Copyright’s “Double Spend” Problem: Digital First Sales
By Lance Koonce
For those interested in the evolution of digital currencies, I highly recommend Steven Levy’s “E-Money (It’s What I Want)” article from Wired way back in December 1994. It’s a great read, and presages many current developments.
One problem with cryptocurrencies that Levy just touches on briefly is the problem of the double-spend. In brief, when hard currencies are exchanged, the transaction can only happen in one place and at one time, but in the digital realm transactions do not involve the physical transfer of data. Rather, digital transactions technically involve the copying of data to another party, which leaves open the possibility that the original party can attempt to spend their digital currency twice.
Put differently, someone can potentially attempt to broadcast two digital currency transactions at approximately the same time, using the same digital coins, but the parties receiving the transactions have no incentive to agree that one transaction takes priority. Before the arrival of Satoshi Nakamoto’s Bitcoin, with its revelatory proof-of-work scheme, most solutions to the double-spend problem involved having a trusted third-party intermediary resolving double spends.
Interestingly, copyright law faces the same double spend problem, in form of the first sale doctrine.
The first sale doctrine allows resales of copies of physical works such as books, DVDs and CDs, which otherwise would be precluded by the Copyright Act as in infringement of the original author’s exclusive right of distribution. It is embodied in Section 109 of the Copyright Act, which says:
the owner of a particular copy or phonorecord lawfully made under this title, or any person authorized by such owner, is entitled, without the authority of the copyright owner, to sell or otherwise dispose of the possession of that copy or phonorecord.
The first sale doctrine is why you can find used bookstores in every town, and buy used copies of the Purple Rain vinyl album on eBay or at a flea market.
In economic terms, physical copies are “rivalrous” assets — while being consumed by one person, it cannot be consumed by another person. By contrast, non-rivalrous goods — like broadcast television — can be consumed by many people at once.
In any event, while you can find used bookstores and music stores, good luck finding a used mp3 website. They don’t exist, or if they do, that site would be deemed to be engaged in copyright infringement. That’s because, as interpreted by the courts in recent years, the first sale doctrine does not permit resales of digital content.
Jack Browning recently posted a great analysis on this blog regarding whether a new technology like the blockchain could revitalize the first sale doctrine under copyright law. Jack used a terrific analogy to explain the first sale doctrine, from the Hugh Jackman movie The Prestige. In the film, in order to create the illusion of teleportation, Jackman’s character — who had access to a cloning device only (just watch the movie, okay?) — had to murder his own duplicate each time one was created, so that there was only one copy in the world at any given time.
In the digital world, the problem is that it is difficult to kill the original work when you sell the original version to a third party — you are not effecting a physical transfer of your copy. Instead, what really happens is that you are making a copy of the work on a new computer, and the only way to get rid of the original is to delete it from the original location. To make sure that occurs, some trusted third party would have to ensure that the original copy was truly wiped from existence.
Sound familiar? It’s precisely the same issue cryptocurrencies face, which was (many believe) solved by the blockchain. The only difference is that it the Copyright Act that prohibits making duplicates in the first, whereas when one double spends cryptocurrencies it is up to the market (or perhaps fraud or similar laws) to stop it.
In the Capitol Records, LLC v. ReDigi Inc. case, ReDigi had created a system whereby it attempted to act as a trusted third party intermediary to eliminate the double spend of copies of a copyrighted piece of music, and thus to allow its users to take advantage of the first sale doctrine. It uploaded a seller’s music file to a cloud service, and then eliminated the copy on the seller’s computer, after which the buyer would download a copy of the file from the cloud and pay for it.
However, the court did not buy this approach. The first sale doctrine, it said, does not allow this type of digital “transfer”, but rather only covers “material items, like records, that the copyright owner put into the stream of commerce.” The court believed that ReDigi was not distributing such material items but instead was “distributing reproductions of the copyrighted code embedded in new material objects.”
Jack’s analysis focused on whether the court in the ReDigi case would have reached a different result had the technology in front of it been based on blockchain technology, which arguably might provide a much more secure way of guaranteeing that multiple digital copies of a work are not in circulation. He concluded, I believe correctly, that this court would not have reached a different conclusion given the way the judge interpreted the Copyright Act’s language.
But judicial interpretation is not the only path to change. In fact, the ReDigi court also said the following:
At base, ReDigi seeks judicial amendment of the Copyright Act to reach its desired policy outcome. However, “[s]ound policy, as well as history, supports [the Court’s] consistent deference to Congress when major technological innovations alter the market for copyrighted materials. Congress has the constitutional authority and the institutional ability to accommodate fully the varied permutations of competing interests that are inevitably implicated by such new technology.” … Such defence often counsels for a limited interpretation of copyright protection. However, here, the Court cannot of its own accord condone the wholesale application of the first sale defense to the digital sphere, particularly when Congress itself has declined to take that step.
The question then becomes whether the blockchain, or some similar approach, might provide enough certainty around digital transfers to persuade Congress that the first sale doctrine can be extended to cover such transfers, at least under certain circumstances. (Of course, this also flags a larger policy discussion: Do we want a robust secondary market in digital copies? In a world where streaming has largely overtaken holding digital copies for many forms of content, would this mark an advance or a retreat?)
From a technological standpoint, however, the blockchain on its own cannot solve copyright’s first sale doctrine problem for digital copies. While a blockchain-based system can provide a potentially unassailable record of ownership of a particular copy of a work and the substitution of a second copy that for all intents and purposes becomes , some other system still has to be implemented to automatically remove or disable the prior copy, which still resides with the seller. Some entity still has to act as the ReDigi in the equation.
Unless, that is, we can come up with a proof-of-work type of solution for digital resales as well. Some way for a decentralized network of peers to verify the deletion of the original copy and the creation of the new copy that’s replacing the original, perhaps in a shared database?
Any budding Satoshis out there who want to take on this problem?