Amongst one of the most archaic legal arguments one can engage in, the question of whether free software licences are contracts or mere licences is up there with similar arcane questions about monkeys having rights. Thankfully I seem to be a specialist in those exact types of questions.
“In US law, as in other common law systems, property law distinguishes between contractual licenses and “bare licenses”. A license can be considered as contractual if certain conditions are present: offer, acceptance and consideration. But a license can also be considered as a “bare license”: a unilateral legal permission by which a licensor permits a licensee to do something she would not have been allowed to do under the law. Therefore, in our case, bare licenses are “copyright licenses”, unilateral permission to use works under certain conditions, whose violation is sanctioned by copyright injunctions.
The differences between these two regimes are numerous. Unlike a contract, a single license does not require the consent of the licensee: it simply indicates the conditions where the use of the work is not a copyright infringement. If these conditions are not met, the license ceases to exist, and the offender is committing a copyright infringement. So violations of bare licenses are sanctioned under copyright’s strict liability regime, with harsh statutory damages. Moreover, while the termination of a contract is limited by law (and contractual terms), bare licenses are revocable at any time.”
In other words, whether a free licence such as the GNU General Public Licence is considered a contract or not could determine which legal regime applies to it, and whether or not it would be easy for a licensor to bring an action under copyright infringement or breach of contract. There are lots of legal implications, including damages and contract formation that could be different if one considers one or the other.
This debate has been more prevalent in the United States, and to a lesser extent in other Common Law jurisdictions such as England and Wales. This is because in these systems contract formation includes something called consideration to the normal requirements of offer and acceptance. Consideration is usually understood in these systems as a form of payment, or more accurately, something of value has to be given in return for the contract to be valid. Many early objections to open and free licensing schemes opined that these licences were invalid because there was no consideration, they were pretty much offering software for free. The argument that many of us made to defend the validity of open licences is that many of the obligations present in the terms and conditions are enough to contain consideration (such as the copyleft clause), and also the fact that in most legal systems of the world, consideration is not part of contract formation, offer and acceptance are enough to have a valid contract.
The main ruling dealing with the contract/licence dichotomy came in the US decision of Jacobsen v Katzer, where the 9th Circuit had to decide on this very legal question for the purpose of declaring the validity of the Artistic License, an open source software licence. In that decision, the 9th Circuit judges opined that open licences contain “enforceable copyright conditions”, which means that a violation of an open source software license exposes the licensee to more than a breach of contract claim, but also to a claim of copyright infringement. This is because the contract does not exist, and therefore any use of the works covered by the licence would amount to copyright infringement.
This was hailed as a victory for open source in general, as it makes it more enforceable in many ways.
Now we have a new decision that sheds new light on the arguments, and it is in the ongoing case of Artifex Software v Hancom (the decision relates to a motion to dismiss). The case involves plaintiff Artifex Software, the developer of Ghostcript, a popular postscript and PDF interpreter; and Hancom, a South Korean company that develops alternatives to word processors and office suites such as Word and Microsoft Office. Hancom started incorporating Ghostcript into its own product in 2013, but Ghostcript is released under the GPL, which contains certain conditions, which Hancom failed to comply with. Amongst other, the GPL requires that the licensee clearly identifies the licensor in any derivatives, as well as making the derived source code available to the community. After receiving a complaint, Hancom removed Ghostcript from its software in 2016, but the plaintiffs sued anyway for breach of contract and copyright infringement. The main argument posed by the defendant was to argue that the claimant had not made a valid breach of contract argument, and that such a claim pre-empts the copyright infringement claim.
This is at the heart of the question, and why this is yet another interesting addition to the corpus of open source software litigation. The argument by the defendants is that there is no contract, and if there is no contract there cannot be a breach of contract, ergo the licence is still valid and there is no copyright infringement. However, the court did not buy this argument. In a strong declaration that online open source licences are contracts, the court declares:
“Defendant contends that Plaintiff’s reliance on the unsigned GNU GPL fails to plausibly demonstrate mutual assent, that is, the existence of a contract. Not so. The GNU GPL, which is attached to the complaint, provides that the Ghostscript user agrees to its terms if the user does not obtain a commercial license. Plaintiff alleges that Defendant used Ghostscript, did not obtain a commercial license, and represented publicly that its use of Ghostscript was licensed under the GNL GPU (sic). These allegations sufficiently plead the existence of a contract.”
There are further considerations by the court related to pre-emption in copyright law and jurisdiction, but these are less relevant to the current post. The court denied the motion to dismiss, and the case continues.
There are a couple of interesting aspects here for future reference. Firstly, we continue to get decision after decision that declares the validity of open source licences; not only that, we get decisions about the contractual validity of the licences. It was not so long ago that I had to defend in a conference the claim made by some lawyers that free software licences were invalid. Secondly, we are starting to see that the question of whether these licences are contracts or mere licences will continue to play out in US courts, and that despite Jacobsen there is more nuance in the debate. It seems like judges will have to analyse the facts of the case in each particular situation.
However, we can rest assured that the GPL continues to be a contract where and when it really matters.