An Empirical Study of Copyright's Substantial Similarity Test

Document Type



The substantial similarity test is copyright law’s dominant means by which courts determine whether a party has infringed another party’s copyright rights. Despite this, we have very little empirical evidence about what the test is and how courts apply it. To date, only a few empirical studies exist, and these are limited in several important ways, including with regards to scope, time periods covered, and volume of opinions. Mostly, courts, commentators, and scholars rely on anecdotal accounts of the test in both their conceptualizations and critiques of it.

To help provide a clearer empirical assessment of the test, this study examines a random sample of over 1,000 substantial similarity opinions issued between 1978 and 2020. This study covers opinions from district and appellate courts in every circuit and tracks a number of these opinions’ characteristics. These characteristics include: the subject matter and copyright rights in dispute; procedural posture; opinion date; the subtests, expert evidence, and copyright limitations used in the opinion; the sources of authority that courts rely on in their opinions; and outcomes for each part of the test and the case overall.

This review reveals a number of important findings. First, similar to other types of copyright litigation, courts in the Second Circuit and Ninth Circuit dominate the substantial similarity space, with the Ninth Circuit more recently displacing the Second Circuit as the primary venue for substantial similarity cases. Courts also rely on opinions from these two circuits more than any other source in interpreting and applying the substantial similarity test. Second, courts typically spend little time assessing whether a defendant actually copied from the plaintiff’s work. Courts mostly decide this first prong of the substantial similarity test as a matter of whether defendants had access to the plaintiff’s work, and they mostly favor plaintiffs on this question. Courts rarely rely on expert evidence or assess the two works’ similarities on this first prong, despite courts and commentators frequently opining otherwise. Third, the second part of the test, where courts assess whether the defendant’s copying amounts to improper appropriation, is characterized by significant heterogeneity. No dominant means exists for resolving this question, and, in any given opinion, a court typically uses multiple subtests and copyright limitations to decide this inquiry. Courts also use expert evidence more frequently under this prong of the test than the first part, a result that defies conventional wisdom. Finally, defendants win substantial similarity cases slightly more frequently than plaintiffs. This is further evidence that prong one of the test, where plaintiffs enjoy significantly greater success, appears to be largely inconsequential. The data also suggest that one of the keys to winning, for either defendants or plaintiffs, is the extent to which the court engages with and discusses copyright limitations.

We conclude by considering several implications. First, courts should maintain the two traditional prongs of the substantial similarity test as distinct and reinvigorate their assessments of similarity under the first prong of the substantial similarity test. Second, courts should make discussing and applying any relevant copyright limitations the heart of their prong two analyses to ensure that copyright law serves its constitutional purposes.


UC Irvine Law Review

Publication Title

UC Irvine Law Review

Digital Object Identifier (DOI)