy,” and “car(s),” noting that if readers had no more than these five words to go on, they could read them in this order and see no meaningful difference at all. The different words – CarMax’s “new” where AutoNation uses “better” – were both simple modifying adjectives. Likewise, CarMax’s “used cars” differed from AutoNation’s “a car,” but these were small differences: an adjective plus a pluralization of the noun. I also pointed out that AutoNation’s executive, Gerald Weber, stated on one news program: “This is the way to buy a used car,” echoing the “used”


in CarMax’s slogan and illustrating the unimportance of these minor adjectival and plural form differences. One of AutoNation’s affiliates, Car Stop, had advertised: “A Whole New Way to Buy a Car!”, also echoed the use of “new way” in CarMax’s advertisements. Depositions of the linguists Reports submitted, reviewed, and critiqued, Butters and I next faced the ordeal of being deposed. The depositions were within a week of each other, during the month of April, 1997. AutoNation’s expert linguist’s deposition Butters’ deposition was held in Fort Lauderdale, Florida on April 17, 1997. He reiterated the points that he had made in his expert report and, not surprisingly, attorney Schill challenged many of them. There is no good way to capture all of a long deposition with completeness and total contextual fairness. Nor do I attempt to be selective in a way that favors one side or the other here. The following points are noted only to show small bits of how the expert reports of two linguists were treated in their depositions. Both linguists held their ground rather well, I believe, and both remain good friends, despite some different points of view here. 1. The issue of AutoNation vs. AutoNation USA. Butters agreed that readers focus on the beginnings of words rather than on the ends of them, but added that readers skip only text that is redundant. “USA,” he said, is not redundant and would not be skipped since it is capitalized and separated from AutoNation with a word boundary space. This, he believed, would make the reader’s eye focus on “USA” since it began a new word. He made a good point. As I suspected, Butters argued that journalistic economy was the probable reason why these media articles contained so many mentions of AutoNation without the “USA.” He correctly pointed out that the first mention of the company in the articles was virtually always AutoNation USA, and, after that, the reference was to AutoNation. He noted that this practice was consistent with journalist’s use of pronoun references to the initial nouns to which those pronouns refer. He pointed out that it is equally true that journalists shorten company names in their subsequent references, citing the their practice of using Post Office for US Post Office, Volkswagen for Volkswagen of America, and Fed Ex for Federal Express.


2. The issue of my database of media articles. When asked about the media articles that I had described in my report, Butters objected that they did not constitute a random collection and were therefore not a representative sample. He noted that AutoNation’s own advertisements were more relevant data than secondary sources such as these media articles. He agreed, however, that media articles could be used to suggest how the public generally viewed the trademark issue if they were gathered and studied properly, since the writers would have no fear that they might be confusing or misleading their readers. 3. The issue of pronouncing the “o” of AutoMation. Butters continued his report’s assertion that the “o” vowel of AutoMation was pronounced with a schwa (“uh”) vowel (like the second vowel of “sofa”) because this word is not a compound noun. In contrast, he said, was the “o” vowel in AutoNation, which was pronounced to rhyme with “go” or “so,” because of the fact that AutoNation is a compound noun, made up of two free morphemes. Butters did not agree that it was easy to confuse the nasal coninuant sounds, “m,” and “n.” He stated that the sounds of “b” and “p,” among others, are even more easily confused than the sounds represented by the letters, “m” and “n.” Attorney Schill asked about Butters’ claim that the “o” of Auto- in AutoMation reduces to schwa when it occurs in a free morpheme. Schill asked if words like “Vegomatic” “Autoclave,” and “Automatic” contain reduced vowels as well. Butters agreed that AutoMation, if stress were placed on the third syllable, could preserve the “o” sound, but he added the caveat that if such pronunciation occurred, people would think that the “o” sound in AutoMation would sound stilted. 4. The issue of the relevance of context. Butters agreed that meaning is sometimes determined by context, but explained that context does not affect the underlying grammar of the structure. Context, he said, can help readers select which of the multivalent meanings of the lexical item is to be chosen and may also affect the connotation of the word, but it does not alter the grammar or the meaning. 5. The issue of the meaning of “Auto-” in AutoMation. Concerning CarMax’s use of “Auto-,” Butters described it as a pun. Within the context of an automobile story, the pun becomes germane and salient, creating the potential for connotation. But the speaker, Butters continued, never forgets the primary meaning. Schill then asked Butters the meaning of the prefix, “Mc-”, in the words, “McLawyer” and “McNewspaper.” Butters responded that it meant cheap, inexpensive,


and easily accessible. Schill then asked him what happened to “Mc-’s” original meaning as a patronym. Butters agreed that at least some people will see the “Auto-” of AutoMation as meaning cars. 6. The issue of my quoting an AutoNation executive. Butters also disagreed that the statement made by AutoNation’s executive, “This is the way to buy a used car,” was a concession to CarMax’s expression, claiming that this was purely anecdotal and not subject to scientific analysis. Schill then asked if slips of the tongue had any relevance to linguistic analysis. Butters replied that such slips were largely phonological, explainable by certain basic constraints that were often found in the same syntactic constituents or intonation units. But, he concluded, the executive’s addition of “used” here does not fit such constraints and was not a slip of the tongue. CarMax’s expert linguist’s deposition I was deposed on April 23, 1997. It was a difficult time for me, since my wife had just had hip replacement surgery and at that time I was her sole caregiver. To accommodate my situation, both attorneys graciously decided that my deposition would be held in my home city, Missoula, Montana. On re-reading my deposition, I was embarrassed to find that I was sometimes inarticulate and confused. Most embarrassing of all was that when asked about my previous depositions, I completely forgot about one in which I had worked on behalf of a client of now opposing attorney Colombo. He reminded me of this, of course. Perhaps both Butters and I found it awkward to be deposed by attorneys with whom we had worked previously on the same side. In any case, I know this was true for me. 1. The issue of my database. Attorney Colombo probed deeply about whether or not the database I used was selective or complete. I responded that I analyzed what was sent me as the universe of my information, but that I had no idea whether or not the database represented everything that was out there. For me, the issue was not about the randomness (or lack thereof) of the sample. I was given the totality of the data found by the electronic searches within the time period specified. 2. The issue of what was meant by context. Butters had used a more semiotic definition of “context” than I was comfortable with. He discussed the use of color and other issues from which I tried to steer away. I defined the discourse context as that of a car store and the use of AutoMation as a computerized inventory system that was the first


important step in CarMax’s car store operation. I noted that I was not “a color expert” and could only answer questions about color as a layperson since this went beyond what I considered my linguistic expertise. The extent to which semiotic issues are appropriate to linguistic analysis is an open question. Butters is comfortable with it; I am less so. This implies only our different opinions, not what is right or wrong. 3. The issue of the slogan/expression marks. I explained that “better” and “new” were clearly different words but that they overlap a great deal in meaning, as is clear from their entries in any dictionary definitions. Citing such dictionaries, I pointed out that “better” indicates an improvement over something that preceded it. I agreed that “a used car” was a more specific category of automobiles than “a car” and that the plural “-s” morpheme meant more than one, but I maintained that these were superficial differences. 4. The issue of AutoNation vs. AutoNation USA. I agreed that it is commonplace for journalists first to mention the complete name, in this case AutoNation USA, then to abbreviate the name in later references, in this case as AutoNation. I also agreed that in CarMax’s advertisements, the mark, AutoMation, regularly occurred closely connected to a picture of a computer screen. This, of course, made little difference to me since I had already indicated that the discourse context was that of a car store in which this computer inventory process was central. It became clear from this deposition that this issue was going to be a major one at trial. 5. The issue about what a reader/listener will understand. I was asked what I meant in my report when I said, “the names ‘AutoNation’ (with or without ‘USA’) and ‘AutoMation’ and the expressions, ‘The New Way to Buy Used Cars,’ and ‘A Better Way To Buy A Car’ may be capable of confusion by consumers.” I replied, as I always do, that there is no way that a linguist can get into the mind of a consumer. The linguist examines the text and determines the range of possible meanings that the language of the text can reveal, including its inferential potential. It is possible to tell what the language used is capable of meaning but it is not possible to tell what was actually meant by the sender or what is actually understood by the receiver. If the terms and expressions look alike, sound alike, and mean alike, there is a rather good potential for consumer confusion, but there is no way for linguistic analysis to determine whether or not consumers were actually confused.


Conclusion of the case After expert reports and depositions, the next step was the trial itself. Butters and I had been through this before and were by now anxious to get the whole thing over. From the time of our depositions we had both done some consulting with our respective attorneys and prepared our thoughts for trial testimony. In my case, however, things were a bit different, for since the time of my deposition, CarMax had changed law firms for this case. The new law firm’s attorney, Richard Knoth, came to Missoula so that we could get oriented to the way we both worked and to feel somewhat comfortable with what we both were going to say and do. To make matters worse, I was beginning to feel that CarMax did not have as good a case as AutoNation’s. In October, 1998 the jury trial began in West Palm Beach, Florida. CarMax sought $477 million dollars in damages based on what it believed the AutoNation logo and marketing slogan would be worth over a period of 20 years. It also sought punitive damages, claiming that AutoNation’s executives deliberately tried to confuse customers. AutoNation, of course, vehemently denied this. The week of the trial, both Butters and I were flown to West Palm Beach and housed in carefully selected hotels that were nowhere near each other. We knew better than to discuss this case with each other up to that time, but the attorneys, knowing that we were good friends, did not want to give us the opportunity to meet even casually. We spent whatever time we could with our respective attorneys, but such time is obviously limited once the trial begins. So we sat waiting in separate law offices through the eight-day trial. I cannot vouch for what Butters was being told but I know that I was kept pretty much in the dark about what was happening in the courtroom. By Friday, I was told that my testimony would not be needed but I wasn’t told why. I later learned that on that Friday the judge decided that much of Butters’ testimony had already been duplicated by a marketing expert who had just testified. Since I was supposed to rebut Butters, mine wasn’t needed. So I flew back home. Only much later did I learn the six-person jury’s verdict in this case. Surprisingly, both CarMax and AutoNation had been claiming victory. The jury’s verdict was that there was no trademark violation by AutoNation. On the other hand, the jury awarded CarMax 50 million dollars in punitive damages, concluding that AutoNation had purposely cut into CarMax’s business by using its AutoMation trademark. In a halfday of deliberation after the eight-day trial, the jury also found that CarMax’s marketing slogan was not distinctive enough to be protected,


but at the same time ruled that AutoNation had purposely engaged in unfair competition by using that slogan. The good news for AutoNation was that the jury verdict meant that it was not forced to give up the name and slogan for which it had invested millions of dollars. The bad news was that up to that point, AutoNation had not been able to register these trademarks with the government since such registration was pending until after this case was settled. The punitive damages awarded to CarMax might be expected to affect their achieving such registration. But a few days after the jury’s verdict, on December 12, 1998, the judge vacated the judgment of the 50 million dollars in punitive damages. CarMax’s appeal was denied on November 29, 1999, and the lower court judge’s act of vacating the punitive award to CarMax was affirmed. Since the trial, AutoNation has all but abandonded the used car business to focus instead on new car sales. While CarMax has held on to its some of its superstores, the original idea that prompted the whole operation, developing a user-friendly way to buy used cars, has not yet appeared to have taken the nation by storm. In retrospect, it may have been unwise for me to have taken this case on. AutoNation USA clearly had the better evidence and Butters brilliantly developed and brought it out. But the case illustrates an important point none the less. For one thing, no matter what the evidence, it is the obligation of linguistic experts to use all the resources available to them for their client’s good, and to do this as effectively as possible. However, in doing so they must adhere to the fundamental principles of science. They cannot omit data, distort scientific evidence, use secondary rather than primary data, choose data selectively, use unvalidated data, or manipulate the data to serve their own purpose. They may make points that are technically weak, but when this happens and opposing points are shown to be stronger, there is no more honorable (and effective) path than to concede. Nor can they take on the attorney’s role as trial advocate. If Butters and I had testified at trial, it might have been the case that each of us would have had to concede some things. In my opinion, I would have been the worse for this, since, as noted earlier, CarMax’s case was less compelling.

12 Using Foreign Language Words in Trademarks

The use of foreign languages in trademarks is of great interest to most linguists. Some countries, such as Portugal, Spain, Argentina, Mexico, and China have prohibited nationals from using foreign languages at all in their trademarks, believing that such use involves an inherent likelihood of deception with regard to the foreign origin of the products. The issues of descriptiveness and suggestiveness are also interesting linguistic issues when foreign language words are part of a trademark. Should a mark’s descriptiveness or suggestiveness be considered from the perspective of speakers of the country where the protection is sought or from those who speak the foreign language itself? Can a mark be descriptive or suggestive in English when it is not in French, Spanish, or Arabic? Is a French name for expensive perfumes more suggestive than the English translation of that mark? Then there is the doctrine of foreign equivalents, noted in chapter 3. McCarthy discusses the problems that occur when the foreign language word would or would not have a descriptive connotation to American buyers. McCarthy appears to be troubled by this doctrine in many ways, since it depends greatly on the knowledge and acquaintance of consumers with the specific foreign languages or foreign language words. In 1980, this doctrine was tested in a slightly different way when a seller of high-priced furs and fur products in New York City charged another fur and fur products dealer with infringing on the foreign language derived names of the two companies. Let us begin with the fact that Alexander is common English language name. One of the companies used a French version of this name, Alixandre, while the second company used a Greek version of this name, Alexandros. The involement of opposing linguists turned primarily on the pronunciation of these two company names. 144

Using Foreign Language Words in Trademarks 145

A company called Alixandre Furs had been doing business at 130 West 30th Street in New York City continuously since 1955, adopting the Alixandre trademark, registered in the US Patent and Trademark office in 1956, in the sale of fur wearing apparel. A second company, Alexandros Furs, located nearby at 150 West 30th Street, began doing business in September, 1980. This area of New York is known as the “fur district,” and it might seem unusual that two companies on the same block have such similar names.

Alixandre Furs, Inc. v. Alexandros Furs, Ltd. and Alex Furs, Inc. Noticing the similarity of the names, Alixandre Furs brought forth a complaint charging that Alexandros’s use of this name was calculated to deceive the trade and purchasing public into believing that defendants’ fur products were derived from or somehow were sponsored by plaintiff, an unfair business tactic which tended to confuse the purchasing public and create the false impression that the Alexandros fur products were manufactured and designed by plaintiff, or that there was some actual connection between the two companies. Shortly after the complaint, Alixandre filed for a preliminary injuction against Alexandros, which was followed by a hearing on November 20, 1980, an opposition memorandum by Alexandros, and a reply memorandum by Alixandre to Alexandros’s oppositon to the request for a prelimary injuction. Alixandre’s request for preliminary injunction against Alexandros noted the similarities in looks, sounds, and meaning of the marks, Alixandre Furs and Alexandros Furs. No complaint was made about the use of defendant’s other trademark name, Alex Furs. The fact that the two fur dealers were located in such proximity furthered the issue of confusion, according to Alixandre’s complaint. The gists of the claims of both parties are as follows: Defendant Alexandros’s claims Difference in sounds Addressing sound and meaning similarities between the two marks, defendant Alexandros relied on the assistance of an alleged linguistic expert (who will remain unnamed here), who claimed that there was no difference between the pronunciation of Alixandre and the somewhat common English masculine name, Alexander, specifically referring the sounds of the last syllables, which are reflected in the spellings of


“-re” and “-er.” This linguist said that these two names yielded the same pronunciations of “-er.” That is, he claimed that English speakers would see the spelling of Alixandre, which he identified as from French, and consider it the same as it might be pronounced in English. If his observation were true, it was hoped that the descriptiveness of Alixandre’s mark could be established, since it was nothing more than a variant spelling of a common English name (albeit from French). Defendant Alexandros, admitting that Alixandre was a French name, attempted to buttress the opinion of the similarity of pronunciation by citing the variant spellings of “center” and “centre,” both of which are pronounced following the “-er” sounds in English. Difference in meaning Defendant’s expert linguist also claimed that Alixandre’s French name is feminine in gender, whereas their Alexandros Greek name is masculine in gender, differentiating the meaning of the two marks. Difference in looks Without use of its linguistic expert, Alexandros submitted a brief which compared the script-style mark of Alixandre with the block letter “A” design of Alexandros. The name “Alexandros” was written in script up the left side of the block “A.” The claim was that the differences in Alixandre’s script differed significantly from Alexandros’ block “A” with its name written in script. Plaintiff Alixandre’s claims Similarity in sounds Alixandre’s attorneys addressed this issue with several types of evidence. For one thing, they noted that if Alixandros’s representatives would only have telephoned Alixandre’s listed number, they would have heard the phone answered with the French pronunciation of the company’s name, the same way it had been prounounced for the previous 25 years, in contrast with the final syllable “-er” pronunciation claimed for it by defendant Alexandros. Alixandre’s attorneys also cited Webster’s Third New International Dictionary’s entry for “Alexander,” noting that the listed name is pronounced “alig’zander,” and not the way Alixandre Furs pronounced its name. It was on this issue that Alixandre’s attorneys called me. On November 24, 1980, after plaintiff’s complaint was filed and the hearing had


taken place, I was called by attorney, Simor Moskowitz, who asked me how Alixandre normally would be pronounced. The name was spelled for me but not given to me orally. I advised him that in my opinion and based on my experience, it would be said, “al-ix-’andre.” The second “a” sound would rhyme with the words, “pond” or “fond.” The last “-re” would be pronounced in the French style, sounding something like “-ruh” (not “-er”). This seemed to please him, so he then asked me to write an affidavit saying this, which I did, submitting it on the evening of that same day. My affidavit also noted that language must always be seen in its social context, and this includes names. Names that have a foreign origin or spelling, such as Alixandre, particularly names of expensive or exclusive products (such as designer clothing, furs, or perfumes), tend to preserve a foreign pronunciation longer and more successfully than do names associated with inexpensive or common household products. As for the opposing linguist’s use of the words “centre” and “center,” I explained that “centre” is the British spelling of “center,” borrowed originally in both pronunciation and spelling but no longer used in pronunciation, even in Britain. I noted that the French pronunciation of the “-re” of “centre” parallels the French pronunciation of the “-re” in “Alixandre,” as described above. I also noted that the alleged similar pronunciation of “centre” and “center” is not always applicable to English words ending with “-re,” as is noted by the different pronunciations of “timber” (trees or wood) and “timbre” (the quality of a sound given by its overtones). Similarity of meaning I was not asked to comment on issues of meaning. On their own, Alixandre’s attorneys cited a book, Name Your Baby (Bantam, 1980), for name entries and descriptions of both Alexander and Alexandra. They noted that Alixandre was not listed, leading the attorneys to claim that it is a coined, arbitrary mark. They also took issue with the opposing linguist’s claim that Alixandre was a feminine name while Alexandros was masculine, noting that since Alixandre was not listed among the names in this book, there was no proof that it was feminine in gender. On the contrary, the attorneys continued, that book of names for babies described Alexandros as the Greek equivalent to the Italian name Alessandro, to the Spanish equivalent, Alejandro, and to the French and Portuguese equivalent, Alexandre. Thus, the attorneys argued, if Alexandros and Alexandre are equivalent male names, the public will


assume that Alixandre and Alexander are also equivalents. This reasoning, I assume, was intended to destroy defendant’s claim that there was a difference in gender meaning of the two names. Similarity of looks The attorneys pointed out that it is true that plaintiff’s registered trademark is in script. But Alexandros is also used in block letter form, as defendant’s own exhibits selectively showed. Although Alexandros did not mention this, their own mark had appeared in a script form as well as in block letters. At the hearing, Alexandros had relied heavily on their stylized script letter “A,” and plaintiff pointed out that this stylized “A” does not appear in published articles about that company in Women’s Wear Daily (November 11, 1980 and November 13, 1980). The issue here appeared to be more one for the expertise of a graphics designer or handwriting expert, than a linguist. Without the assistance of linguistic expertise, therefore, Alixandre’s attorneys demonstrated what they believed to be visual similarities of the script and block letters used by both companies. Despite the use of foreign language words in this case, it appears that the doctrine of foreign equivalents may not have been seriously tested after all. The words in question here are proper names, not exactly the type of translatable terms found in the cases discussed by McCarthy (1984: 464–6). The doctrine of foreign equivalents specifies that to American buyers familiar with the specific foreign language, the word would have a descriptive connotation. It is difficult to imagine how either of the proper names Alixandre or Alexander could be found descriptive of a fur product in the way that Telechron (electric clocks) might be argued as descriptive because of its Greek meaning of ‘time from a distance’ (for the record, it was not so found). Defendant’s implication that Alixandre was descriptive because it signified a masculine name like Alexander in English tried to create a quite different meaning of descriptive than seems to have been intended by the doctrine.

Conclusion of the case The last document made available to me was plaintiff Alixandre’s reply to defendant Alexandros’s opposition memorandum for a preliminary injuction, to which my full affidavit was attached. Alixandre argued, in addition to the above arguments about similarity of sound, look, and meaning, that there was a strong likelihood of confusion by consumers based on their similarity of business within the same block of New


York’s fur district. Defendant was asked to remove any signs, labels, advertising, etc. which use the Alexandros mark, and should, instead, operate under any of the other names being used, such as Alex Furs, that were not likely to be confused with the trademark, Alixandre Furs. As in a few other cases I’ve worked on, I was never informed about the conclusion of this case. My search of legal references to it yielded nothing, suggesting that the case was settled before trial. Like other linguists, I am still anxiously awaiting the outcome of a salient case in which the doctrine of foreign equivalents applies.

13 Disclaiming Dealership Authorization

The right to resell a genuine branded item in an unchanged state normally carries with it the right to advertise to others that the dealer is selling that product, as long as such advertising does not mislead the public into thinking that the dealer is an agent or authorized distributor of the owner of the trademarked item. McCarthy elaborates on this: One court observed that when plaintiff’s goods were usually sold through an authorized dealer, defendant’s use of plaintiff’s marks in selling genuine goods can lead customers to mistakenly think that defendant is part of plaintiff’s authorized dealer network. Therefore, the distributor who is not an authorized dealer can use the trademark of the goods, but not in a way that would suggest that the distributor is part of the authorized dealer’s network. However, the unauthorized retail sale of a product in a type of outlet that the manufacturer does not approve does not per se create a false impression that the retailer is an authorized dealer. (1997: 25.43) Therefore, if resale of a trademarked product causes damage to that mark, the reseller is subject to the charge of trademark infringement. One remedy for this is to produce a disclaimer. Disclaimers are known in the US, UK, Scandinavia, and selected other parts of the world. In the following case, the manufacturer of a line of hair products specified that the sellers obtain specialized knowledge and training in the sale and use of its line of hair care products. These products were to be sold by authorized dealers, such as hair salons operated by qualified specialists who had the knowledge and skill to instruct customers to use those products properly and without causing damage to the user’s hair and skin. If the products were sold elsewhere or under other conditions, 150

Disclaiming Dealership Authorization 151

a clear and readable disclaimer must be made, alerting the consumer to the possible danger of using such products without supervision and advice, and indicating clearly that the seller was in no way associated with the manufacturer of the trademarked item. Under most conditions, the sale of these products is precluded at most drug stores and otherwise popular outlets. Although the mark’s owner sells only to qualified retailers, it can happen that certain merchants are sometimes able to acquire the products in other, unknown ways, possibly through the resale of bulk quantities that are purchased from the authorized dealers themselves.

Matrix Essentials, Inc. v. F & M Distributors, Inc. However it came into their hands, F & M Distributors had placed Matrix Essentials hair care line on its shelves for retail sale in several of its chains of stores. The issue in this case was not that F & M stores had Matrix products in their possession, but rather whether or not the sale of the products caused damage to the mark’s owner. In such cases it is required that the seller prominently display a disclaimer that would make clear to customers that they are purchasing from a seller that is not an authorized hair care professional and that has no other relationship with the product’s maker. Black’s Law Dictionary defines “disclaimer” as follows: 1. a renunciation of one’s legal right or claim 2. a repudiation of another’s legal right or claim 3. A writing that contains such a renunciation or repudiation. (Black’s Law Dictionary: 76) In the fall of 1991, upon hearing reports that Matrix Essentials products were being sold at F & M stores, representatives of Matrix checked two F & M stores to see if they had disclaimers on display and, if so, to see what those disclaimers looked like. One of the two stores had no Matrix products on display but it did have a disclaimer located near the professional hair care products shelves. It was a hand-printed sign, in block capitals, approximately –12 to 1 inch high, done with a black magic marker pen. It was printed on an 8 ¥ 10 inch piece of cardboard and mounted on a silver stand. The stand was positioned on the edge of the top shelf of the products offered, toward their left. This sign read: F & M’S SALE OF PRODUCTS DOES NOT CONSTITUTE ANY REPRESENTATION THAT F & M IS AN AUTHORIZED DISTRIBUTOR OF


SUCH PRODUCTS OR THAT F & M HAS ANY OTHER RELATIONSHIP WITH MANUFACTURERS. The second F & M store investigated did have Matrix products available for sale. A disclaimer was present, typset on a piece of cardboard approximately 5 ¥ 8 inches in size and mounted on a metal frame stand. The lettering was approximately –12 inch high. At the top of the sign, typset in yellow letters on a red background, were the words, “F & M IF IT’S FOR HAIR IT’S HERE.” Approximately 1–12 to 2 inches below that, the following statement was typeset in black letters on a yellow background: F & M’s sale of products does not not constitute any representation that F & M is an authorized distributor of such products or that F & M has any other relationship with the manufacturer of such products. With this information, Matrix Essentials’ attorney, Louis Colombo, called me to make a linguistic analysis of these disclaimers. On September 24, 1991, I submitted the following report to him: The impact of communication is a direct result of two factors, specificity and unpredictability. The more specific a communication is, the more it will be noticed, attended to, and comprehended effectively. Conversely, the more broad or general a communication is, the less it is noticed, attended to, and comprehended effectively. Unpredictablity of communication is at the heart of effective advertising, speech making, creative writing, and virtually every other form of communication. The more predictable, stereotyped, trite or clichéd the communication is, the less impact it will have on receivers of that communication, that is, the less it will be noticed, attended to, and comprehended effectively. The purpose of a disclaimer is to clarify a potentially ambiguous relationship of a potentially false assumption on the part of a receiver by stating the true facts about such potentially false assumptions and relationships. In order for a disclaimer to be effectively presented, the following conditions must obtain: 1. It must be placed contextually contiguous to the item for which the disclaimer is made. 2. It must attract the receiver’s attention.


3. It must be relevant to the receiver’s state of mind at the time the disclaimer is appropriately received. 4. It must be legible and comprehensible to the receivers. 5. It must follow the four maxims of effective communication outlined by linguistic theory and research: a. be relevant b. be clear, orderly and brief c. be truthful d. be as informative as is required. The disclaimer displayed by F & M Distributors, Inc. falls short of meeting these conditions, as follows: 1. Empirical evidence of site visits to F & M stores indicates that in one store, the disclaimer sign was displayed some 8–9 feet to the left of where the Matrix products appeared. The disclaimer was not centered on the display, but, instead, was well off to the side. 2. Disclaimer signage not contiguous to the product for which the disclaimer is relevant will not attract the reader’s notice and attention. 3. The structure of the disclaimer sentence does not follow widely recognized principles of recipient design. That is, the sign does not begin with the consumer’s frame of reference or state of mind. Instead, the sentence fronts F & M and its sale of products as the first words on the sign. Words that address the consumer, such as “you,” “your,” “customers,” “purchasers,” etc. are totally absent. The message is entirely about F & M, while the consumer’s role is at best implied, or at worst ignored. 4. It is a well-recognized principle of document design that sentences printed with all capital letters are more difficult to read and to process than are sentences written in more conventional orthography. In one site where the F & M disclaimer was displayed, the sentence contained all capital letters. Readability is highly questionable in such cases. Comprehensibililty of the disclaimer’s syntax also is highly questionable, at least for average readers. The disclaimer contains one 30word sentence. The average reader can effectively process sentences of 10–15 words in length. The disclaimer’s sentence contains three clauses (one main clause and two dependent clauses) and four prepositional phrases. Such a sentence cannot be considered the product of a writer’s careful effort to be comprehensible and clear. Equally salient is the disclaimer’s use of negation. A disclaimer is, by definition, a negative concept. As such, it requires a negative. The most


logical negative would be one that attaches to the authorization, i.e., “F & M is not authorized to. . . .” Instead, in this case, we find the negative transported to the verb, “constitute,” following a heavily nominalized subject: “F & M’s sale of products does not constitute any representation that F & M is an authorized distributor. . . .” Thus the negative is primarily attached to the words, “constitute any representation.” If one manages to process the rest of the sentence, one can figure out what that representation relates to, but the physical association of the negative with “authorized” has been successfully avoided. The second dependent clause is accomplished in the same way, by avoiding the use of the negative with the words, “has any other relationships.” Having placed the negative with “constitute any representation,” it became grammatically unnecessary to use it where the negation most closely serves the appropriate verb. To be most comprehensibly negative, the negative should be contiguous to the verb it negates. For example, had the disclaimer sentence said, “F & M is not an authorized distributor” and “F & M has no other relationship,” the meaning and force of the negatives would have been much clearer to the average reader. But by couching the negative in the independent clause, associated with the constitution of F & M’s representation, it was grammatically possible to avoid the negative in the two dependent clauses where the negative must be comprehended in order for the intent of a disclaimer is to be processed and understood. Therefore, although the sentence is grammatical, its comprehensibility is highly questionable and its effectiveness as a disclaimer is seriously reduced. 5. It has already been pointed out how the maxims of relevance has been defused in this disclaimer. The 30-word sentence begins with F & M’s sale of products as its main, independent clause. It appears, from this, to be more about F & M than about the needs of the consumer. Although the disclaimer may be considered brief, it can hardly be considered clear (as noted above) or orderly, placing as it does the main idea of a disclaimer in the dependent clauses. I was not asked to discuss the truthfulness of the disclaimer. As for informativeness, the disclaimer lacks specificity. It is an allpurpose disclaimer, apparently applicable to any product sold in the store for which authorization to sell has not been awarded. Consumers cannot be expected to know which products are authorized items and


which are not. The general, all-purpose disclaimer, therefore, is relatively useless to the consumer. Its usefulness can be only to F & M, to protect F & M from whatever penalties might accrue from not displaying a disclaimer. The maxim of informativeness specifies that the communication should be as informative as is required (including not pointing out things that consumers need not know or be informed about). All-purpose, general disclaimers of a company policy constitute both information overload for consumers, since company policy is not particularly useful to the process of purchasing individual products, and infomation underload, since consumers can have no idea which products are covered by the general, all-purpose disclaimer. Conclusion The disclaimer sentence offered by F & M Distributors, Inc. is inadequate and incomprehensible in that: • it is not contextually contiguous to the Matrix products for which it disclaims authorization; • it does not attract the customers’ attention; • it is not relevant to the customers’ state of mind; • it is not readable when appearing in all capital letters; • its 30-word, three-clause, four-prepositional phrase construction, with the main point placed in dependent clauses, is not easily or effectively understandable; and • it violates three of the four maxims of effective communication: relevance, clarity, and informativeness. F & M then called its own expert, who will remain anonymous here, to rebut my report. It should be noted that this expert advertises his availability in law journals but, from his curriculum vitae, it appears that his training and experience are only marginally related to forensic work. In his rebuttal report, the opposing linguist opined that the F & M disclaimer was written in “good, clear, straightforward English, leading directly to an untortured conclusion.” From this description, it might appear that he had seen different F & M disclaimers than the ones that I had reviewed. Using them, he disagreed totally with the points I had made in my report. Both the opposing expert and I were deposed late in the final months of 1991. In my deposition, which was relatively brief as depositions go, the opposing attorney posed challenges to my report especially relating to the sources of my linguistic opinions, as follows:


• my definition of “disclaimer” (several dictionaries); • my list of conditions for an effective disclaimer (primarily based on H.P. Grice’s (1975) maxims), and • research supporting my claim that capital letters are more difficult to read and process than regular script (Felker et al. 1981; Foster and Coles 1977; Poulton and Brown 1969; Tinker 1969). The opposing expert’s deposition might be described as more difficult for him than mine was for me. He admitted that F & M’s disclaimers did not explicitly mention the consumer or customer. He argued strongly, however, that they did so implicitly. Interestingly, he testified that the use of “hair” in the beginning words, “F & M IF IT’S FOR HAIR IT’S HERE,” is an explicit mention of the consumer, asking “Who else has hair?” He said further that the disclaimer’s use of the word “store” was what he called “a necessary implication” that consumers were mentioned. To illustrate the differences this expert had with my analysis, I quote the following exchanges from his deposition by Matrix’s attorney: Q: Would you agree that good clear English requires the use of definite, specific and concrete language? A: No. Q: Would you agree that to write effectively you should omit needless words? A: Not always. Q: Would you agree that to write an effective disclaimer, you should omit needless words? A: Not sure. . . . needless in whose eyes? [. . .] Q: When you review . . . disclaimers . . . are there any fundamental rules that you apply? A: No. . . . I don’t think I’ve ever stepped back and laid down guidelines. [. . .] Q: Is a principle of modern English writing that you should attempt to keep related words together? A: . . . I do not universally agree with it. Q: Would you agree in a disclaimer it’s a good idea to keep related words together? A: I’m not sure. . . . I haven’t thought about it. [. . .]


Q: You write, “Since when is ‘all capital letters’ hard to read?” Are you familiar with any studies that have been done on the question of whether or not it is more difficult or less difficult to process information printed in all capital letters? A: I believe I’ve read studies or have read of studies about that . . . I believe I read in the newspaper, once upon a time, that a study showed that speech scripts were clearer if they were not in all caps. . . . I haven’t read them. [. . .] Q: [In your report] you talk about Professor Shuy’s statement that 31 words are too long for a sentence in a disclaimer. . . . Can you refer to any studies that have been done regarding the effectiveness of sentence length? A: No. [. . .] Q: In the evolution of the English language over the years, has there been an overall trend to shortening sentences? A: No. The opposing linguist went on to say that the King James Bible is the touchstone for correctness of our language. He completely misunderstood and misinterpreted one of the sentences in my report and, when asked what other criticisms he had of my analysis, listed the following things: • My analysis was labeled “linguistic” analysis but I said things that were other than linguistic. • My use of “specificity” was “terribly vague” (although later he stated that it was not vague). • My statement, “the purpose of a disclaimer is to clarify a potentially ambiguous relationship or a potentially false assumption on the part of the receiver” was “nonsense.” He claimed that there are other purposes as well. When asked to list some, he mentioned, “to keep from being sued,” “to show off,” “to show how clever the writer is,” and “to show that the author is a poor writer.” • My use of Grice’s maxims of communication, all of which he he disagreed with, noting: “it sounds like double talk . . . a tautology.” Conclusion of the case Trial was scheduled for June 15, 1992 but it never happened. Apparently negotiations took place for the next few months and, on December 10,


1992, I received a letter from attorney Colombo’s office informing me that Matrix had settled its litigation with F & M Distributors and that there would be no trial. F & M requested that the terms of the settlement agreement be confidential but, the letter concluded, “I can tell you that Matrix is satisfied with the settlement.” If there is an obvious lesson to be learned from this case, it is that not all forensic linguists are the same, just as not all who consider themselves linguists are the same. It is unthinkable to me that a current, practicing linguist would, like the opposing expert in this case, be unfamiliar with Grice’s maxims and some of the more obvious grammatical issues discussed here. I felt sorry for the opposing attorney and if any lesson can be taken from the use of linguistic experts in this case, it is that lawyers should do considerable checking before hiring one. One place to check is with linguists who have already testified in trademark cases. They have some knowledge of what is expected of a linguistic expert is such cases and even may have run across that expert in other cases.

14 Using Linguistic Tools and Thinking in Trademark Cases

Chapter 2 briefly described some of the analytical tools available to linguists. Mentioned were phonetics and phonology, morphology, lexicography, syntax, and meaning (semantics and pragmatics). Chapters 4 through 13 described actual trademark cases in which various these tools were used. For purposes of emphasis, I now offer a summary of which linguistic tools and techniques were used in each case. The most frequently used tools were morphological and semantic analyses, with phonetic and lexicographical analyses slightly less frequent. In these cases, the use of syntactic and pragmatic analyses was somewhat surprisingly uncommon.

Using linguistic tools 1 In the Registry Hotel v. Hospitality Management case (chapter 4), the linguistic issues were fairly straightforward. Was “hospitality management” generic? If we could show that it was, there would be no need to analyze its descriptive characteristics. The linguistic tools used were lexicography and morphology. Lexicocraphy and morphology I posited that generic words would have the features of grammatical flexibility and the ability to be recently attached to another generic word, in this case the one that was central to the case. Two generic words do not normally make for uniqueness or suggestivity. I traced the lexical history of the word “hospitality” from the early French from which it was borrowed to modern-day uses in English, showing that there had been no meaning change throughout a period 159


of some 700 years. Finally, I checked current-day usage in the media and found a corpus of uses showing that the expression, “hospitality management,” was currently in broad usage in a variety of different contexts. 2 In the Warren v. Prestone case (chapter 5), the linguistic applications were to the areas of lexicography, morphology, and semantics. Lexicocraphy Prestone played the lexicography hand first, but used only one of the 18 senses listed in the Oxford English Dictionary that define “life” and only one of its 12 senses that define “long.” It was not difficult to prove this to be a very selective and incomplete use of the available dictionary entries. Then I went on to show how a leap of imagination was required to use the dictionary’s single inanimate definition of “life,” which could only be considered metaphorical. Morphology I pointed out that Prestone’s analogy of “long life” with “long-lived” was actually a morphological dissimilarity, “long life” being a construction in which an adjective is joined with a noun to create the proper noun, LongLife, while “long-lived” was an adjective joined with a verb to create an new adjective. Prestone’s analogy was simply not morphologically equivalent and was therefore not a useful indicator of Prestone’s alledged meaning of “long” and “life.” Semantic meaning In terms of semantic meaning, I addressed Prestone’s claim that its selective, individualized meanings of “long” and “life” had remained the same over time, in their attempt to prove that Warren’s LongLife mark was generic term for a class or type of antifreeze/coolants. In response I pointed out that there were actually two meanings, not one, used by Warren’s LongLife. One meaning was for the long life of the vehicle and the other for the long life of the antifreeze coolant. That a mark contains two meanings is an example of where the consumers need to exercise thought or imagination. If there had been any need to declare Warren’s LongLife a suggestive mark, I was prepared for the imagination test. This was never found to be necessary, however. In this case it was Prestone that first conducted a NEXIS search in an

Using Linguistic Tools and Thinking in Trademark Disputes 161

effort to show that “long life” was a category or type of antifreeze and therefore was generic. Unfortunately for them, their search turned up many false positives that were easily discounted and two alternative terms, “extended life,” and “eternal,” which later proved to be difficult for them, since these words gave a clear indication that there were other terms for that category of antifreeze and that “long life” was not generic even in that sense. 3 In the ConAgra v. Hormell case (chapter 6), the primary linguistic issues were phonetics, morphology, and semantics. Phonetics There was not much disagreement between the opposing linguists about the phonetic differences between Healthy Choice and Health Selections. We agreed that eight of the 13 phonetic sounds were different. We disagreed about the importance of the beginnings of words for consumer comprehension. Morphology I agreed with the opposing linguist that both marks were compound nouns that take a singular verb grammatically. He disagreed with my view of the important morphological difference between the noun–noun compound, Health Selections, and the adjective–noun compound, Healthy Choice. He considered the plural “-s” form of Selections trivial; I considered it important, noting that it means something that was not present in Healthy Choice. Semantics The opposing linguist claimed there was no meaning difference between Healthy Choice and Health Selections. He placed his opinion on the ability of consumers to get the gist of the meanings, which he said were one and the same. In contrast, I pointed out that “Choice” and “Selections” have different meanings. For “Selections,” somebody else picks out something for the consumer while for “Choice,” the consumer has to do the picking. I also noted that the meaning of “Selections” involves an array of things to select from among, while “Choice” narrows that array down to two things to choose between. I noted that acceptable English usage is to use the word, “between,” with two items and the word, “among,” with more than two items.


4 The Woodroast v. Restaurants Unlimited case (chapter 7) involved the linguistic tools of morphology, semantics and syntax. Morphology It was clear from my database that Shelly’s Woodroast used “Woodroast” as a proper noun while no other comptetitor did. Other restaurants used “wood roasted,” sometimes with a hyphen between the morphemes, both in their company names and for their menu items. My electonic searches of media articles showed that they invariably used the common noun forms of “wood roasted.” Lexicography My check through many dictionaries yielded no entry for “woodroasted,” either as a proper or common noun. Proper nouns are capitalized, as “Woodroast” was. Common nouns are not, as were none of the references to “wood roasting” or “wood roasted” found in any of the data base references to anything but Shelly’s Woodroast. Semantics From Woodroast’s own menu it was clear that that “Woodroast cooking” was a style of food preparation. Media articles called “wood roasting” a “form of cooking,” one article noting, “wood roasting is the beginning of all cooking.” This contrasted with Woodroast’s claim to be “a new style of cooking.” Having concluded that it was a style of cooking, the next analysis was of the syntax of the menus. Syntax Here I made use of a somewhat old-fashioned approach to syntax, tagmemics, primarily because I thought it would communicate most effectively to a judge or jury. Using the slot and filler approach of tagmemics, I surveyed hundreds of restaurant menus and discovered the syntax of items listed on them. The five slots were: 1. 2. 3. 4. 5.

self-congratulations, method of cooking, style of food, food item, and serving modification.


All slots were optional except for the obligatory slot number 4, food item. Whereas other menus made optional use of slots 1, 2, 3, and 5, Shelly’s Woodroast’s menu never did. It contained only slots 3 and 4. For Woodroast, slot four was the food item, say chicken. Slot 3, the style of food, was filled by “Woodroast.” This supported the semantic analysis that “Woodroast” means style of food and not, as Shelly’s Woodroast claimed, the food item itelf. All other menus used “wood roasted” as method of cooking. Shelly’s did not. This analysis also supported the idea that “wood roasted” is used as an adjective by the other restaurants but not by Shelly’s Woodroast. 5 The linguistic aspect of the case of McDonald’s v. Quality Inns International (chapter 8) was largely one of trying to determine the current meaning of the prefix, “Mc-.” In terms of this trademark case, the task of the two opposing linguists was to show whether or not McDonalds’ proprietary claim that “Mc-” always means McDonald’s, and not anything else, was accurate. Semantics An electronic search was made in media sources for all uses of the “Mc-” prefix when the articles had nothing to do with McDonald’s or with fast food in general. On a wide variety of topics and contexts that were all unrelated to foods, I concluded that the current central meaning of “Mc-” was basic, convenient, inexpensive, and standardized, no matter what the topic of the article was. 6 In the case of AMR Pharm v. Acme Home Products (chapter 9), the salient linguistic tools were phonetics and semantics. Phonetics To find the recommended pronunciation of OVRAL, I relied on the Physician’s Desk Dictionary. Since AMR’s B-OVal had not yet been advertised, I had to rely on the pronunciations given by company officials. I then compared the two and discovered that the sounds of the two marks contain many important differences. Semantics Both OVAL and B-OVal are associated with the Latin, “ovum,” meaning egg. The latter adds an important “B” to its beginning, suggesting


“breast” in this context, one in which the fear of breast cancer had been closely connected with the use of birth control pills. 7 In the case of Pyewacket v. Mattel (chapter 10), the linguistic issues were phonetics, morphology and semantics. Phonetics I pointed out that the marks, GUK and GAK, differed in one-third of their phonetic sounds and that this difference, the vowel, is an important one, especially since vowels form the nucleus of syllables. Morphology GUK has a robust ability to move beyond the confines of its noun form. In the media and advertising materials I discovered the adjectival forms, “Guk bucket,” and “gucky,” and the verbal form, “Guking.” Descriptive words develop the ability to extend themselves into other grammatical forms, as GUK clearly evidenced. GAK, in contrast, showed no such tendency. Semantics By searching many dictionaries, I eventually found one that included the word, “guk.” Other dictionaries include its phonetic equivalent, “guck,” but not the spelling without the “c.” Dictionary definitions of “guck” lead to a synonym trail which leads back again to “guck.” No such dictionary definitions could be found for either “gak” or “gack.” 8 The linguistic issues in the AutoNation v. CarMax case (chapter 11) went across a broad spectrum – phonetics, lexicon, semantics, morphology, syntax, and pragmatics. Phonetics Both the opposing linguists agreed on the number of sounds in the two names up to “USA.” They disagreed about the stressed syllable in AutoMation, the phonetic quality of the “o” vowel in that word, and the significance of the minimal pair, AutoNation vs. AutoMation. Lexicon A huge disagreement between the linguists had to do with what we were comparing, AutoNation or AutoNation USA. I focused on the uses of


the name in the media, advertising, and news programs. Other disagreements were on the meaning and significance of the differences in the slogan uses of the words, “new,” “better,” “used cars,” and “a car.” Semantics The major issue was kindled by the fact that AutoNation was a car store brand name and AutoMation was a computerized inventory browsing system used in a car store. Obviously they referred to different things in the same industry. Another issue had to do with the meaning of “auto-” in both names. Morphology There was no question but that there were two morphemes in both AutoNation and AutoMation. The continuing issue was whether or not to make the comparison with AutoMation be AutoNation or AutoNation USA. Also at issue was the bound form status of -Mation, and the ensuing issue of the use of a noun compound vs. a noun with a bound morpheme. Pragmatics There was a difference of opinion between the opposing linguists about whether or not extralinguistic context could determine the meaning of the marks and expressions. 9 In the Alixandre Furs v. Alexandros Furs case (chapter 12) opposing linguists were used primarily on the issues of phonetic similarities or differences. Phonetics The opposing linguists expressed two quite different opinions about how Alixandre was pronounced, particularly with respect to the final “-re” of Alixandre and the final “-er” of Alexander. The Alexandros Furs linguist also suggested that there was a semantic difference based on the masculine feminine contrast of Alixandre and Alexandros, but plaintiff Alixandre’s attorney argued against this claim without linguistic help. 10 The opposing experts in the Matrix Essentials v. F & M Distributors, Inc. case (chapter 13) differed diametrically, probably because linguistic analysis appeared to be absent in the opposing linguist’s report and


deposition. In short, this case was less valuable as a comparison of differing linguistic positions than it was an example of how attorneys should be careful in selecting linguists to help them with their cases. Nonetheless, the two linguists differed in matters of syntax and pragmatics. Syntax The opposing linguist had nothing negative to say about the syntax of the disclaimer, finding in clear and straightforward English, easily understandable by consumers. I argued the opposite, noting it it is constructed in such a way that it failed to begin with the consumer’s perspective and was composed of a single, 31-word, very complex sentence with three clauses and four prepositional phrases, with the main point well hidden in one of the dependent clauses. Pragmatics In contrast with the oppositing expert’s view, I analyzed the disclaimer using Grice’s maxims, which showed that it had serious problems that opposing linguist was apparently unaware of.

Using linguistic thinking It is probably the case that every field learns a special way of thinking. Lawyers, physicians, teachers, and linguists, all seem to learn slightly different ways of thinking in their graduate training programs. As I recall from my own graduate school teaching, I noticed that the good students began to learn to think like a linguist at least by the second year of their masters degree programs. If they didn’t, they stood a good chance of never making it through the doctorate. None of these characteristics is unique to linguists, but they may offer a picture of some of the important ways that linguists think about their data. Although, to my knowledge, these characteristics of thinking like a linguist have not been codified elsewhere, some of them include features that often set this discipline off from other fields, including the following. Linguists let the data drive their analyses Rather than starting with a thesis, then proving it, linguists begin with the mass of data made available to them and let the data tell them what the solution is. This is called letting the data drive the analysis. Preconceived notions of what is right or wrong have no place in a truly data-driven study. The data facts are what they are.


Linguists look for recurring patterns Many other fields do this, of course, but the type of patterns linguists look for are often very minute and unseen by non-specialists. For example, in the Alixandre Furs v. Alexandros Furs case (chapter 12), the attorneys for both sides were focusing on the “-er” of the last syllable as evidence that the words were pronounced alike or differently. When I first got involved, I noticed that the vowel in the third syllable, spelled “a,” was also different in the two words. The attorneys on both sides never noticed this difference, thinking that it was pronounced to rhyme with “can” or “man,” rather than the French pronunciation of the vowel, “ah,” rhyming with “John” or “fond.” Simply tallying up the number of instances that a given linguistic feature occurs does not always show the maximum patterning that can be revealed. For example, in the AutoNation v. CarMax case (chapter 11), I not only counted the instances in which AutoNation and AutoNation USA occurred, but also discovered a pattern that differentiated the use of these terms found in advertisements written or controlled by AutoNation versus the television news programs that this company did not write or control. Likewise the patterned usage of restaurant menu items and names in the Woodroast v. Restaurants Unlimited case (chapter 7) went far beyond being a simple list. Linguists are willing to disagree with other linguists Like all scientists, linguists do not tend to agree with each other on all points. The academic world thrives on disagreement and, in this respect, linguists are not unique. But lawyers should know this and expect that the linguist may also disagree with them. As preparation for a court case, this can be a real asset. Attorneys should not be surprised to learn that linguists can also disagree with each others’ analyses in trademark cases as well. But, like all scientists, linguists will also agree with the other side when the evidence points only in that direction. Butters and I strongly disagreed on what constituted the proper name to be compared with AutoMation (chapter 11) and on the relative importance of some of our less crucial findings, but on relatively little else. There was considerably less disagreement between my analysis and the findings of the opposing linguists in all but one of the other cases where linguists appeared on both sides. In two of them (chapters 6 and 8), our differences seemed to stem more from different interpretations of what linguistics is and what linguists can do, but these were still major disagreements. In the case of Alixandre Furs v. Alexandros Furs (chapter 12), there was practically nothing that the two linguists agreed


on. The case of Matrix Essentials v. F & M Distributors (chapter 13) did not really show opposing views of linguists, since linguistics played no apparent role in the opinions of the opposing expert. Linguists would rather be accurate than win a case Ten trademark disputes were described in this book. The attorneys I worked with prevailed in six, lost in two, and I have no idea how the other two ended. When there were opposing linguists, my attorneys prevailed in three and lost in two. But it is relatively meaningless to tally up the victories and defeats, since there are many factors that contribute to the conclusions of each case, most of which had little to do with the linguistic analysis. It can never be very clear what the out-of-court settlements actually meant. Three of the four settled cases appear to have favored my clients (chapters 4, 7, and 13) and one appears to have favored the oppositon (chapter 10). But in the two other cases (chapters 9 and 12), I still don’t know the ultimate results. From the linguist’s perspective, winning or losing a court case is not germane anyway, since linguists are not advocates for one side or the other and, as such, linguists neither win nor lose. The linguists examine the data and come up with analyses that might differ from those of another linguist, and whether their side’s cases prevail or not is not the linguist’s main concern. Sometimes, as in the McDonalds v. Quality Inns International case (chapter 8), damaging testimony by Quality Inns’ own president is said to have played a crucial role in the judge’s untimate verdict. Sometimes, as in the AutoNation v. CarMax case (chapter 11), the nonlinguistic facts accompanying the linguistic issues that the linguist has to work with are simply not as salient as those facts enjoyed by the opposing linguist. Linguists use counterexamples One of the common techniques that linguists like to use is that of searching their data, memory, and intuitions to find language examples or counterexamples that may help prove or disprove points. This was illustrated in several cases here, but perhaps most noticeably in the ConAgra v. Hormell case (chapter 6), in the battle over whether or not the beginnings of words were more important signals of the readers’ attention than their endings. This characteristic of thinking like a linguist can work very well for attorneys. Linguists work with whatever data are available Unlike psychologists or psychiatrists, who often use tests and interviews to create their data after an event has occurred, linguists work with the


data accumulated in the event itself. In criminal cases involving tape recordings, for example, the linguist may never even see the participants. The data for analysis consist of the tapes alone. Trademark cases are somewhat analogous. The data are the trademarks, the uses of the words of or related to the trademarks, and the recorded talk about those words by other parties. The linguists’ part in a trademark case is only as good as the quality and quantity of data with which they have to work. I was not particularly surprised or disappointed by the outcome of the AutoNation v. CarMax case (chapter 11), for example, since the facts in that case did not seem to favor my client’s side anyway. There is nothing much that a linguist can do about the quantity or quality of the data. Nor is there anything a linguist can do about a court decision largely based on secondary meaning, as in the McDonald’s v. Quality Inns case (chapter 8). Linguists work with the data they have and do the best they can with it. Unlike many other out-of-court research contexts in which large amounts of data must be elicited and analyzed to prove a point, the data in a law case are prescribed by the limits of the evidence in that case. If the disputed term is in use at a given time, virtually all uses of it can be found in the media, in the public records, in testimony, or through electronic searches. Thus, the linguist considers this corpus to be the universe of data that can be worked with, not just a sample of it. Linguists teach their audiences One of the common threads in these cases is the usefulness of a linguist in helping teach the triers of the facts. This was especially evident in the Warren v. Prestone case (chapter 5), when it was one of my designated assignments to help the jury figure out what the four categories of trademarks mean. In such cases it is always useful to find analogies in real life as pedagogical tools. In this case, I used the hypothetical marks of LongLife Sox and LongLife Vitamins to make certain points. I also called on a television advertisement that was well known at that time, that of the Energizer bunny, which added familiarity to what the jury might have thought of as a seemingly arcane battle over antifreeze names. In all of the cases described here, it was necessary for the linguists to teach attorneys, judges, and/or juries about the qualities and claims of lexicographical practice, and to convey technical linguistic information about the sounds, grammar, meaning, and discourse of the English language in ways that could be grasped by laypersons. If anythings characterizes applied linguistics, it is this.


Linguists quantify their data whenever possible The power of quantitative illustration was evidenced in several of these cases described here. Electronic searches for the current uses of disputed words expanded the database and allowed for simple and understandable percentage comparisons. This technique was especially salient in the Registry Hotel Corporation v. Hospitality Management case (chapter 4), the Warren v. Prestone case (chapter 5) and the AutoNation v. CarMax case (chapter 11). I have found that triers of the facts are impressed when simple percentages are used for comparison purposes. It adds number evidence to what otherwise, rightly or wrongly, might be considered mere anecdotal information conveyed by words. It is also the case that audiences, such as juries, can be impressed by what they hear, but will better remember what they can see. Thus, linguists tend to use charts and graphs to supplement their oral testimony and written reports. Throughout the cases described here are brief and simple charts of various findings, often presented in the form of numbers or percentages.

15 Some Suggestions for Linguists

Linguists who agree to work with attorneys on trademark cases might consider the descriptions of the ten cases described here and consider the following general suggestions. 1. Be a good team member with your attorney, but don’t be become an advocate for his client The expert linguists are perfectly within ethical boundaries in working with an attorney to make the best possible case for the client. Without distorting, being selective, or ignoring important linguistic facts, linguists should use their knowledge to show where marks are similar or different, where they mean or do not mean the same thing, and where the linguistic evidence shows signs of the mark being generic, descriptive, suggestive, or arbitrary. Lawyers expect their experts to be objective team players, not advocates. But, as in sports, team players also want to win and this is where scientific objectivity and advocacy can become blurred. Impeaching a witness is one of the predictable characteristics of trials of any kind. Lawyers can try to impeach the accuracy of what is said, if possible. They can also try to impeach the expert’s qualification, status, background, age, and many other things. For example, the prosecutor in a criminal case in which I was an expert witness for the defense once asked me my age and when I had last had my hearing checked. He made much of the fact that I was then in my sixties and could probably not hear the tape-recorded evidence accurately. This was pointless of course, since good audio equipment can compensate for most minor hearing loss and the issue was more of listening skill rather than hearing. I had not had a hearing test since I had no hearing problem, but the point was made loud and clear to the jury. I was impeached for 171


being 64 years old, not for my expertise, my status, my qualification, my training, or, for that matter, my hearing. This is fair play for attorney advocates I suppose, but it is not appropriate for experts to aid and abet them in the process. For example, in the McDonald’s v. Quality Inns case, I firmly believe that my analysis was the right one. In fact, the linguist for McDonald’s never disagreed with my conclusions. He discounted only the fact that his interpretation of the field of linguistics, as he defined it, could not lead anyone to reach such a conclusion. His testimony was not about my conclusions but it appeared to impeach my definition of the field of linguistics, an internecine battle that in my opinion has little, if any, place in such a venue. I don’t know to this day if his actual intent was to impeach my linguistic qualifications, but the result had the possibility of being the same thing. When one expert gets on the witness stand and indicates, directly or indirectly, that his or her specialization is better than that of the expert on the other side, the authority of the entire field is often what is actually impeached. Some of that same dispute about the acceptable boundaries of linguistics also played a role in the ConAgra v. Hormell case, where the opposing linguist may have implied that his linguistic specialization was superior by describing himself as a “cognitive” linguist. This label allegedly trumped anything that a expert not claiming that label might offer. Again, the battle seemed to be less over the content or methodology of the two analyses than it was over the specializations of the linguists involved. It is obvious that I have chosen not to name the opposing experts in the McDonald’s v. Quality Inns, ConAgra v. Hormell, Alixandre Furs v. Alexandros Furs, and Matrix Essentials v. F & M Distributors cases. I have no desire to discredit them personally or academically. Two of these linguists are very competent and I have no way of knowing for sure whether or not they may have stepped over the line of objectivity and into advocacy on these occasions. The ConAgra linguist, in particular, presented his written position brilliantly. What may have looked like becoming an advocate could have happened simply because most linguists are unacquainted with the distinction between arguing for the overall goals of a law case and being an objective expert witness whose job it is to avoid the very appearace of advocacy, present their best analyses, and let the chips fall where they may. The more experienced expert witness steers a different path. I feel comfortable identifying the opposing linguist in the AutoNation v. CarMax case, Ron Butters, who has testified many times at trial and

Some Suggestions for Linguists 173

remained objective and professional throughout the process. To be sure, our analyses do not always agree, but when confronted with information that we have overlooked or misunderstood, we both do the honorable thing. We modify our positions. As noted earlier, the ironic thing about this ability to concede when necessary is that it often works to the advantage of the case by humanizing the witness in the minds of the triers of the facts. In any event, experts can look very foolish when hanging on to a position which has been become quesionable. Better to stand corrected and move on. Unfortunately, life provides us more opportunities to do this than we may fully appreciate. By saying that linguistic experts should not take on the role of advocacy, I should clarify that they most certainly should be team players on the side they are asked to serve. Their job is to make the best case they can with the data available. Sometimes those data can be taken only so far, as was clear in the AutoNation v. CarMax case (chapter 11). Experts cannot disguise the facts or ignore difficult data. Their role as dispassionate scholar is crucial and should not be compromised by the desire to help their side win the case, however tempting it may be to do so. 2. Give a little when it is right to do so In my early years as an expert witness, I was often dogged in my determination never to give in to the other side, whether right or wrong, giving evidence of the fact that I had not yet learned to avoid advocacy very well. Years ago a jury was polled after a criminal case in which I had testified on behalf of a defendant who was subsequently acquitted. They pointed out that they were impressed with everything about my courtroom presentation except that I “just never would give in to anything.” After that I changed my approach, acknowledging the other side’s position when it was reasonable and accurate. Giving in a little seems to make the expert more human and real in the minds of a jury. It has been noted in the AutoNation v. CarMax case (chapter 11) that linguists on both sides were willing to concede points when it became obvious that this was the right thing to do. And they did. 3. Write brief expert reports The less you say in your expert report, the less you may have to say in the almost inevitable deposition that follows. There is a fine line here, though. Expert reports are supposed to make clear to the opposing attorney every opinion that is planned to be presented at trial, as well as everything used to support those opinions. But writing an expert report


is not at all like writing an academic paper. For good reasons, most scholars tend to be verbose and overly detailed. They anticipate every counterargument to their ideas and try to answer it before it is even asked. The context of a law case is a bit different. Experts should still be ready to answer any possible objections and questions, but they do not have to invite such questions and objections in their reports. There is little to be gained by anticipating all objections in this way, since there is often a very good possibility that the opposing attorney will not even think of the objections that the writer might envision. In any case, the deposition is the time and place for the basic opinions of the expert report to be probed and detailed, if the opposing attorney wants to and can think of how to do it. Otherwise, the expert may be better off to be as brief and accurate as possible. 4. Prepare well for the deposition and trial Depositions often occur long after the analysis is made and the expert report has been written. Other things occupy the expert during the interim and it is easy to lose the focus and clarity that were present at the time the report was constructed. Careful and complete file-keeping is essential and an occasional review of the matter, at intervening times before the deposition and trial events can keep the expert on top of the facts. It is often the little things that trip one up and detract from the effectiveness of any presentation. I recall with considerable embarrassment one deposition in which I got confused and used the wrong name for my own client’s company. At that time I had been working on several cases simultaneously and simply did not keep the threads of different cases separate enough. My gaff didn’t hurt the case in the long run, but it provided a laugh or two at my own expense. It could have been worse. 5. Be brutally honest about the downsides of your case It is difficult to work for lawyer-clients and, at the same time, be able to tell them that they may be wrong or that the evidence in their cases has certain flaws. A well-known fact about law is that if lawyers have the law on their side, they argue the law. If they have the facts on their side, they argue the facts. If they have neither the law nor the facts on their side, they just argue as hard as possible. As mentioned above, one way to argue is to try to discredit the other side and to impeach their witnesses. Another way is to prevent unnecessary argument by helping the attorneys you work with see the linguistic weaknesses of their own cases at the earliest possible stages. If the weaknesses that are thereby


exposed are so objectionable to the attorneys that they decide not to use your expertise, this can be the best thing that can happen, both for you and for the attorney. 6. Don’t underestimate the linguistic competence of attorneys and judges Attorneys may not be trained linguists, but they most certainly have inner linguistic skills and the best ones have an amazing breadth of knowledge and ability to assimilate whatever new knowledge experts, including linguists, have to offer them. Most linguists providing expert testimony are professors who, as they deal with students regularly, find, as I have, that attorneys are some of the best students they have ever had. Good lawyers learn new information very quickly, offer intelligent suggestions, ask very good questions, and work extremely hard. Most of the attorneys I worked with certainly fit this description. Bartholomew McLeay, Bruce Little, John Alex, Lawrence Hefter, Gilbert Schill, Louis Colombo, and Richard Knoth, in particular, are all outstanding trademark attorneys and intellectually exciting people with whom to work. Good linguistic ideas may not prove to be good trial ideas and it behoves the expert linguist to listen to the attorney’s view of what will be effective and what will not. One of the worst things a linguist can do is to insist that some bit of linguistic information be included when the advice of the attorney is that it should not be. The odd position that any expert faces is that of communicating technical expertise without overwhelming the jury or judge with information or language unnecessary, or even counterproductive, to the context of a trial. As teachers, we have control of what we think our students need to know but we do not have similar control in the courtroom. You should not cut corners, of course, but exactly how much of our knowledge a judge or jury needs to know is often better assessed by the ones who live in such contexts daily. The preparation phase in any legal issue should work out such matters well in advance of expert reports, depositions, and trial testimony. 7. In deposition and at trial, listen to the questions and answer those questions only, then stop Before giving deposition or trial testimony, the expert’s attorney regularly gives this time-honored advice. Before deposition, it is usually possible to anticipate most of the questions that will be asked. The same is true at trial. What is never clear, however, is how the questions will be


posed and what new ones may come up. The expert’s job is to make sure that the opposing attorney’s question is clear enough to be understood and to request clarification if it is not. Compound questions should be broken down into separate ones. In a trial I was once asked, “Dr. Shuy, when you carried out your subjective analysis of these tapes, what kind of tape recorder did you use?” This compound question was, of course, a trick to get me to admit that I had done a subjective analysis. I managed to catch this, however, and replied, “My analysis was objective, not subjective.” Unclear questions should be responded to with words such as, “I don’t understand your question.” The expert should never assume what a vague or ambiguous question means and then go on to give an answer based on that assumption. Obviously, no answer should be made based on the expert’s partial knowledge, imperfect memory, or guesswork. This can only muddle what follows and possibly even make the expert look silly. One good strategy for you to develop is to pause briefly before beginning any answer. This gives your own attorney the chance to make an objection if one is called for. It also provides you a brief opportunity to consider the best way to respond rather than to blurt out whatever first comes to mind. If a question that the expert has never thought about before is posed, it is sometimes good to admit that this is a new perspective, pause to consider it, and then give the best possible honest answer. I have found this to happen to me, particularly when I’m asked a hypothetical question. These can be troublesome since the expert witness often does not always know or understand the direction that such a question is taking. One has to quickly think about such a direction while not getting panicked by it. Asking a couple of clarification question buys some time while the expert has time to reorient. The expert sometimes opens the door to trouble when volunteering information that is not requested in the question. This is where academics often have great difficulty and most attorneys are well enough aware of this fact to use it to their advantage. Experts tend to enjoy displaying their knowledge and sometimes confuse their roles as teachers with those of expert witnesses. They forget that it is their attorneys who are controlling the flow of information at the trial and that anything that they might want to have said can be inserted later, in crossexamination, or in the recross. You should always listen carefully to the question that is asked and answer that only and volunteer nothing else, no matter how difficult this may seem at the time.


8. Don’t panic if you forget something, give an unclear answer, or say something that you wish you hadn’t said In court cases, as in life, there is always an opportunity for redemption. If expert witnesses forget to make important points, give inaccurate information, or are hopelessly unclear, the attorney on their side will usually sense this and give them an opportunity to “rehabilitate” the exchange at a later point in their testimony. Unfortunately, there are occasions when this is not possible and the damage is done. An example of this occurred in my testimony during the ConAgra v. Hormell case (chapter 6), when I was trying to stress the point that the meaning of the word, “selection,” differed from that of the word, “choice,” in the semantic array of things that these words control. In my opinion, “choice” meant that only two things could be chosen between, while “select” indicated that there were more than two things to select from among. But later in my testimony, on another topic, I used “choice” to indicate an array of things that could be chosen, a clear stumble that did not go unnoticed by the judge. I caught it immediately and corrected myself with a laugh, but there was no way to rehabilitate this gaff. In the long run, it didn’t make much difference, but at the time my cheeks burned brightly. 9. Keep clear and readable records One of the curious things about law is that it regards itself as a written activity. Records are kept on paper, despite the fact that depositions and trials are conducted orally. What matters is the written record and the way it is kept. Court reporters hear the oral testimony and record it on paper, which then becomes the total evidence of what has happened. The expert makes a written record as a report or affidavit, but after that is expected to perform orally in depositions and trials. In the days before electronic word processing, experts prepared paper drafts of their reports that were expected to be preserved and disclosed to the other side, as evidence of the stages of their thinking and the changes that they made during the process of analyzing the data. Today, one tends to preserve no such draft copies, since it is now possible to make corrections over and over again on the original disk or hard drive of a computer. This, of course, makes it possible to leave no paper trail about whatever changes may have been made during the process, since there is no evidence to disprove it. Thus an ethical issue evolves. Since it is more honest not to obscure the evidence of earlier drafts, a better approach may be to not write them at all. Instead, experts can organize their thinking and discuss it


with their attorneys well in advance of setting anything down on the computer screen. Then, once an oral agreement is reached about what the report should contain, a draft can be written. Naturally, any writer will have second thoughts about ways to make a report better. Editing is common and necessary. Things are removed or added. It behoves the expert to keep track of such changes, however, and report them honestly and completely when asked at deposition or trial. It is unbelievable for experts to claim that they constructed their reports at a single, first sitting, making no changes thereafter. It is better to have the gist of the report well in hand before writing it, then to admit honestly and clearly, if asked, what editing or other changes were made from that point on. Another record-keeping issue relates to the notes that experts take during their analysis of the data. Sometimes opposing attorneys will ask to see such notes at deposition. Again, it is unbelievable that in a case involving a great deal of data, experts will take no notes at all. One can always say that such notes were once made and were later destroyed, thus having none available if such are requested. Even if this is the truth, it is not always a good idea. Better is to produce any notes at deposition. But when this happens, the notes had better be clear and readable both by the opposing attorney and by the person who wrote them. Again I report a problem that I once had, this time with my own notes. I took notes during a tape-playing session at the attorney’s office during the AutoNation v. CarMax case (chapter 11). Previously, I had made careful notes on the media print materials back at my home office. These were clear and readable. But at the attorney’s office I was rushed. This was the only chance I had to hear and view the radio and television advertising of both companies and the television news programs that discussed the two enterprises. My notes, scribbled in pencil, were hurried and sloppy at best. When I produced them, as requested by opposing counsel at deposition, even I was unclear about some of them, causing considerable confusion to both of us. It was not a happy time and, if I had it all to do over, I would have made my notes more legible and neat in the first place. 10. Be patient Linguists who are inexperienced in working with law cases will need to develop a sense of patience that differs from the patience commonly used in their everyday life. Time is no longer under the linguist’s control. Deadlines are set by judges and lawyers; the linguist has little or nothing to say about when things are needed or the time limits for


getting them done. A civil court case is characterized by times of frantic rushing followed by long periods in which it seems that absolutely nothing is happening. Like the army, it’s hurry up and wait. For linguists, the initial period of a trademark case is probably the most interesting. A batch of data is received and the task of organizing and analyzing it is more like the work that linguists are used to doing in the academic world. Having come to some preliminary conclusions, linguists then want to share what they have found with an audience. In their university life, they can test out their hypotheses and findings with their advanced students, they can write a journal article about them, or they can discuss them with colleagues. Confidentiality makes this impossible in a court case (some cases even require experts to sign a confidentiality agreement, which is in force during the proceedings of the case). The only viable audience to share things with is the attorney with whom they are working. One problem at this point is that of getting the attention of that attorney, who is often mired in many other aspects of the case, including various briefs, depositions, relationships with other experts and the client, and negotiations with the opposing party and the court. The attorney has a sense of timing that does not always meet the linguist’s need to share exciting findings or problems. To make matters worse, it is often the case that the linguist and the attorney are in different parts of the country. Phone discussions are possible, but it is often difficult to connect with each other. Sometimes weeks or months go by with little or no communication. Experts begin to wonder if their work is being ignored, considered unimportant, or whether the case has been settled without their having been informed. Face-to-face meetings are expensive and time-consuming, but there finally comes a time when one is scheduled. Not surprisingly, something may come up and the meeting has to be rescheduled for a later time. Meanwhile, the linguist’s excitement begins to wane and other things naturally begin to occupy his or her interest. When the face-to-face meeting finally takes place, everyone is relieved and the linguist’s excitement about the analysis is revived. But meanwhile a new kind of patience is required. When the deposition is scheduled, a new waiting game is created. Linguists worry about just how much to say, whether or not to try to avoid any troublesome parts of the data, and how well they will do when faced with the opposing attorney’s questions. It would seem as though they should be spending a lot of time with their own attorneys getting ready


for this event. Sometimes only a day, or part of a day, is spent in such preparation with the attorney. Sometimes they meet only for a couple hours the night before the deposition. Linguists may wonder whether this small amount of pre-deposition preparation is evidence of the fact that their contribution to the case is so minimal that it doesn’t merit any more attention or whether it means that the attorney is so confident in them that little preparation is actually needed. It would be nice to be told which, but the attorneys seldom discuss this issue. They have their own concerns, distractions, and plans. A new kind of patience is required. Then there is the trial itself. The linguist could be asked to be present for only that part of the trial where he or she is required to participate. Trademark cases are usually short – a week or two – and this usually means that experts should be there for most, if not all, of the trial. In civil cases, especially in bench trials, it is possible to take experts out of order when scheduling problems dictate. In a trial before a jury, such practice is less desirable, since juries are thought to be less able to piece together the overall thrust of the case. So the expert is expected to be within reach at any time during the trial. This means doing a great deal of waiting. The expert can use the time to go over proposed testimony, think up possible cross-examination questions, and review the data another time or two. But there is only so much of this that can be done before it becomes a bit stale. The expert does not want to appear to be stale on the witness stand. A new kind of patience is required. * These ten suggestions are offered to alert linguists who may have the opportunity to be involved in litigation on a trademark case but they can be equally applicable to involvement in other types of cases as well, especially civil litigation. One of the most exciting aspects of such work is that it expands one’s knowledge base, simply by the need to apply or revisit aspects of linguistics that may not have been used recently. It can also cause one to dig into newer areas that developing. Sir Francis Bacon is known for his belief that all knowledge should be the province of humankind. During the 25 years that I have been working on the intersection of linguistics and legal issues, I have been delighted to have the opportunity to learn a great deal more about the world. In only the cases described here, I’ve had the opportunity to learn things that I never knew about the industries of lodging, automobile products, the food merchandizing business, restaurants (including fast


foods), pharmacology, the toy business, and how used cars can be sold. In other civil cases, such as product liability, defamation, and contract disputes, I’ve learned other things about how the world outside of academia’s ivory tower exists. Criminal cases have exposed me to things about how large corporations work, and about the more seemy sides of life that I might otherwise have never known about. Being invoved in the practice of taking my field to other parts of life has been a truly educational experience, perhaps not exactly as Bacon had envisioned, but nontheless valuable in many ways. For this experience alone, I recommend this work to others.

16 Some Suggestions for Attorneys

The description of the cases discussed above should make it clear that there are different specializations within the field of linguistics. Fifty years ago, this was less true, but linguistic study has expanded rapidly since that time and it is now possible for you to call a local university for a linguist and be asked what kind you want. Some specialize in semantics, some in phonology, some in syntax, some in one or more of the many other areas of linguistics now offered and taught, including applied linguistics. But one thing should be true about all people who call themselves linguists, no matter what their specialization. They should be well trained in the core areas of the field, namely phonology, morphology, syntax, lexicography, semantics, pragmatics, and discourse. A linguist who specializes in psycholinguistics, historical linguistics, applied linguistics, or sociolinguistics may focus on those areas but, at the same time, should be able to effectively use the core areas named above as they carry out their work. One thing a trademark attorney may want to find out about prospective linguistic experts is whether or not they are trained and competent to deal with the issues of phonology, morphology, syntax, lexicography, semantics, pragmatics, and discourse that are likely to be relevant in a trademark case. Unless they are, you may not want to talk further. You should also understand that some academic linguists may have no interest in getting involved in legal disputes. This may result from fear of the courtroom, lack of time, a dislike of attorneys in general, or, more likely, a sincere desire not to involve themselves in anything that takes precious time away from their current academic interests or work. Even linguists who have experience in law cases may have time and interest problems. This does not mean that you will not be able to find 182

Some Suggestions for Attorneys 183

one for your case, however. One suggestion is to call linguists who have testified in earlier cases and ask them for suggestions. There is no “registry” to consult for such specializing linguists but word of mouth is often just as efficient. One last warning. Be careful about linguists who advertise their wares in journals or newspapers. Once it is decided that a linguist might be helpful, attorneys who find linguists to assist them with their trademark cases might consider the following ten suggestions, parallel to the list given to linguists in chapter 15. 1. Keep your expert linguist from becoming an advocate As noted above, it is a natural human tendency to take sides in controversies and battles. Academics are prone to do this out of their background and training. Your expert linguists may have such a tendency by nature. Help them avoid it. It is very important to have team players but the greatest value of your experts will be as arm’s-length, neutral outsiders to the very advocacy that is your own duty to have in abundance. Since linguists may confuse being a good team player with becoming an advocate, help them to make this distinction early on in your case, then critique them when they appear to cross the line. Your case is better served when they stick to their analysis and let you take it to the level of showing the jury or judge how it helps your case. Since trademark cases often involve the issue of confusion, it is best not to expect your linguist to conclude that a jury will or will not be confused. Linguists cannot make an absolute judgment about this. They can only report whether or not the language data they analyze indicates a potential for such confusion. Your survey consultants must ask the public what it claims that it understands. Don’t expect the linguist to be able to do this for you. On this same general point, avoid submitting to the court a statement about what you “expect” your linguist to testify about without first checking with that linguist to be certain that your statement is accurate and complete (see chapter 10). Your understanding of what the linguist will say may not match that of the linguist. 2. Encourage your linguistic expert to give in a little when there is no other way to do the right thing Academics are trained to stick doggedly to their positions. That is fine for the advocate but can be dangerous for the expert witness. Furthermore, as noted above, giving in a little often makes the expert appear more human in the minds of the triers of the facts. Naturally, you don’t


want your experts to give away the store or to have their testimony hurt you in vital points, but on things that matter less, advise your experts that it is not too harmful to give in when it is appropriate. To help them be able to tell the difference between the big things and the little ones, tell them know the broad outlines of your case from the beginning, make clear the resources you have to address them, and tell them which issues that matter most and the ones not worth fighting about. A good team player needs to know these things anyway. 3. Advise your experts to keep their reports as brief as possible Academics love to talk and write, especially when they are experts on a subject. You know how nice it is to cross-examine academics when this happens. Just when you thought you were not getting anywhere, the expert suddenly volunteers a lot of information that provides you with new directions to take. Linguistics experts are no different in this regard. They are unfamiliar with the culture and pace of law cases. Help them be brief and to the point. The long-standing seminary advice given to ministerial students is to say what you want to say, then stop. This is not easy for experts who really know and love their field. Many of them are professors who sincerely want to teach the jury or judge more about linguistics than your case may require. 4. Give your expert linguist enough attention Nothing is worse than feeling ignored or useless. During periods of frantic activity, your case may seem dormant or dead from your experts’ perspectives. Try to find time to give them a call to report on where things are, especially when you haven’t had contact for a while. For example, I was involved in a civil case that had enormous amounts of negotiation going on over a two-year period after I had finished most of my analysis. The attorney gave me a call every month or so to keep me generally informed and to let me know that my services would still be needed. This felt very good and made me feel connected. If possible, schedule several face-to-face meetings with your expert before the reports, depositions, and trial take place. There have been cases where I had information or serious questions which I did not get the opportunity to discuss in depth before these events took place. My issues may have seemed unimportant from the attorney’s perspective, but they left me with a less than adequate feeling when the deposition or trial events transpired. If face-to-face meetings are not feasible, keep in contact by telephone or electronically, as long as problems of discovery are not an issue.


5. Encourage your expert to take the other side’s position, for the sake of the clarity of your own argument It has long been recognized by the field of intelligence analysis that it is practical and efficient to have more than one hypothesis about a crime. Good law enforcement practice is to form multiple hypotheses, including the attempt to interpret events from the defendant’s point of view. Weak law enforcement efforts have only one hypothesis, that of the defendant’s guilt. When faced with a different defense hypothesis at trial, the single hypothesis approach often founders. The prosecutor simply hadn’t looked at the case from a different perspective or anticipated the other side’s position. This failure led to a classic oversight in the criminal case of US v. John Z. Delorean, when the prosecution did not consider that words they thought were so incriminating could be understood by the target in an entirely different manner (Shuy 1990; 1993). In trademark cases, it may be wise to try to put together, as well as one can, a way of defending the opposite side’s case. Good trademark attorneys always do this, for it is a good way to evaluate their own arguments. My point to attorneys here is to invite your linguistic experts to do the same thing. How would your linguists treat the data if they were working for the other side? The answer should be that they would come to the same conclusion no matter which side they worked on. This is a reasonable test to determine how the experts will stand up in deposition and at trial. But it is also a good test for the possible weaknesses of your own case. If there are holes in it, it is good to know this as you go into the matter, and not simply learn about it later. If your linguist finds serious problems with the linguistic aspects of your case, you may even consider seeking an out-of-court settlement before such problems surface at trial. 6. Help the linguistic expert understand that lawyers and judges are already skilled in the use and knowledge of language Linguists are used to facing what they believe to be naive opinions about language by laypersons. Throughout I have noted that linguistics is, in many ways, an invisible field. Everyone claims expertise about language, despite knowing very little about how it actually works. Since they are used to this, linguists sometimes believe that attorneys are no different. Lawyers may not use the same terminology as linguists or have training in depth that language specialists achieve, but they are far from ignorant about the practical uses of the field. Fortunately, lawyers are very good at absorbing new information from disparate fields of study.


I have found attorneys to be quick students in fields that are new to them. The sooner a linguistic expert comes to understand this, the sooner the exchange between linguist and lawyer can be more productive and efficient. 7. Teach your linguistic expert how and when to listen Many academics are, on the whole, better talkers than listeners. We are trained to put forth our ideas, our research, our opinions as forcefully as possible. This is how we convince our colleagues that we are right. We also have learned to infer the meaning of what others are saying, often answering questions that have not even been asked. However effective this may or may not be in the academic world, it can be dangerous in the courtroom. Simply listening to the question without inferring what underlies that question or what further questions might be expected to follow it is not necessarily easy to do. However much resistance academics may put up to your advice to answer only the question as it is stated, attorneys must learn to impress their experts on how crucial it is to do so. This does not come easy. 8. Make very clear to your experts that errors in their testimony can be rehabilitated at a later point “Rehabilitated” sounds like a terrible word. Perhaps a better expression might be found. Again, the academic world is a different culture from that of the courtroom culture. Academics usually feel that they are not allowed to make mistakes. When they do, they can feel humiliated, embarrassed, and outcast. It is difficult for experts to understand that errors, forgetfulness, or confused testimony can be fixed up when their own attorney gets another turn at them. Well in advance of any deposition or trial, this fact should be impressed upon your experts. It can be very comforting to know. That the expert has erred seems humanizing to many jurors, as long as the error is not crucial to the case. That line between crucial and marginal is not as clear to experts as it may be to attorneys. I was introduced to this concept in the first trial in which I testified, in 1979. It was a criminal trial in Fort Worth, Texas, and I was on the stand for three days as an expert witness for the defendant. At each break, in Texas style, the attorney announced to me the “football score,” which was actually an evaluation of how well I was doing. At the first break I was ahead 7-0. By noon it was 14-3. At the end of the second day I was ahead 24-6. After cross-examination, in which I was apparently effective, the attorney announced that I had won, 40-10. He was an experi-


enced and rather famous criminal attorney and his “score” held up when the jury came back with an acquittal for his client. But I was concerned about the 10 points scored on me by the prosecutor. The attorney shrugged it off as something that endeared me to the jury. Later jury polling showed him to be correct. They liked the fact that this hotshot expert could take a hit. In this case, no “rehabilitation” was apparently thought to be needed. 9. Help your expert know how to keep appropriate records My view of what I was required to provide to the opposing side in a trademark case included absolutely everything that I had ever scratched on paper, including summaries of telephone conversations, copies of correspondence, and notes taken during my analysis. At one time in my life, my handwriting was pretty clear but in later years it deteriorated to a strange combination of semi-decipherable abbreviations and partial words. Seldom do actual sentences appear. I write notes in pencil, sometimes impossible to read later, even by me. The view of what should be preserved and kept was never disputed by any attorney with whom I worked. When facing deposition, I was told to bring everything that I had in my file. So I did. I later came to realize that my obsession with total compliance and accuracy, though honest and above board, often created some problems for me and for my attorney. Long periods of depositions were spent in trying to decipher what I had scribbled. Cross-outs were evaluated. Guesses were made about what had been written. I also noticed from reading depositions of other experts that they often reported having taken no notes whatsoever. Most replied either that they had never taken notes at all or that their notes had been destroyed at some point. If destroying old notes was acceptable practice, then I guessed I could probably do that too. But better yet, perhaps, would be to make my notes in a more presentable fashion, one that would make them readable at a later time. Drafts of expert reports posed another interesting issue. In the days before computers I produced several drafts of the reports that I finally submitted. When asked by opposing attorney to see those drafts, I brought them along and made them available. There were the usual rewordings, reorganizings, deletions, and additions, but usually nothing that seemed interesting enough to probe deeply about. I entered the computer age a bit later than others and in 1991, during the ConAgra v. Hormell case, I heard the opposing linguist say that he had no drafts of his report since he simply corrected the single version on his word


processor. That seemed like a good idea, and I’ve done it this way ever since. The major issue for the case, I suppose, is whether or not the attorney I worked with had influenced me to change anything factual about my report. This never happened. No attorneys has ever asked me to do this and I would refuse if they did. My advice to attorneys here is to let your linguistics expert know clearly what the rules are about preserving original notes and drafts of expert reports. In my experience, the attorneys seemed to prefer to be silent or as non-committal as possible about this. Somehow I feel that I caused more trouble than necessary by taking a totally above board, letter-of-the-law approach to my notes. As a result, I tend to no longer take elaborate notes of telephone conversations with attorneys, noting only who called and the date. For people with less than perfect memories, this can pose a bit of a problem. Nor is the whole issue of discovery very clear to most non-lawyers. It would be nice to be better informed. 10. Be considerate of the expert’s need to know what is going on Although your case may be absorbing and time-consuming for you, keep in mind that your experts have little or no knowledge about the meaning of delays and other matters, leading to their impatience and feeling of being ignored or devalued. Long stretches of silence can be easily misinterpreted. And by all means, have the courtesy to let your expert know how your case came out in the end. As in a marriage, effective and frequent communication is essential, even if it is only to report that nothing is going on. Recently, an attorney called me about possibly helping him with a case involving a mountain of data to be analyzed within a two-month period. After discussing this possibility, he told me he’d get back to me with final authorization for my work in a day or two. Shortly after this, I received a call from a different attorney and I felt that I had to turn his case down because I wouldn’t have time for both cases. As it turned out, the first attorney never called me back and I was left with neither case to work on. A bit more consideration of my own work schedule would have been nice. * Trademark cases provide attorneys with an opportunity to make use of a valuable resource, linguistics. Linguistics waves no magic wand and


most certainly will not win or lose trademark cases on its own. But digging into the way marks sound, look, and mean is clearly the domain of the trained linguist. Trademark law has developed sophisticated categories of dealing with possible infringements, many of which can use the expertise of an linguist who can bring the tools of phonology, mophology, syntax, lexicography, semantics, pragmatics, and discourse analysis. Linguistics offers the technical reasons and measures to support or deny the claims of similarity or difference in product names or services. As some of the cases described here show, using a linguist early in a case may even prevent the need for you to have an expensive consumer survey carried out. The cultures of law and linguistics differ in many ways. We have not resolved the issue of who owns language. We have not resolved whether the principles of trademark law take precedent over those of social science or linguistics. This battle will probably continue. Nor can this book expose all of the ways that linguistics and trademark law intersect, but it is offered as a step in that direction, with the hope that other linguists and other attorneys will find still more and better connections in the future.

17 Power, Control, and the Ownership of Language

So what can be said about who owns language? The fields of law, linguistics, and other social sciences have their own standards, based on the knowledge, research, and principles of their respective fields. These differences grow out of quite conflicting views of where the standardization comes from and how it works. As was pointed out in chapter 1, the authority of law, attorneys, money, as opposed to the way linguists document the natural processes of language change, can come in conflict when placed together in the same arena, such as a trademark case. The practical consequences of authority and its resulting standardization are applicable to the field of trademark law, where there is an important and practical need for someone to have the authority to settle disputes clearly for once and for all. But when legal issues deal with linguistic and sociopolitical concerns involving human behavior, a second practical consequence emerges: the need to reconcile the overarching importance of law’s prescriptive language authority with the contrasting language principles of fields that define it in a quite different way – by the people who actually do the talking and acting. Such conflict is apparent in some of the cases described in this book. The task for linguists and trademark attorneys is to together negotiate the rough waters of conflict without compromising the integrity of the whole ship.

Power over language use As Conley and O’Barr (1998: 2) suggest, it is possible that the issue is not really so much one of the authorities supporting law, linguistics and other social sciences, as it is a matter of their differences in power – in this case the power of law versus the power of virtually everyone else, including linguists and social scientists. If power is defined as authority 190

Power, Control, and the Ownership of Language 191

with greater muscle than is given to others, then law is the obvious winner. Behind the authority of law is the government. It has enough muscle to impose physical penalties on those who don’t comply. It can be the bully on the block if it wants to. But is legal power so strong that it doesn’t need to pay attention to the power of language and society? At present, trademark law seems to have the power to ignore the power of other fields, but we are led to wonder if law even knows that other forms of power exist, including those that can help them meet the goals of law. Different power relations are based on underlying belief systems or ideologies. Ideologies are social representations shared by members of a group and are used to govern group behavior through internalized group norms. We normally think of ideologies as relating to political programs, philosophical orientations, or religious views. Ideologies also underlie attitudes and actions regarding race, gender, ethnicity and other current issues. But ideologies are also found in the basic, shared understandings and behavior of arenas of life such as law, linguistics, or social science. Having accepted our field’s ideologies, we see issues from that position, and are sometimes shocked to learn that others do not share them with us. As fields with different ideologies are thrown together, either by choice or by accident, a clash over ideologies and beliefs can become troublesome. When this happens, the prudent choice is to claim the process of negotiation for, after all, power is actually a set of potentials that can be variably shifted once it becomes apparent that conflict exists. Power implies distance in most societies of the world. It suggests a hierarchy in which an unequal relationship precludes closeness and solidarity. Tannen points out one of the problems in the dynamics of language power and solidarity: Communication is a double bind in the sense that anything we say to honor our similarity violates our difference, and anything we say to honor our difference violates our sameness. (1994: 29) The result of this dynamic is that similarity becomes a threat to hierarchy, whereas clearly setting off differences promotes it. It is quite natural for linguists and lawyers to celebrate their differences. It is easy for linguists to say that lawyers don’t know anything about linguistics or for lawyers to say that linguists don’t know anything about law. However overgeneralized these statements are, the simple truth is that the respective powers of law and linguistics are different, but not


unequal. It is also true that the hierarchy of a trademark case places the field of law, not linguistics, at the top of that hierarchy within the legal context, which is buttressed by the power of coercion that linguistics does not share. But in matters of analyzing language evidence, law and linguistics can be equal power partners if they really want such a relationship. The double bind that Tannen speaks of can be a distinct deterrence to the effective cooperation of trademark attorneys and their linguistic experts within that hierarchy. But not all cultures think like Westerners. In Japanese society, for example, being linked in a hierarchy tends to bring individuals closer, uniting them in a joint activity, while Americans in the same activity see themselves as individuals participating in that same event (Watanabe 1993). Other anthropological studies also show how different cultures actually view hierarchical relationships as mutually empowering (see Beeman 1986 for Iran; Yamada 1992 for Japan; Wolfwitz 1991 for Java). It would seem that we can learn from such cultures.

Recognizing the language power conflict Power is a social event, and, like communication, it has pattern and structure. According to Mead (1934: 63), there is conflict over power only to the extent that all protagonists participate actively in the conflict. So conflicts must be recognized as conflicts before anything can be done about them. The real danger of power is that it is best exercised at times when it is not the main topic of the agenda or when it is not even explicitly on the floor (Varenne 1982: 131–2). The conflict between the power of trademark law and the power of the human mind to use language creatively and flexibly has not been subjected to much public discussion or debate. Neither lawyers nor linguists have given it much consideration. In fact, many have probably never thought about it at all. When important conflicts are not on the public agenda, they can take on a mythic quality. Public myths tend to disguise important truths, as when people believe that medical science can solve all problems or that the stock market will always rise. When such conflicts are not on the public agenda, they can fester, driving participants farther and farther apart. We do not offer a final solution to this power conflict here, but we have to recognize its existence and try to put the issue on the agenda, begin talking to each other about it, and asking each other a few well-placed “why?” questions.


Power versus control One path to a possible solution comes from the realization that power is not the same thing as control. Without giving up their muscle power, the courts can easily retain control of trademark laws but ground them more firmly on the wider knowledge of the workings of language and society. Two decades ago, Tannen expressed this issue poignantly: Is there any situation in which power is equal? It is misleading, I believe, to reify power as if there is one source of it and somebody has it and somebody else doesn’t. I suggest that there are many different kinds of power and influence that are interrelated and have varied manifestations. When people are taking different roles, it may not be the case that one has power and the other doesn’t, but that they have different kinds of power, and that they are exercising it in different ways. (1982: p. 5) Tannen’s observation seems to be relevant to the conflicting powers of law, language and society. Yes, the law has great power. Yes, linguistics has its own kind of power. But does one power really have to trump the other? Are these not really two different kinds of power being exercised in different ways? Control is a different issue. Law must certainly have control of the courtroom, the methods of determining cases, and the ultimate outcomes of trials. But at the same time it would be prudent for law to learn to better appreciate and use the power of linguistics as law exercises control of these things. Most of the chapters in this book try to demonstrate how linguistic power can be used effectively in individual trademark cases. Another goal is to open the door to the possibility that such power could be used to help trademark specialists rethink some of their current approaches in using law’s major tool, language, in the ways that laws are couched and interpreted. This goal can be accomplished without attorneys having to give up any control. Until trademark lawyers come to appreciate how best to make use of the power of linguistics without surrendering their own control of their cases, ideological differences such as the ones concerning the ownership of language will remain at the center of the battle over standards. Unintentionally perhaps, law may give the impression that it is king over language and that the fields that feed it must humbly submit to law’s ownership of words and expressions (even prefixes, as it turns out).


And the unfortunate product of such an impression will continue to have an impact on linguists who are asked to help with trademark cases. They are forced to tread a slippery slope. They must be true to their ideology that the people who speak a language are the real owners of it while, at the same time, they must operate as team players with the attorneys whose ideology is that law owns language. Of course, as the cases in this book relate, linguists can continue to call on their expertise to analyze the words and expressions in specific trademark cases and to describe the usage that they find in dictionaries, electronic searches, depositions, and any other language data in evidence. At the same time they are forced to suspend their disbelief in such matters as the right of one side to own words and expressions, the paradox of “secondary meaning,” and the elusive definitions of key concepts in trademark law. And they also must live with the outcomes of the cases, whether or not they believe them to conflict with their own field’s value of the power of people to own words and expressions. So far, the burden has been all on the side of the linguists. Some linguists find this burden difficult, if not impossible. They believe that trademark law ideology is so ill-informed and so counter to what linguists know to be true about the ownership of language that there is no point for them to even get involved. Other linguists are more optimistic, trying to add their linguistic knowledge to at least some parts of the individual disputes. Still other linguists may not even be aware of these problems at all. Attorneys who call on linguists to help them with their trademark cases might do well to discover their experts’ feelings about the battle over who can own language at the very beginning of their association. Linguists who haven’t even considered the issue may not prove to be very helpful anyway, for if they haven’t even thought about the problem, they may not be very good linguists. If they believe that the power of law supersedes any other kind of power, including that of their own field, they may also fail to be convincing in the long run. Some linguists who are drastically opposed to the view that law owns language may not even want to help attorneys with their cases. Those linguists who have come to the conclusion that the law cannot really own language but, at the same time, are willing to do what they can as a team player, without prostituting their underlying differences about the ownership of language, may prove to be the most helpful of all to trademark attorneys. But this issue should be addressed early on in the association. As in all battles, allies often do not share the same frames of refer-


ence, knowledge, beliefs, or ideas about authority, power and control. Trademarks are largely about language, making it useful, if not necessary for lawyers to move in the direction of linguistics for help. Linguists are not as knowledgeable about law as they need to be, making it necessary to move in the direction of lawyers. Such efforts have great possibilities for mutual empowerment and successful trial outcomes. Attorneys stand a better chance of working together with linguists rather than trying to go it alone in battles where language is the central issue.

Linguists and lawyers working together If trademark lawyers were to consider how to conjoin the power of linguistics to the power of law, the things they might do together, as a team, include the following: 1. Law everywhere agrees that a trademark must perform the function of identifying and distinguishing. Although a mark must be distinctive, definitions of “distinctive” are seldom made explicit, if they are given at all. There have been efforts of many international trademark law conventions to attempt this, but we are still a long way from clarity (Ladas 1975: 974). Defining in such a way that distinctions can be made is a task for which linguists are professionally trained. In the US, the Lanham Act, borrowing from British law, not only speaks of distinguishing the applicant’s goods and services, but also refers to marks that are “capable of” doing so. Language capability is clearly an issue to be determined by linguistic analysis, which is, of course, what linguists do. 2. Trademark law makes much of concepts such as “invented words” and “newly or freshly coined terms,” yet does not give procedures for determining or defining inventiveness, newness, or freshness. Case records are replete with inconsistent decisions, as law scholars like Ladas and McCarthy freely admit. For example, a drink derived from whisky and liqueur called “Whikeur” was judged to be invented while a similar product, derived from the words, “Scotch” and “light,” called “Scotchlite,” was judged to be not invented. As things now stand, if the Court determines that a word is invented, it is therefore invented, regardless of whatever insights linguistics might bring to bear on the issue. Linguists are specialists in making such determinations. 3. Trademark law focuses on the meaning that is conveyed to the “ordinary person,” yet it is not often clear exactly who the ordinary person is. Sociolinguists and other social scientists have ways of getting


closer to a meaningful definitions than any that appear to be evident in trademark law or trial practice. 4. Trademark law and linguistics together can surely figure out a way to make some sort of compromise about the prohibiting dictionaries from documenting accurate and scientifically discovered instances of the public’s generic use of commercial trademarks, especially by competent lexicographical practice. 5. Perhaps the power of money could be reconsidered in pursuing decisions based on “secondary meaning,” which is not really “meaning” at all but, rather, an economic condition that is used to interpret other economic conditions. It is a stretch to consider “secondary meaning” as meaning in a linguistic sense. As the term currently applies, the law appears to be using “secondary meaning” as a euphemism for “previous financial investment.” If it’s money that matters here, why not openly say so? 6. More linguistically enlightened procedures could be followed in helping untangle the muddled doctrine of “foreign language equivalents,” which even McCarthy complains about. The increased internationalization of trademarks will face this issue more and more. 7. Linguists could be called upon to help better define the “imagination test.” If addressed by linguists, the hierarchical series of inferences that need to be made in order to determine how far a word must be taken so that it can fit into the category of suggestiveness, could add some clarity to this admittedly subjective measure. 8. As Ladas (1975: 1005) and other legal scholars point out, the distinction between generic and descriptive terms is not very precise. It is said to be “more a difference in degree.” One currently used requirement is that the “descriptive character must strike the purchaser as a matter of first impression without mental operation.” Etymology doesn’t count since consumers cannot be expected to dissect words. Linguists can be called upon to develop a more explicit continuum that will help explain this “difference in degree” as well as the stages in this till now undefined “mental operation.” 9. Trademark law’s concept of generic words is somewhat inconsistent as well. To be considered generic, it is sufficient that the principal significance of the word is an indication of the nature or class of an article rather than an indication of its origin (Ladas, 1975: 1165). Yet in 1955, for example, a Court found in the case of Marks v. Polaroid that the origin of Polaroid’s name is protected, even though the public is required to understand that name as a generic term, a class of the


product. The search for the genericness of a word is clearly a linguistic issue. 10. In the US, the issue of infringement, according to the Lanham Act, section 2(d) and 32(1), centers on “whether there is a similarity that is likely to cause confusion or mistake or to deceive purchasers.” The general impression of the purchasers is the standard. But how is this general impression to be discovered? In the US, attorneys are to look at the marks “as a whole” since the ordinary purchaser won’t dissect them. UK judges have ruled that the marks should be examined “side by side.” Swiss judges have ruled that they should be looked at separately. German judges have determined two different dangers of confusion: narrow (literal) and broad (non-literal), suggesting a direct danger of confusion (the marks look the same and have the same origin), and indirect danger of confusion (the marks do not look the same, but have the same origin). Linguists examine the language itself, not the consumers’ impressions, for what possibilities of confusion the language itself permits or encourages. Consumer surveys of what the public understands are sometimes useful but such findings can tell us only what the consumers say they understand. One cannot get into the minds of consumers to find out whether or not what they say is actually the case. At any rate, this can be considered “self-report” information, a type of data that is not highly regarded in scientific research. If linguistics can tell the courts what the language is capable of meaning, we may still not be into the minds of specific consumers, but we are every bit as close as the self-report evidence found by consumer surveys. Linguists also deal with directness and indirectness in language, particularly salient in the case of the German definitions of direct and indirect dangers of confusion, but also very important in analyzing the overall meaning of trademarks. 11. Since trademark law concerns the ethics of honesty and the need to prevent deceit, it is prudent for lawyers to get whatever help they can in such matters. Spotting deception is not a simple task, as those who have spent their careers on this topic have discovered. Paul Ekman, one of the top researchers on deception, has concluded that the human ability to identify spoken deception is woefully weak (Ekman 1985). In fact, one could flip a coin and do almost as well. From their extensive research on spoken deception, Miller and Stiff (1993) conclude that the clues that observers claim to rely on to make judgments about spoken deception are, on the whole, not really related to actual honesty or deceit at all. A comprehensive review of research on deception by


Robinson (1996) agrees with the findings of Ekman, and Miller and Stiff, that social psychologist’s research on detecting deception shows it to be at about the level of chance. Can linguistic analysis of deception help lawyers determine deceit in trademark cases? Deceptiveness in written language has not been well studied. Since trademark cases provide such data, it could prove useful for linguists to be encouraged to address this area.

So what really matters here? It should be apparent that linguistics has a great deal to say to law, especially in trademark cases. It should be equally apparent that trademark lawyers can make effective use of linguistic expertise in their cases. Important differences in beliefs about who can own language may or may not be resolvable, but two thing are fairly clear. First, linguists and lawyers can use their knowledge to work together not only to resolve trademark disputes, but also to create satisfactory language that accomplishes the goals of law without sacrificing the goals of human identity and behavior. Secondly, the general public cares a great deal about language, a fact that should be clearly recognized by the field of law. People may not articulate their concerns very well, but there is evidence that language issues play a very important role in their lives. In fact, some nations have even been willing to fight bitter wars over it. Language is a vital part of human identity, one worth doing battle over. Less dramatic evidence than pitched warfare over language, but nevertheless an important indicator of the human concern about words, can be discovered in recent technological developments such as those used by Merriam Webster, Incorporated. This publisher developed a web-site for its dictionary that attracts users to the tune of 1 million page-views a day. Site logs provide a record the general public’s electronic inquiries. For example, commonly asked questions are about when to use “affect,” “principal,” and “ensure” rather than “effect,” “principle,” and “insure.” Recent daily news broadcasts lead users to ask questions about the spelling and meaning of words like “millennium,” “recuse,” and “quid pro quo.” They even ask about newly coined words, such as President George W. Bush’s amusing use of “subliminable.” The most frequent questions asked are about words such as “paradigm,” “ubiquitous,” “hubris,” “chad,” and “oxymoron.” The general public cares about language enough to regularly use lexicographical products, such as dictionaries and usage guides. They clearly recognize language as important enough to spend their time and effort to get it right.


When people care about language, they also care a lot about being told how they can or cannot use it. We may not yet be seeing much resentment from them about a corporation legally prohibiting them from using a word, a prefix, or an expression, but this may well happen soon. They may not even know yet that the dictionaries they use every day are restrained from reporting actual generic uses of terms that are probibited by trademark battles. They may not be aware that the legal doctrine of “secondary meaning” allows the company who spends the most money on advertising and promotion to have sole rights to a trademark. They may be ignorant of the fact that the legal establishment’s power overwhelms the power of other fields. If they knew these things, they might be more likely to harbor deep concerns and resentments. Just as the previously assumed power of the medical profession is now coming under more scrutiny, so the field of trademark law would be prudent to prepare itself for eventual public awareness of these things, for change comes quickly and sleeping giants are easily awakened. Conley and O’Barr (1998: 9) express it this way: “Power may exclude, but those who are excluded remain on the scene, ready to turn local level episodes of oppression into moments of resistance.” It is a truism to say that we live in a rapidly changing world. Linguists have to deal with amazingly fast changes in language itself, in attitudes about language, in theories about language, and in the relationships between institutions and language. Law tends to prefer an orderliness with little or no change. But when law intersects with language, it must face the facts of constant fluctuation and change. That language changes very rapidly can be seen in dictionaries themselves. In 1991 we saw the first citation of “world wide web.” Within five years entries such as “browser,” “netiquette,” and “dot com” could be found. Ten years later dictionaries offered new senses for existing words such as “surf,” “web,” “bookmark,” and “clipboard.” New words are entering dictionaries today faster than ever before. Trademark law appears to want to freeze the natural and healthy phenomenon of language change, instead prescribing fixed meanings that allow for no variability. This may make sense to the field of law, but it runs counter to human creative and intellectual behavior. Thus the battle between the ideologies of law and linguistics goes on. Trademark law is largely about language planning based on commercial advantage and gain. Linguistics is about describing what the people do naturally with language. But does this mean that the two fields are hopelessly locked in an unwinnable battle? I think not. In addition to the current examples that linguists contribute in individual trademark


cases, there is much that can be accomplished if law were also to call on linguists to assist them in making clear and unambiguous some aspects of the legal concepts and procedures used in trademark cases. In short, the power of linguistics and power of law need not be in competition. Working together in harmony could be more empowering for law as well as linguistics. And in the long run, it is the rest of the people who will benefit most.

References American Institutes of Research, 1981. Guidelines for Document Designers. Washington, DC. Aronoff, Mark and Janie Rees-Miller (eds.), 2001. Handbook of Linguistics. Cambridge: Blackwell. Baldwin, John and Peter French, 1990. Forensic Phonetics. London: Pinter. Beeman, William O., 1986. Language, Status, and Power in Iran. Bloomington: Indiana University Press. Berk-Seligson, Susan, 1990. The Bilingual Courtroom. Chicago: University of Chicago Press. Black’s Law Dictionary, 1999 (7th edition) (ed.-in-chief Bryan A. Garner). St. Paul: West Group. Blackett, Tom, 1998. Trademarks. London: Macmillan Press. Bright, William (ed.), 1992. The International Encyclopedia of Linguistics. New York: Oxford University Press. Chomsky, Noam, 1966. Linguistic theory. In Northeast Conference on the Teaching of Foreign Languages, working committee reports, 43–9. Reprinted in Lester, Mark (ed.), 1970. Readings in Applied Transformational Grammar. New York: Holt. Chomsky, Noam and Morris Halle, 1964. The Sound Pattern of English. New York: Harper & Row. Clyne, Michael, 1997. Multilingualism. In F. Coulmas (ed.) The Handbook of Sociolinguistics. Oxford: Blackwell, 301–14. Conley, John and William M. O’Barr, 1998. Just Words: Law, Language and Power. Chicago: University of Chicago Press. Crystal, David, 1987. The Cambridge Encyclopedia of Language. Cambridge: Cambridge University Press. Cushing, Steven, 1994. Fatal Words. Chicago: University of Chicago Press. Daoust, Denise, 1997. Language planning and language reform. In F. Coulmas (ed.) The Handbook of Sociolinguistics. Oxford: Blackwell. 436–52. Ekman, Paul, 1985. Telling Lies: Clues to Deceit in the Marketplace, Politics, and Marriage. New York: Norton. Felker, Daniel B., F. Pickering, V. Charrow, V.M. Holland, and J. Redish, 1981. Guidelines for Document Designers. Washington, DC: American Institutes for Research. Ferguson, Charles A., 1959. Diglossia. Word, 15, 325–40. Foster, J. and P. Coles, 1977. An experimental study of typographical cueing in printed text. Ergonomics 20, 57–66. Garvin, Paul, 1973. Some comments on language planning. In J. Rubin and R. Shuy (eds), Language Planning: Current Issues and Research. Washington, DC: Georgetown University Press, 24–33. Gibbons, John, 1994. Language and the Law. London: Longman. Gilson, Jerome and Anne Gilson Lalonde, 1999. Trademark Protection and Practice. Cumulative Supplement. Vols. 1 and 3. New York: Matthew Bender. 201

202 Linguistic Battles in Trademark Disputes Grice, H.P., 1975. Logic and conversation. In P. Cole and J. Morgan (eds.), Syntax and Semantics (Vol. 3, 41–85). San Diego: Academic Press. Heath, Shirley B., 1992. Why no official language? In J. R. Crawford (ed.) Language Loyalties: A Sourcebook on the Official English Controversy. Chicago: University of Chicago Press, 220–30. Hollien, Harry, 1990. The Acoustics of Crime. New York: Plenum. Labov, William, 1970. The Study of Non-Standard English. Champaign, IL: National Council of Teachers of English. Ladas, Stephen P., 1975. Patents, Trademarks, and Related Rights: National and International Protection, Vol. II. Cambridge, MA: Harvard University Press. Landau, Sidney I., 1989. Dictionaries: The Art and Craft of Lexicography. Cambridge: Cambridge University Press. Lentine, Genine and Roger W. Shuy, 1990. McMeaning in the marketplace. American Speech 65:4, 349–66. Levi, Judith N. and Anne G. Walker (eds.), 1990. Language in the Judicial Process. New York: Plenum. Matoesian, Gregory M., 1993. Reproducing Rape. Chicago: University of Chicago Press. McCarthy, Thomas, 1984. Trademarks and Unfair Competition. San Francisco: Bancroft-Whitney. McCarthy, Thomas, 1997. McCarthy on Trademarks and Unfair Competition. 4th Edition. Vols. 3 and 4. West Group. McMenamin, Gerald R., 1993. Forensic Stylistics. Amsterdam: Elsevier. McMillan, James B., 1949. Five college dictionaries. College English 10:4, 214–21. Mead, G.H., 1934. Mind, Self and Society. Chicago: University of Chicago Press. Mellinkoff, David, 1963. The Language of the Law. Boston: Little, Brown and Company. Miller, Gerald R. and James B. Stiff, 1993. Deceptive Communication. Newbury Park CA: Sage. Milroy, James and Leslie Milroy, 1999. Authority in Language. London: Routledge. Newmeyer, Frederick J., 1983. Grammatical Theory. Chicago: University of Chicago Press. Okawara, Mami Hiraike, 1999. A trademark dispute in the Japanese courts – the Snack Chanel case. Unpublished paper presented at the meeting of the International Association of Forensic Linguistics, Birmingham. Philips, Susan U., 1998. Ideology in the Language of Judges. New York: Oxford University Press. Pinker, Steven, 1997. How the Mind Works. New York: Norton. Poulton, E.C. and C.H. Brown, 1968. Rate of comprehension of existing teleprinter output and possible alternatives. Journal of Applied Psychology 52: 16–21. Rieber, Robert W. and William A. Stewart (eds.) 1990. The Language Scientist as Expert in the Legal Setting. Annals of the New York Academy of Sciences, Vol. 606. Robinson, W. Peter, 1996. Deceit, Delusion, and Detection. Thousand Oaks, CA: Sage. Shuy, Roger W., 1990. Tape recorded conversation. In Andrews, Paul and Marilyn B. Peterson (eds.) Criminal Intelligence Analysis. Loomis, CA, Palmer Enterprises. Shuy, Roger W., 1993 (reprinted 1996). Language Crimes. Oxford: Blackwell. Shuy, Roger W., 1996. Bureaucratic Language in Government and Business. Washington, DC: Georgetown University Press.

References 203 Shuy, Roger W., 1998. The Language of Confession, Interrogation, and Deception. Thousand Oaks, CA: Sage. Shuy, Roger W., 2001. Forensic linguistics. In Mark Aranoff and Julie Rees-Miller (eds.). The Handbook of Linguistics. Oxford: Blackwell. Solan, Lawrence, 1993. The Language of Judges. Chicago: University of Chicago Press. Stygall, Gail, 1994. Trial Language: Differential Discourse Processing and Discursive Formation. Amsterdam: John Benjamins. Tannen, Deborah, 1982. Remarks on discourse and power. In Leah Kedar (ed.) Language and Power. Norwood, NJ: Ablex. Tannen, Deborah, 1994. Gender and Discourse. New York: Oxford University Press. Tiersma, Peter, 1999. Legal Language. Chicago: University of Chicago Press. Tinker, M.A., 1969. Legibility in Print. Ames, IA: University of Iowa Press. Varenne, Hervé, 1982. Analytic ambiguities in family communication. In Leah Kedar (ed.) Power through Discourse. Norwood, NJ: Ablex. Watanabe, Suwako, 1993. Cultural differences in framing: American and Japanese group discussions. In Framing in Discourse (ed. Deborah Tannen). Oxford: Oxford University Press. Wolfwitz, Clare, 1991. Language Style and Social Space: Stylistic Choice in Suriname Javanese. Urbana, IL: University of Illinois Press. Yamada, Haru, 1992. American and Japanese Business Discourse: A Comparison of Interactional Styles. Norwood, NJ: Ablex.

Cases Discussed AMR Pharm Holland, B.V. and Applied Medical Research, Ltd. v. American Home Products Corporation US District Court for the District of Columbia Civil Action 91-1830 TFH AutoNation, Inc. v. Acme Commercial Corporation et al. US District Court for the Southern District of Florida Ft. Lauderdale Division Civil Action No. 96-6141 Counterclaim Circuit City Stores, Inc., Circuit City Stores West Coast Inc. and Acme Commercial Corporation v. AutoNation, Inc. ConAgra, Inc. v. Geo. A. Hormell & Company US District Court for the District of Nebraska Civil Action No. 8: CV 91-119 Pyewacket Enterprises, Inc. v. Mattel, Inc. US District Court for the Western District of Washington at Seattle Civil Action No. C94-434D McDonald’s Corporation v. Quality Inns International, Inc. US District Court for the District of Maryland Civil Action No. PN-87-2606 Registry Hotel Corporation, d/b/a Hospitality Management Corporation v. American Hospitality Management Company US District Court District of Minnesota Third Division Civil Action No. 3-92-852 Warren Distribution, Inc. v. Prestone Products Corporation US District Court for the District of Nebraska Civil Action No. 8: CV95-106

204

Index abbreviation in marks, 34 Adams, President John, 6 Alex, John, 69, 175 Alixandre Furs, Inc. v. Alexandros Furs, Ltd., 144–9, 165, 167, 172 American Heritage Dictionary, 13, 21, 55, 85, 113, 119 American Society for the Testing of Materials, 56 AMR Pharm Holland v. American Home Products Corp., 15, 110–15, 163– 4 ancillary mark, 128 applied linguistics, 9, 25, 83, 112, 169, 182 arbitrary or fanciful marks, 14, 32, 37, 39–42, 47, 51, 65–6 Aronoff, Mark and Janie Rees-Miller, 8, 9, 18 Armani, Georgio, 2, 30 asymmetrical speech event, 126 Aunt Jemima Doctrine, 96, 97 authority, x, 1–2, 4, 5, 6, 7, 55, 190 of dictionaries, 12–13, 21 of law, x, 2, 4, 5–7, 12, 55 of linguistics, 7–10, 12–13, 17, 55 of money, 13–14 of trial attorneys, 11–12 AutoNation, Inc. v. CarMax, 15, 125–43, 164–5, 167–70, 172–3, 178 Baldwin, John and Peter French, 8 Berk-Seligson, Susan, 7 Beeman,William O., 192 Blackett, Tom, 29, 44 Black’s Law Dictionary, 33, 96, 151 Bright, William, 49 British Trade Marks Act of 1938, 29 British Trade Marks Act of 1994, 29, 44 Butters, Ronald, 129, 133–6, 138–40, 142–3, 167, 172

capitalized marks, 54, 87, 131, 137, 153, 157 Chomsky, Noam, and Morris Halle, 11, 107–8 Clyne, Michael, 4 cognitive linguistics, 10, 71, 107, 172 Cohen, Michael, 110 Colombo, Louis, 128, 140, 152, 158, 175 Community Trade Mark, 29 competitor’s need test, 37 competitor’s use test, 37–8 comprehensibility, 153, 155 computerized data base search, 47–53, 63–4, 74–5, 84–5, 98, 101, 129–31, 139–41, 160, 162 ConAgra, Inc. v. Geo. A. Hormell & Co., 69–80, 115, 161, 168, 172, 177, 187 confidentiality, 53, 94, 155, 158 confusion of marks, 15, 28, 30, 40, 41, 43–4, 56, 67–70, 80–2, 96, 111, 116, 118–19, 127–8, 141, 145, 183, 197 Conley, John and William O’Barr, 7, 190, 199 consumer surveys, 15–16, 26, 44, 53, 67, 80, 102–3, 197 control, 6, 9, 13, 15, 23, 130, 133, 193–4 Compact Edition of the Oxford English Dictionary, 57 context, 101, 104–5, 115, 139–41, 147–8, 152 core meaning, 105 Crystal, David, 22 culture of the courtroom, 7, 15, 189 customer perception, 43 Daoust, Denise, 3 deception, 30, 36, 67, 69, 82, 111, 197–8 205

206 Index deceptively misdescriptive marks, 31, 36 degrees of confidence in testimony, 123 descriptive marks, 15, 30, 31, 32–6, 40–1, 47, 53, 56–7, 61–3, 65–6, 70, 80, 82, 117–18 dialect variation in marks, 72, 76 Dictionary of American English, 85 Dictionary of Regional English, 85 dictionary test, 37 disclaimers, 150–5 discourse analysis, 17, 141 distinctive feature analysis, 11, 75, 112 distinctive marks, 31, 42–3, 111, 195 doctrine of foreign equivalents, 15, 34–5, 144, 148–9, 196 document design, 77, 146 Doty, Judge J., 81 Ekman, Paul, 197–8 Felker, Daniel et al., 156 Ferguson, Charles, 4 forensic linguistics, 8–10 Foster, J. and P. Coles, 156 Garvin, Paul, 3 generic marks, 15, 32, 40, 46–7, 49–54, 57–8, 62–3, 65–6, 81–2, 99, 109, 129, 132, 196, 199 Gibbons, John, 8 Gilson, Jerome and Anne Gilson Lalonde, 26 graphemics, 42, 77 Grice, H.P., 87, 156–8 Gruber, Thomas, 130 Heath, Shirley B., 6 Hebrew Academy, 3 Hefter, Lawrence, 98, 104, 175 “high” variety of language, 4–6 Hollien, Harry, 8 Huizenga, Wayne, 127 hyperbole in marks, 60 hyphenated linguistics, 8–10 ideology, 6, 191–5, 199 imagination test, 15, 37–9, 60–1, 65, 70, 80, 102, 196

information overload, 155 information underload, 155 internet, 2, 4, 30 intuition, 38–9 Jacobs, Shelly, 82, 86 Knoth, Richard, 142, 175 Kopf, Judge Richard G., 79 Labov, William, 6 Ladas, Stephen, 29, 109, 195, 196 Landau, Sidney, 13, 21, 50 language academies, 6 language loyalty, 4 language planning, x, 2–5, 53, 55, 199 Lanham Act of 1946, 31–3, 195, 197 laudatory marks, 34, 36, 79–80 Lentine, Genine and R. Shuy, 99 Levi, Judith and Anne G. Walker, 8 lexical change amelioration, 60 combining, 50–1, 59 contracting, 51 flexibility, 50–1 gaps, 92–3 generalization, 60 pejoration, 60 recency, 51–2, 90, 94 shift, 100 specialization, 60, 100 lexicography, 17, 20–1, 49–52, 54–5, 133, 159–60, 162, 164 linguistic tools, 9, 17, 42, 83, 115, 159, 182 Little, Bruce, 47, 53, 54, 82, 83, 84, 175 “low” variety of language, 4–6 Macquarie Dictionary, 199–20 Mani, A.R., 2, 30 marked forms, 88, 90–1, 93 Mc-, 2, 19, 33, 96–109, 139–40 McCarthy, J. Thomas, x, 31–2, 34, 35–7, 39, 40, 42, 70, 95, 144, 148, 150, 195–6 McDonald’s v. Quality Inns, 2, 14, 33, 95–109, 163, 169, 172

Index 207 McKee, Toni, 116 McLeay, Bartholomew, 57, 64, 68, 175 McMenamin, Gerald, 8 McMillan, James, 20 Matoesian, Gregory, 7 Matrix Essentials, Inc. v. F & M Distributors, Inc., 150–8, 165–6, 168, 172 maxims of effective communication, 153–4 Mead, G.H., 192 Mellinkoff, David, 8 Merriam Webster’s Collegiate Dictionary, 13, 55, 85, 198 metaphors in marks, 36, 67, 102 Miller, Gerald and James Stiff, 197–8 Milroy, James and Leslie Milroy, 6 minimal pairs in marks, 118, 122, 133–4, 164 misdescriptive marks, 36 misspelled marks, 34 Moran, Jim, 127 morphology, 17, 19, 42, 72, 76, 100, 119, 132, 134–5, 137, 159–62, 164–5 negation in marks, 153–4 Newmeyer, Frederick, 106 nicknames, 72, 76 note taking, 105, 177–8, 187–8 Okawara, Mami Haraiki, 28 ordinary person, 195–7 ownership of language, 1–3, 109, 190–4 Oxford English Dictionary, 21, 49, 52, 58–9, 85, 90 passing off, 44 patronyms in marks, 95, 100, 103 Phillips, Susan, 8 phonetics and phonology, 17–19, 42, 72, 75–6, 80, 111–14, 118–19, 121–2, 129, 133, 134–7, 139, 145–7, 161, 163–5 Pinker, Stephen, 38 Polton, E.C. and C.H. Brown, 156 power, 4, 190–5, 199, 200

pragmatics, 17, 23–4, 104–5, 134–5, 165–6 prefixes and suffixes in marks, 34, 95, 97 prescriptiveness, 2, 6, 190 punctuation in marks, 87, 92, 114, 131–2, 153, 157 Pyewacket Enterprises, Inc. v. Mattel, Inc., 116–24, 164 Random House Dictionary of the English Language, 58, 85 readability, 153 recency of marks, 50, 63, 90, 92, 94 recipient design, 153 referential meaning, 58 Registry Hotel Corp. v. Hospitality Management Co., 46–54, 159–60, 170 rehabilitation of testimony, 177, 186–7 Rieber, Robert and William Stewart, 8 Robinson, W. Peter, 197 Roget’s Thesaurus, 93 roles of attorneys and linguists, 14–15 Sadock, Jerrold, 75 Schill, Gilbert, 129, 136, 138–9, 175 sealed settlements, 94 secondary meaning, 14, 33, 36, 40–1, 43, 47, 53, 57, 68, 82, 103, 110, 117, 194, 196, 199 selective dictionary definitions, 58 self-report data, 107, 197 semantic change, 60 semantics, 17, 23–4, 59, 61, 73–4, 77–80, 86–7, 99–100, 115, 119, 122, 135, 137, 139, 146–8, 160–5 semiotics, 23, 73, 134–6 Shuy, Roger, 7, 8, 22, 185 slips of the tongue, 140 slogans, 125–9, 132, 135–8, 140–1 Solan, Lawrence, 8 specificity, 152, 154–5, 157 speech community, 4 standardization, 2, 6 strong and weak marks, 15, 39–40, 42–3 Stygall, Gail, 7

208 Index suggestive marks, 15, 32, 37–40, 47, 61, 65–6, 70 Swift, Jonathan, 6 synonyms, 38, 77, 93, 120–1 syntax, 17, 21–2, 42, 72–3, 87–90, 162–3, 166 Tannen, Deborah, 191, 192–3 theoretical linguistics, 10, 104–8 Tiersma, Peter, 7, 8, 42 Tinker, M.A., 156 Thorndike-Barnhart Dictionary, 49 trademark law in Ethiopia, 29 in France, 29, 125 in Germany, 29, 197 in Hungary, 30 in Japan, 28 in Jordon, 30 in Kuwait, 30 in the Netherlands, 32 in Scandinavian countries, 150 in Spain, 29 in Turkey, 30 in the UK, 28, 29, 30, 44, 150, 195 in the US, 28, 29–30, 41–2, 44, 150, 197

in Yemen, 30 Trademark Registration Act of 1905, 3 UK Trade Marks Act of 1994, 44 unfair business competition, 44–45, 145 unpredictability, 152 US v. John Z. DeLorean, 185 Varenne, Hervé, 192 Warren Distribution v. Prestone Products Corp., 56–68, 160–1, 169–70 Watanabe, Suwako, 192 Webster’s Collegiate Dictionary, 49, 91, 119 Webster’s New World Dictionary, 91 Webster’s Second New International Dictionary, 20, 85 Webster’s Third New International Dictionary, 20, 59, 85, 146 Wolfwitz, Clare, 192 Woodroast Systems, Inc. v. Restaurants Unlimited, Inc., 81–94, 162–3, 167 Yamada, Haru, 192 yes/no questions, 7

Linguistic Battles in Trademark Disputes

Battles

The Terror's Trademark

The trademark registration kit

Linguistic Categorization: Prototypes in Linguistic Theory

Studies In Linguistic Motivation (Cognitive Linguistic Research)

Studies In Linguistic Motivation (Cognitive Linguistic Research)

Causation and Delay in Construction Disputes

Mediation of Construction Disputes

Body’s Battles

Applicable Law in International Investment Disputes

Linguistic Method in Philosophy

Dependency in Linguistic Description

Essays in linguistic ontology

The Expert Witness in Construction Disputes

Trademark Surveys: A Litigator's Guide

Ancient Battles - (Rulebook)

Four Battles For Kharkov

Chariot Wars: Ancient Battles

Us Tank Battles In France 1944-45

Are There Unresolvable Moral Disputes?

Adjective Classes: A Cross-Linguistic Typology (Explorations in Linguistic Typology)

The Seventy Great Battles in History

The Painter of Battles

Between The Battles

Shieldwall (Warhammer Ancient Battles)

Battles of the Clans

The Painter of Battles

Episodes Of Chess Battles

Landmarks In Linguistic Thought III: The Arabic Linguistic Tradition (History of Linguistic Thought)

Linguistic Battles in Trademark Disputes

Battles

The Terror's Trademark

The trademark registration kit

Linguistic Categorization: Prototypes in Linguistic Theory

Studies In Linguistic Motivation (Cognitive Linguistic Research)

Studies In Linguistic Motivation (Cognitive Linguistic Research)

Causation and Delay in Construction Disputes

Mediation of Construction Disputes

Body’s Battles

Applicable Law in International Investment Disputes

Linguistic Method in Philosophy

Dependency in Linguistic Description

Essays in linguistic ontology

The Expert Witness in Construction Disputes

Trademark Surveys: A Litigator's Guide

Ancient Battles - (Rulebook)

Four Battles For Kharkov

Chariot Wars: Ancient Battles

Us Tank Battles In France 1944-45

Are There Unresolvable Moral Disputes?

Adjective Classes: A Cross-Linguistic Typology (Explorations in Linguistic Typology)

The Seventy Great Battles in History

The Painter of Battles

Between The Battles

Shieldwall (Warhammer Ancient Battles)

Battles of the Clans

The Painter of Battles

Episodes Of Chess Battles

Landmarks In Linguistic Thought III: The Arabic Linguistic Tradition (History of Linguistic Thought)

Recommend Documents