This post explains why §301 copyright preemption splits matter and previews how each circuit treats contract‑based scraping claims.
I’ve written over 3000 words on this subject before, but this is probably the most understudied and under-litigated issue in the world of web scraping.
Preemption means that federal law overrides state law when the two are in conflict. The idea behind preemption is that certain legal issues must be handled at the national level rather than varying by state. Indeed, it is written into Section 301(a) of the Copyright Act that copyright law expressly preempts any state law that is “equivalent” to copyright. (17 U.S.C. § 301(a)).
But courts often disagree about when state laws are “equivalent” to copyright laws.
Let me provide a simple example, and the example that is most relevant in the context of web scraping. Almost all online terms of use prohibit copying and reusing data from their websites without permission. But when it comes to intellectual property law, copyright law is the domain of law that is typically used to determine when someone has the right to copy and reuse content without permission. This includes what types of content are subject to legal protection (creative works, not facts), what types of uses cannot be restricted (in the context of fair use, limited non-commercial uses, transformative uses), and what types of content belong in the public domain (old works, facts without unique organization or presentation).
In a very literal sense, online contracts that prohibit copying and reuse of data in any context expand on the protections of copyright law in favor of the website owner. But you’re not supposed to be able to unilaterally expand copyright law. Copyright law is what it is, and either you’re entitled to protection or you’re not. And a contract shouldn’t be able to expand that.
Which brings us to the key question here and what creates our first “circuit split.”
Does federal copyright law “preempt” state contract laws governing the enforceability of online contracts? In the Second Circuit, it usually does. So, too, in the Sixth Circuit. In the Ninth Circuit, it often does. And this was most clearly demonstrated in the recent X Corp. v. Bright Data case. In that case, Judge Alsup concluded that X Corp.’s prohibition on scraping in its terms of use was unenforceable because it was preempted by copyright law. According to Professor Guy Rub:
When you hear someone smart like Guy Rub explain it like that, it makes total sense. Online contracts that provide extensive expansions of copyright law should be preempted by copyright because they undermine the policies underlying copyright law.
Unfortunately, in much of the country, that’s not the law. In the Fifth, Seventh, and Eleventh Circuits, there is categorically no preemption of contracts by copyright law.
With that, in maybe 30% of the country, if a host website tries to sue a scraper for “copying and reusing” their information without permission in arguable violation of a terms of use, that claim would be preempted by copyright. In another 30% of the country, the web scraper would not be allowed to make that argument in their defense. In the rest of the country, the question of preemption of online contracts would be handled on a case-by-case basis.
Because web-scraping claims almost always boil down to a “you can’t copy and reuse our data” argument, that’s a big deal.
And given the sheer volume of state laws that apply to web scraping conflicts, a uniform federal understanding of preemption of copyrights is essentially the only way that I can think of where we’re going to develop uniform national law on web-scraping legal issues. And because we’re nowhere close to that, a comprehensive understanding of web-scraping law requires you to get into the nitty-gritty of state and federal differences. And that’s what the rest of the series is going to be about.