A few thoughts about today’s Perfect 10 v. Google decision.
First, I think the court was absolutely correct to reverse the District Court on the question of prima facie infringement by applying Kelly v. ArribaSoft.
Second, the court correctly rejected the possibility of Google’s vicarious liability for infringing posts of Perfect 10 material on various websites picked up by Google’s search engine.
Third, I am troubled by language in the opinion that appears to accept the possibility of contributory liability because a defendant knows with substantial certainty that infringement will occur. Although knowing with substantial certainty is a form of intent recognized in common law tort, Google’s behavior doesn’t fit the normal application of such intent.
Most large actors know with substantial certainty that their behavior will harm the interests of others. Bicycle manufacturers know that riders will eventually suffer injuries in crashes. Bridge builders know that workers will get hurt on the job. Alcohol sellers know that customers will eventually drive while intoxicated or that minors will eventually obtain liquor. Gun manufacturers know that someone will eventually use their weapons to shoot innocent people. None of these actors are liable for intentional torts. Their liability, if any, depends on a nuanced analysis of reasonableness of the sort found in negligence or products liability law.
Accordingly, Google may know with substantial certainty that its search engine will eventually help an infringer, but that is not enough to create liability. A court must still assess whether Google’s decision to operate its search engine was reasonable under the circumstances.
Granted, the court’s opinion leaves open the possibility of this nuanced analysis by remanding to the District Court. There probably does need to be fact finding about what Google knew and whether its behavior was reasonable in light of its knowledge. However, I’m still troubled by the court’s language that “Google could be held contributorily liable if it had knowledge that infringing Perfect 10 images were available using its search engine, could take simple measure to prevent further damage to Perfect 10’s copyrighted works, and failed to take such steps.” On the surface, this statement seems reasonable. Why shouldn’t Google take “simple” steps to protect copyright? My concern is that “simple” will be interpreted as “easy for Google” without any consideration for the social benefits that “simple” actions might destroy.
For example, Perfect 10 may argue that it’s “easy” for Google to program its search engine so that any inquiry including “perfect 10” returns only results from perfect10.com. To be sure, this removes assistance from those infringing Perfect 10’s images. However, this would also remove perfectly legitimate results like those concerning gymnastics scores, wikipedia entries or other editorial writings about Perfect 10, and references to the litigation. Thus, “simple” steps that Google could take to protect copyright make little sense for society at large. Accordingly, it’s perfectly reasonable for Google to use an unedited, automated process to find myriad references to “perfect 10,” even it some reference to infringement occurs, and Google ought not be held liable for failing to ameliorate all of its consequences.
I have elaborated on these views about third party copyright liability in my article “Third Party Copyright Liability After Grokster,” 91 Minnesota L. Rev. 184 (2006) (available from the Minnesota Law Review’s website).