Author:Lemley, Mark A.
Position:Separation between validity, infringement and defenses of intellectual property rulings - II. The Problems with Separation C. IP Doctrines and the Nose of Wax 2. Patent through Conclusion, with footnotes, p. 2240-2285
  1. Patent

    As we will see in the next Part, utility patent law does more than its sister doctrines to evaluate the scope of a patent right in an integrated fashion. The patent claims themselves are an effort to define the scope of the legal right ex ante. And the pretrial Markman hearing requires the parties to resolve the scope of the patent right before either validity or infringement is established. (163)

    Nonetheless, patent law too has gaps that result from its separation of validity and infringement. The first and most obvious of these gaps involves defendants who practice a technology that is already in the prior art. Because patent law requires that an invention be novel and nonobvious, (164) it should be clear that a patent owner is not entitled to sue someone for using technology that existed before she ever "invented" it. Nonetheless, the Federal Circuit has repeatedly held that there is no "practicing the prior art" defense in patent law. (165) As Timothy Lau has ably explained, this rather bizarre rule results from the separation of patent law doctrines into validity and infringement. (166) Because the patent owner bears the burden of proving infringement but the defendant must prove invalidity by clear and convincing evidence, courts worry that defendants can skimp on their burden of proof that the defendant's technology was in fact in the prior art. (167) But the fact that the courts will not recognize practicing the prior art as a defense means that defendants who are in fact engaged in what should be permitted activity do not always have a fair opportunity to prove it. In theory, the Markman hearing solves this problem by treating validity and infringement together in an integrated proceeding. (168) But the Federal Circuit has discouraged courts from focusing much attention on validity during Markman. (169) A district court may construe the patent claims for infringement purposes without focusing on the possibility that a claim construed so broadly might be invalid. And when the issue of validity does arise, district courts and juries may be reluctant (even apart from the high standard of proof) to hold the patent altogether invalid.

    Patent owners can and do exploit this gap with some regularity. Relying both on the higher standard of proof for invalidity than non-infringement and the demonstrated reluctance of juries to invalidate patents, patent owners are often willing to trade a greater risk of eventual invalidity to gain a broad scope for their possibly invalid patent claims. (170) Sometimes this gamble fails. In Liebel-Flarsheim Co. v. Medrad, Inc., for instance, the patentee succeeded in persuading the court to construe its claim to be much broader than the thing it had actually invented, (171) only to have the claim ultimately invalidated as insufficiently supported by the patent disclosure. (172) But more often the strategy succeeds, either in court or in settlement before trial. (173) This is a particular problem with "bottom-feeder" patent trolls who are only interested in coercing a nuisance-value settlement based on the cost and uncertainty of litigation. (174) The practical effect of not having a practicing the prior art defense is that some defendants who are actually using old technology are nonetheless held liable, and more are forced to pay settlements to avoid the risk of losing a case that, in theory, they should win.

    A final example of a scope gap in patent law operates against patentees rather than for them. Patent law is designed to cover new inventions but not to preempt abstract ideas altogether. (175) Patentees in computer software, following the playbook that has worked for them in the practicing the prior art example, have sought broader and broader interpretations of their patent claims, to the point where many claims are not limited either to a particular computer algorithm or approach or to a particular hardware implementation. Rather, they claim any computer configured in any way to achieve a particular result. (176) But in this case, the backlash against that functional claiming has been spectacular. The Supreme Court held in (2014) that claims of this sort are not patent-eligible at all, (177) and the lower courts have already invalidated dozens of patents on this ground. (178) The fact that the scope determination got pushed into validity made it an all-or-nothing exercise, and in this case patentees end up getting nothing as a result of their overreaching, even if they would have a valid narrower claim to a particular software algorithm.

  2. Trademark

    Trademark law has, especially relative to copyright, several validity doctrines with some bite. Most importantly for our purposes, valid marks must be distinctive and nonfunctional. (179) But courts tend to rely primarily on the likelihood of confusion factors and on certain defendant-specific doctrines like descriptive fair use to manage the scope of trademark rights. Validity and infringement are interrelated here. Distinctiveness, trademark law's core validity concept, significantly affects scope through the likelihood of confusion factors precisely because distinctiveness can only be assessed in relation to particular goods or services. (180) Thus, though trademark rights are no longer strictly limited to the goods for which the plaintiff uses the mark, (181) those goods continue to anchor the likelihood of confusion analysis through the similarity of goods factor. (182)

    But similarity of goods is only one dimension of relevant similarity, and infringement is determined by evaluating the confusion factors flexibly, with no one factor being dispositive. (183) In general, we can say that the more similar the mark and the more similar the goods or services, the more likely the use infringes, but there is no definitive threshold level of similarity of mark or of goods that makes another's use noninfringing. (184) And, importantly from our perspective, while the confusion factors require assessment of confusion as a result of the defendant's use of a mark and courts therefore often focus on features of the relevant marks, none of the factors explicitly seek to ascertain whether the features of the defendant's mark that are causing the confusion are features that define the plaintiff's protectable interest.

    This isn't to say that the confusion analysis doesn't in some ways take scope into account. In fact, it is black letter law that the amount of similarity necessary to support a likelihood of confusion finding is a function of the strength of the plaintiff's trademark. (185) Strong trademarks get broader protection. (186) And while strength relates primarily to consumer recognition of a mark, in assessing strength many courts consider whether others in the relevant market also use the claimed mark to identify their own goods or services. (187) When a number of parties are using the same or a similar mark, each party's rights are narrow. (188) In Eastland Music Group, LLC u. Lionsgate Entertainment, Inc., for example, the court noted the extensive prior use of "the phrase 50/50 or a sound-alike variant (50-50, fifty-fifty, fifty/fifty)" as the "title of intellectual property" and therefore held that, by entering such a crowded field, Eastland's rights in its Phifty-50 mark, which it used for a rap duo, were "weak and narrow." (189) The Federal Circuit recently expounded on this doctrine at length, suggesting that it may play a greater role at the PTO at least. (190)

    Courts have been most successful in aligning scope at the validity and infringement stages when, as in Eastland, they expressly consider those two issues together and specifically describe the aspects of the plaintiff's mark that warrant protection. In Star Industries, Inc. v. Bacardi & Co., the court found Star Industries's "O" design inherently distinctive because it "had sufficient shape and color stylization to render it slightly more than a simply linear representation of an ellipse or the letter 'O'" and was "a unique design in the alcoholic beverage industry at the time it was introduced." (191) But, the court noted, "the extent of stylization was marginal at best. The outline of the 'O,' though not uniform, is ordinary in its slightly varying width, and the interior and exterior borders are also ordinary." (192) As a result, Star Industries had only a "'thin' or weak mark, which [was] entitled to only limited protection." (193) Having called out the specific aspects of Star's "O" design that made it protectable, the court could compare Bacardi's "O" with those specific features in mind. Unsurprisingly, the court then found that Bacardi's orange "O" design was not likely to cause confusion.


    In our view, a court should come to a similar conclusion in Adidas's claim against Marc Jacobs. Adidas, which owns a three-stripe design mark for shoes that it also applies to clothing, has sued Marc Jacobs for putting four stripes on its clothes. (194) Despite the similarity in some of the colors and the width, we doubt that a mark consisting of three vertical stripes could be broad enough in scope to cover anything but an essentially exact copy. (195)


    But courts have had a much easier time doing this with relatively simple marks--single letter designs like a stylized "O" or combinations of letters--than with compound, contextual marks. Thus, we can imagine a court applying this principle to deal with Lagunitas Brewing's recent suit against Sierra Nevada Brewing. (196) Lagunitas claimed that Sierra Nevada infringed its rights in an "'IPA' family of trademarks." (197) Lagunitas admitted that "other brewers have adopted the shorthand parlance of 'IPA' to market their India Pale Ales," and indeed the "IPA" designation is in widespread use by a variety of different brewers. (198) The term IPA itself is therefore likely generic and unprotectable.


    Nevertheless, Lagunitas insisted that the...

To continue reading