The news has been full of various proposals for addressing the perceived evils associated with patent assertions by a patent owner that have been identified at different times as a patent troll, a patent assertion entity and a nonpracticing entity (hereinafter collectively referred to as a troll). The perceived evil arises from the concern that the troll is using the threat of initiating a litigation or actually commences a litigation as part of its efforts to secure licensing agreements requiring the target company/defendant to pay licensing fees for the use of patents that are controlled by the troll, but are not practiced commercially by the troll. This identification of the problem seems simple enough and there are a multitude of proposed solutions for addressing the perceived problem. However, as noted below, both the perceived problem and the proposed solutions suffer from a lack of clear definition and, ultimately, may be further proof of the law of unintended consequences.

As a starting point, there is no clear definition of a troll. If the definition is as simple as looking to whether the patent owner is not practicing the asserted patent, it is bound to include a variety of organizations that are otherwise perceived to be contributing to society. Thus, a General Electric that secures something on the order of 2,000 patents a year becomes a troll when it assigns patents to a spinoff that is designed to assert the patents with some of the benefits inuring to General Electric. While most will concede that General Electric is a viable commercial entity, it could fall victim to a law that includes a broad definition of a troll.

If the definition of a troll is refined to exclude entities that are otherwise engaged in truly commercial enterprises, it may still entrap organizations that are perceived to be contributing to society. Consider the number of universities that have very active research programs that generate a number of patents, many of which are in the most advanced technologies. According to the Intellectual Property Owners Association listing at, at least seven universities were in the top 300 entities granted U.S. patents in 2012. The Regents of the University of California received 357; the Massachusetts Institute of Technology received 216 patents; Stanford University received 182 patents; the Wisconsin Alumni Research Foundation received 155 patents; the University of Texas received 141 patents; the California Institute of Technology received 136 patents; and the University of Michigan received 97 patents. It is not likely that any one of these institutions will commercialize directly any of these 1,200-plus patents. However, these institutions are likely to license the technology and that means revenue for the school.

According to a Nov. 25, 2012, article in USA Today, Northwestern University received $191 million in licensing revenue in 2011. The University of Minnesota lost $75 million in licensing revenue in the two years between 2009 and 2011 and the University of Iowa saw a rapid $35 million-plus drop in revenues as a result of a patent expiring. The fragility and rarity of these revenue streams was noted by Richard Bendis, CEO of Innovation America, who is quoted in the article as saying, “Royalty income really depends on the hot drugs, hot products or hot technologies, and where they are in the market and the patent cycle.” This is an important point for all patented technologies because the market creates the demand and the patented technology must satisfy that demand in order to justify the payment of royalties.

As most universities are research-focused, many reformers are willing to give them some kind of a pass when it comes to the definition of a troll. While this may be attractive as a simple solution, it fails to recognize that there are many commercial entities that focus on research, and those entities protect the results of that research with patents. In addition, commercial researchers often partner with universities to do some of the very basic research that is not, of itself, commercially important but is critical to further development of the technology.

As can be seen from above, I do not believe there is a definition of troll that will meet the objective of controlling what is perceived as sham litigation in order to force a license. However, I do believe that the broader issue of forcing patent licenses through litigation needs to be addressed. To that end, I suggest that the solution rests in the courts and their control of the litigation process. Many district courts have already adopted patent rules that govern the patent litigation before them. Unfortunately, such rules are not universal and their application within the same district is not always uniform. The issue can be addressed in a manner that is universal to all entities that assert a patent that is not being practiced by that entity or its licensee. By way of example, a company that does not practice the patented technology would be prohibited from using the short form patent complaint set out in Form 18 of the Federal Rules of Civil Procedure. In order to use Form 18, the courts could require the party asserting the patent to identity its product or products covered by the patent or patents or identify a licensee who satisfies the requirement. If an asserting entity did not qualify for the use of Form 18, the patent complaint will need to meet the pleading standards under Bell Atlantic v. Twombly, 550 U.S. 544 (2007), and Ashcroft v. Iqbal, 556 U.S. 662 (2009). As most districts have an established base of case law under these decisions, the pleading standards are known and this places the troll on the same footing as plaintiffs in other litigations.

In addition to changing the availability of the Form 18 pleading standard, patent litigation could benefit from the adoption of uniform rules for the conduct of the litigation. For instance, an early determination of claim terms, as required by Markman v. Westview Instruments, 517 U.S. 370 (1996), will likely lead to the earlier filing of summary judgment motions. In addition to the Markman claim construction, it may be beneficial to clearly express a preference for bifurcation of liability and damages. It is undisputed that the cost of discovery, especially as it relates to sales and damages issues, is one of the swords that hang over a defendant. By limiting the initial discovery to liability, many of the protective order fights can be eliminated and disputes over the handling of commercial information among the litigants and experts can be delayed until there is a determination of liability or at least a determination that the defendant is not entitled to summary judgment on the basis to the court’s Markman claim construction.

Finally, there should be some defined standard for awarding attorney fees. The award cannot be based solely on the fact that one party to the litigation lost. Likewise, it cannot be based on the fact that one party is perceived to be a troll. It seems preferable to look at what stage of the litigation the matter is resolved. For example, summary judgment based on the Markman claim construction may create a presumption that the defendant is entitled to some award and the burden of demonstrating the reasonableness of the original complaint will fall on the plaintiff. Thus, the defendant will not have the burden of seeking fees in the first instance. This presumption would disappear if the case went beyond the grant of summary judgment.

I recognize that all of the above proposals will need to be vetted among the interested parties and that their implementation will not be any simpler than trying to define a troll with enough specificity to correct only the perceived problem without creating new problems.

Anthony S. Volpe is a shareholder at Volpe and Koenig, an intellectual property law firm with offices in Philadelphia and Princeton, N.J. He has corporate and private-practice experience in securing, licensing and enforcing all aspects of intellectual property rights.