Defining music NFT ownership, from the digital to the analog world

tl;dr: Music NFT headlines this year have focused on astronomical sales… but what is actually being sold? In speaking with lawyers and annotating music NFT contracts, we found few details — and in fact many contradictory statements — about what NFT “ownership” really means, let alone how artists and platforms can be held accountable for delivering on their promises to buyers. This gap in education and communication is rooted in a fundamental tension between the simplicity artists crave in their agreements and the complexity of the global music copyright ecosystem.

This is Part II of a five-part, collaborative research report that the Water & Music community has put together over the last two months on the state of music and Web3. Contributors to this research thread on legal issues around music NFTs are listed at the bottom of this page, sorted by role, alongside a list of contracts we annotated as part of our research process and a collection of historical resources on music/Web3 standards.

Part I focused on the burgeoning market for generative music NFTs. You can view the current state of our report rollout, and a full list of our member-contributors ,by visiting stream.waterandmusic.com.

Music NFTs have generated over $80 million in primary sales in the last year, driving an unprecedented wave of development and experimentation around blockchain technology in the music industry. Top proponents of music NFTs have praised the power of decentralized infrastructure to remove entrenched industry middlemen and make the traditional functions of record labels, PROs and other stakeholders obsolete. Instead, this new infrastructure is allowing a new generation of musicians and creators to see the majority of the fruits of their labor.

But behind the explosive financial growth and hype lies a surprisingly shoddy foundation of knowledge and awareness about what people are really buying when they buy a music NFT — and whether, legally speaking, the technology actually “solves” any fundamental issues around the complexities of music copyright ownership.

All this in spite of over five years of development and funding, and hundreds of pages worth of whitepapers that enthusiasts and investors have poured into blockchain-based solutions for the music industry. As early as 2015, artists and developers have been trying to use blockchain to tackle the music industry’s largest problems — from obtuse legal agreements and complicated copyright laws that lag behind innovation, to difficulties around accurate creator attribution and timely royalty payments (let alone the continually shrinking pie of available royalties in the first place).

Early on, the argument was that blockchain would be a B2B panacea for the music industry, providing a decentralized ledger to display music rights and ownership metadata more transparently, along with “smart contracts” to verify and automate royalty flows fully. Many startups and projects like Ujo Music (under ConsenSys), JAAK and Berklee/MIT’s Open Music Initiative sprung up between 2015 and 2018 to take on this ambitious challenge of creating decentralized licensing and payments infrastructure for the global music business.

But most of these projects proved too big in scope, and/or crumbled under the pressure of navigating entrenched industry politics and attempting to convince competitors to share their rights data in a database accessible to everyone in the industry. As Jack Spallone, head of crypto at HIFI Labs and former project lead at Ujo Music, put it recently in a Water & Music Twitter Space: “We tried to work with the incumbent industry to modernize their systems,” but “it was probably a little naive to assume you could register all the global catalogs of music and make it programmatically licensable.”

Fast-forward five years later, and part of the landscape seems to have taken these lessons to heart, most specifically around NFTs. Most of the activity surrounding music NFTs is not only happening on a micro scale — namely, involving a single artist and anywhere between one to a few or perhaps 100 buyers, instead of trying to shift thousands or millions of songs on-chain — but is also concerned with creating entirely new economic models around music from scratch, instead of merely trying to bring legacy legal frameworks onto Web3 rails.

Based on this new emerging paradigm, it seems that there are two distinct philosophies at play with respect to how to apply blockchain to the music industry at scale:

1. Bring legacy music industry structures into Web3; or
2. Break with the past and create new systems.

In this binary approach, the first option excludes the other, and vice versa. Yet, both require uniform adoption at scale to have genuine impact. Which of these ideologies you adopt thus becomes a key question to everyone working in this ecosystem. That said, what should those who are active in Web3 do today? What legal concerns should they be aware of?

For this article, we interviewed several lawyers and annotated 40+ pages worth of contracts of recently released music NFT projects, to think through how the two approaches laid out above play out in the current landscape. (You can view the full list of annotated contracts in the appendix at the bottom of this page.) Given that there has already been plenty of more intellectual, abstract debate about legal issues with music NFTs over the past several years, we hope to add to the ongoing discussion by grounding it in more concrete, analog-world examples of high-profile NFT contracts that are being presented to collectors and investors right now — highlighting major legal gaps and red flags that still remain in the modern landscape and how those gaps could tangibly affect artists and fans in the short term, for better and for worse.

Obligatory disclaimer: This article was not written by lawyers and its contents are written only to convey general information. The contents of this article do not provide legal advice or opinions. The contents of this article should not be construed as, and should not be relied upon for, legal advice in any particular circumstance or fact situation and may not reflect the most current legal developments. No action should be taken in reliance on the information contained in this article [and we disclaim all liability in respect to actions taken or not taken based on any or all of the contents of this article to the fullest extent permitted by law.] An attorney should be contacted for advice on specific legal issues.

Copyright 101 — or, what we’re really talking about when we talk about “ownership”

From a legal standpoint, ownership is a straightforward concept: It’s about who holds certain rights to certain property, intellectual or otherwise. However, ownership of music is anything but straightforward.

Generally, intellectual property rights are an attempt to balance the benefit to society of open IP with the need to incentivize inventors and creators with an initial exclusive right to consideration (a legal term that refers to a benefit in the form of money or goods established between two parties) for their works, in order to promote progress. The most common IP rights are trademarks, copyrights, patents and trade secrets — though there’s many more and and when dealing with public figures, name & likeness rights and rights to publicity are sometimes discussed in the same breath. For our purposes, understanding the basics of a copyright may be most helpful, though trademark rights and rights to publicity are also important when thinking about recent NFT projects.

A copyright is a right of exclusion. Namely, it gives the owner of the copyright the exclusive right to copy and distribute the creative work. What makes music a unique copyright is that within every sound recording, there are multiple copyrights: The right to the underlying musical composition (think, the sheet music), the right to the sound recording (i.e. this specific recorded version of the underlying musical composition) and the right to perform the composition or broadcast the specific version of the recording in public (i.e. “public performance rights”). So, the different instances of 1) playing a recording, 2) making a copy of a song, 3) distributing or broadcasting a song and 4) playing a song in timed combination with any visual content (say, in a Netflix show) may each require different and/or multiple licenses, usually from various entities who own different aspects of these copyrights.

It’s no surprise, then, that copyright ownership in music has a complicated history. Whether it’s Michael Jackson buying the rights to the Beatles catalog and the subsequent to-ing and fro-ing, Taylor Swift trying to wrest back control of her sound recordings from her former label and various investors or a Pulitzer-winning journalist adding himself as a songwriter to hundreds of compositions — most major changes of hands in music copyright are met with some level of controversy, illustrating the flaws in a system where multiple different parties jockey to own multiple kinds of copyright in a single work.

Do NFTs inherently fix these flaws? While they may deliver certain kinds of benefits to artists, and while blockchains do allow for more transparent attribution mechanisms (assuming they are fed accurate data), the technology does not inherently “fix” issues around ownership complexity in the music industry.

Sophie Goossens, a partner at Reed Smith LLP, explains this issue as follows. What an NFT brings to the table is a form of ersatz ownership — namely, an ownership established by contract. So while there is no such thing as legal property in a digital file — and, therefore, no such thing as legal ownership of digital property —two parties can still agree that such a thing exists. Goossens thus sees NFTs as adding a third layer of ersatz digital ownership onto intellectual property, where previously you had solely a tangible item and the IP rights connected to it. The structure then looks like this:

1. Creator owns the IP,
2. Analog-world buyer owns the physical item,
3. Owner of the NFT on blockchain has the ersatz ownership, created by contract.

In other words, this rejected New Yorker cartoon is pretty accurate (if, for Lord of the Rings fans, we imagine Sauron still owning the IP to his creation that he sold to Frodo):

While the NFT adds another ownership layer, a blockchain entry itself isn’t a change of ownership in the original work. That change still needs to be legally verified in the “analog world.” This rings especially true when the acquisition of a music NFT presumably comes with the attribution of associated IP rights, and/or future royalties related to said IP. Several musicians including 3LAU, Lil Pump, Lyrah, Vérité, Jacques Greene and Daniel Allan have experimented with delivering revenue shares on streaming royalties as the primary “utility” of their NFT sales and crowdfunds, with contract terms that are optimized to remove legalese for the average fan.

Upon analyzing the “fine print” in the underlying agreements attached to several of these rights-transfer or royalty-share NFTs, however, it becomes clear there is a large gap between the understanding of legal ownership in music copyrights and recorded ownership on a blockchain. Namely, all the music NFT contracts we analyzed were light on details and often included contradictory statements — particularly conflating ersatz digital property ownership with “analog-world” intellectual property ownership. While perhaps excusable for one-off experiments, these kinds of conflations could have significant negative implications down the line as the music industry tries to adopt blockchain infrastructure at scale.

Red flag #1: It’s never as easy it sounds

With a lot of modern music NFT contracts, artists seem to be trying to leverage blockchain as a tool for simplicity — particularly around more appropriately providing consideration to the original creator of a work, or more transparently splitting royalty revenue among multiple stakeholders. But when you speak to lawyers about how striving for this simplicity plays out in practice, it becomes clear it is at the expense of accurately describing the ownership and control of the IP involved. While simplification in general should be a goal, simple contracts can be very problematic, especially in a music-industry context when an artist wants to exploit a given musical work at a high level across many uses and mediums.

Let’s look at Lyrah’s NFT contract for “Taken” as an example, which was generated with the help of the CreateOS platform:

The first sentence in the green box — “with this NFT, you get 25% of the master recording, meaning you’ll get 25% of the streaming royalties that this song makes from platforms like Spotify and Apple Music” — is wrong. If you get 25% of the master, that doesn’t equal 25% of the total streaming royalties; recall all the different copyright owners who will receive royalties for their different ownership in the song. This is a pretty flagrant error, and highlights the amount of good faith needed between parties for these simple agreements to work. For smaller, more emerging artists, whose copyrights are not yet spread between different entities, and who operate on a scale where good faith is manageable between artist and fan, it’s a much easier task than for an artist signed to a major label and who has a different publisher and a separate sync deal.

Several other examples abound of artists publicly conflating (intentionally or otherwise) different kinds of ownership or copyrights in music. Earlier in 2021, Jacques Greene sold an NFT that included the publishing rights (i.e. the copyright in the underlying composition) to the track involved. However, this was not a true transfer of ownership; rather, it was a right to receive royalties, the terms of which were agreed off-chain between Jacques Greene and the buyer of the NFT. Similarly, Eugy’s Your Touch NFT, listed on Serenade, allowed the buyer to record a verse on the remix of Eugy’s previously released track “My Touch,” and granted the owner of the NFT a 25% share in the streaming royalties of that remix (only after “My Touch” reaches 15 million streams across all platforms). However, the Terms & Conditions related to the Eugy NFT purchase made clear these rights do not constitute an actual transfer of analog-world IP ownership.

Even many music NFT platforms themselves seem to be lacking in basic education about the different kinds of copyrights implicated in a song. For example, the landing page for the Republic / Lil Pump NFT drop states to investors that “you will get a share of potential profits generated by the master record,” and then proceeds to display the following graphic:

As one of our members Jonathan Larr pointed out in our accompanying annotations for this drop, this is “immediately concerning” because many of these listed sources also generate composition royalties, not just master royalties. “Interactive streaming generates sound recording royalties and composition royalties. Downloads pay mechanical royalties which go to the composition. Synch can pay master and/or composition depending on what’s used, etc.,” wrote Larr. “It would worry me from the start that they’re using these terms incorrectly.”

Again, when we are dealing with singular 1-of-1 NFTs and/or independent artists, these nuances may not matter as much. But imagine if 1,000+ people purchase shares in the royalty stream of a song, or if these NFTs are for songs with multiple writers, or if major-label artists (whose labels may actually own or have exclusive rights to the master recording) try to create their own music NFTs. Or, what if the NFTs representing those royalty shares are sold on secondary markets, or the artist records newer versions of their masters (à la Taylor Swift)? Not to mention the fact that copyright laws and notions of “ownership” differ from country to country. Further, what happens if an artist is sued for infringing the work of another artist; are you now also liable for 25% of their losses as an NFT owner? And if you buy the NFT thinking you “own” it, will you be satisfied with a mere right to receive royalties? Or would you want the right to create derivative works, remixes or control the work yourself?

With the Republic / Lil Pump example specifically — if as an NFT investor you want to sue the artist over not receiving your fair share royalties, how would a court accurately determine what royalties they are even supposed to be getting if the initial information presented is inaccurate? If a court is unable to determine what royalty streams should be flowing to them (e.g. whether there should really be masters and composition royalties involved), then, at best, that leads to complex litigation over the intent of the contract drafters in the first place. A court could just as easily decide they are entitled to both kinds of royalties — or to neither.

… You can see how this all can get complicated quickly. Clearly, smart contracts on the blockchain — with all the benefits we’ve described — are still not yet a match for the complex, global music copyright ecosystem.

It is worth noting that the relationship between the NFT creator and the NFT purchaser/investor in these situations can seem analogous to the traditional relationship between the record label and the musician. But that is not in and of itself a negative, as investing in artists can certainly help further their careers. Perhaps the solution isn’t necessarily to alter the way creators offer rights as part of their NFT sales, but rather to ensure purchasers have clear understandings of what rights they are purchasing and how exactly they can benefit from these rights.

Red flag #2: Lack of financial regulation and accountability mechanisms

Smart contracts — transaction protocols that automatically execute when certain, predetermined, conditions are met, or simply a set of if-then functions — digitally establish ersatz(?) ownership related to NFTs on a public and decentralized ledger. However, as discussed above, this ownership isn’t automatically enforced by anything in the analog world — which, at least in a music context, arguably precludes any of these contracts from actually being “smart.”

What, then, secures all the promises that accompany many music NFT drops today, especially ones with ownership- or investment-related perks? And how can we make sure artists make good on these promises beyond a trust-based relationship (i.e. ensure that their NFT drops don’t become scams)?

“We need to track artist’s interactions with fans — do they give their rewards?” Dan Tauhore, head of growth at MODA DAO, a community aiming to create the Web3 tools needed for artists to retain full control over their music, tells us in an interview. In other words, fulfilling the digital or physical benefits that come with certain NFTs will become a key measure of success for artists’ NFT drops.

But most of the contracts we annotated were surprisingly light on even the most basic information one would expect in a contract where one party is expected to deliver some product or return to the other. For instance, most of the contracts had little to no details on the mechanics of how owners of mass-scale royalty investment NFTs will actually be paid their fair share of revenue (e.g. via which cryptocurrency or stablecoin, how often payments will occur, whether payments will be airdropped to collectors’ wallets versus obtained via a claim link, who would cover gas fees for these transactions, etc.). This is likely because this process would be confusing to explain and/or there are not many great tools currently out there to deliver these payments in a smooth, streamlined way — but these still seem like important details that should not be glossed over.

It isn’t just copyright laws that don’t perfectly apply to blockchain-related technologies, or the appropriation of rights through a digital file. There have been many questions raised by founders, lawyers and regulators alike, about whether NFTs, and tokens more generally, should be classified as securities.

The “Howey Test” from a 1946 Supreme Court case is how courts decide if something is an “investment” contract” and thus subject to securities laws. It is not yet established if or how the test applies to blockchain-based pursuits, so creators should always seek legal counsel based on the facts of their specific endeavors. That said, those advisors will question specifically:

1. What’s inside the token? Ownership or the transfer of other rights?
2. Is the token advertised or marketed as an investment derived from the efforts of others?

Would recent music NFT drops provide affirmative answers to these questions? Let’s look, for example, at composer Junkie XL’s NFT drop on AmplifyX in June 2021. The “winner” of the NFT will work together with Junkie XL to create a soundtrack of their life — a.k.a. a 20-minute score created by Junkie XL with input from the NFT owner. While the AmplifyX motto clearly states that buyers (or “winners,” in their terminology) make an “investment” in artists, the actual Terms & Conditions related to the NFT talk about a license, which does not involve a transfer of ownership. While clause seven of their T&C does discuss a potential shift of ownership if the winning bid crosses a threshold of $250,000, even then what the winner can and cannot do with the work reads a lot like what a purchaser could do with, say, a DVD. In other words, it’s a clear case of owning the NFT, the digital item, but having solely a license to the work to be produced: A soundtrack to the winner’s life. A red flag may be raised due to the terminology involved, but the actual NFT may not constitute a security under the Howeytest.

On the other hand, one example that seemed to specifically be offering an investment vehicle is Opulous’ S-NFT, which they created with Republic. From their announcement (emphasis added):

“NFTs offered as a security instrument creates a new landscape of opportunities for artists. Security NFTs are designed to provide efficiency in distributing investors’ share of royalties directly into their crypto wallets.”

And yet, when we look at the actual language of the agreement for the first S-NFT, a drop with the rapper Lil Pump, we find the following (emphasis added):

“Neither the Offering nor the Securities have been registered under federal or state securities laws, leading to an absence of certain regulation applicable to the Company.”

Should Opulous have registered this S-NFT with the SEC? As with other regulatory issues related to cryptocurrencies, there’s a lack of consensus. In general, the SEC’s role is to protect investors through the regulation of securities; Goossens and other lawyers thus argue that what matters isn’t whether something is sold on a blockchain or elsewhere, but rather whether the item — the potential security — is a fungible, and passive, financial asset. NFTs are, by definition, non-fungible, though other kinds of tokens and protocols (e.g. social tokens) are.

While there is no settled law to draw on (yet), it seems inevitable regulators will step in at some point. Whether this is in five years or 10 years depends on the speed of development and, more importantly, whether private litigation increases.

In particular, with a general lack of education around how music copyright royalty flows work, there seem to be many cases where retail investors are completely disconnected from the economic reality of the royalty they are buying. Take, for example, the contract for 3LAU’s “Worst Case” drop on Royal, in which the artist freely distributes 333 “assigned shares” of on-demand streaming royalties to collectors — meaning each share ultimately represents a meager 0.15% slice of the streaming pie for this one song. And yet, many of these 333 NFTs are now listed on OpenSea at a floor price of 3.25 ETH, or around $14,000 as of publishing this article. The number of streams it would take for the “Worst Case” royalty payments alone to make it a 3.25 ETH price tag a worthwhile investment is astronomical; the NFT is being treated much more like a short-term speculative asset than a long-term productive one.

Is the ability to immediately resell the royalty-bearing NFTs on a secondary market putting the retail investor at risk? There would be extra cause for concern if the artist were to retain an interest (royalty) in the subsequent sale, which is common practice in NFTs. In contrast, the Republic/Lil Pump “S-NFT” offering states that they fit within the SEC’s Rule 144, which is an exemption from the regular SEC filing requirements used commonly for seed funding and employee/executive stock benefits. Under this requirement, though, investors are required to hold onto their S-NFT for a 12-month period — which can feel like years in crypto time.

There have already been a few examples of litigation surrounding IP ownership in NFT drops. Roc-A-Fella Records claimed Damon Dash couldn’t sell what he didn’t own when he advertised an NFT for Jay Z’s Reasonable Doubt. In their court filings, the label explained that while Dash owns a third of the label, that doesn’t mean he can sell a corporate asset. An NFT may just be a digital item, but when an NFT involves a transfer of copyright we immediately see a direct collision with analog-world legal boundaries. Similarly, when Quentin Tarantino minted NFTs in relation to his own screenwriting for Pulp Fiction, Miramax stepped in and sued. The issue isn’t so much the NFT in and of itself, but rather whether the rights that Tarantino has in relation to the film extends to this new technology of NFTs. Namely, the issue is about the exploitation of rights, and in the case of Tarantino minting a Pulp Fiction NFT immediately relates to a Miramax-owned product.

In music IP, with its multitude of copyrights and copyright holders in even a single song, this type of uncertainty about the exploitation of rights is only exacerbated — and we don’t even need Web3 for that uncertainty to be laid bare. These early legal tests are the tip of the iceberg which will lead regulators to adopt or adapt laws to tackle these new forms of ownership in analog-world policies.

Conclusion: How can we improve contracts today

So much that is currently happening in music and Web3 focuses on small-scale solutions for artists and their fans, which requires simple documentation to help people navigate new modes of support and interaction. If one artist can make enough money to support themselves and their art through NFT sales, that’s a win.

However, the complex structures of music IP do not provide the framework for simple solutions. This field of tension between the simplicity artists crave, and the complexity of the structures that their art sits in, led us to define two red flags in the modern music NFT landscape:

1. Simple language is taking precedent over the accurate and comprehensive description of rights awarded through NFT contracts.
2. There’s a distinct lack of specificity in these contracts over when an NFT is a speculation tool, versus when it’s a marketing tool.

So, what are some tangible improvements we can make to music NFT contracts today?

First of all, it’s about education and communication. Many artists, and definitely many fans, are unaware of the intricacies of copyright. Musicians and creators need to be open about what they’re promising, and how they plan to deliver on it. Fans who invest in their favorite artists via NFTs need to understand that any potential upsides might not happen, and that, in spite of splashy press releases suggesting otherwise, there are no easy get-rich-quick schemes here.

Beyond communication, there’s also a need to put in place clear mechanisms that allow fans to hold musicians and creators accountable for the utility included within their NFTs. This requires clear definitions of what that utility is, and how to claim it. If it includes any form of future revenues, we need clear communication again — about where exactly that revenue stems from, whether it’s just passive income or can also include active participation on the NFT purchaser’s end, and what happens when the rights to those future revenues shift hands again.

Blockchain-based solutions can make all of the above seem easy, in the sense of being resolved by if-then structures that trigger any utility included in an NFT. But the underlying problem is that of ownership, and recognition of ownership. And there, we hit a very fundamental roadblock for music in Web3.

In one corner, there are those people who have a higher risk appetite and are keen to experiment and transact with more emergent technologies and business models. An example of an organization with this mindset is the previously mentioned MODA DAO, which launched their token last month before the standards they aimed to develop were even clear.

In the opposite corner, there are those people who want to bring the entire legacy music industry on-chain, and are keen to establish broad-ranging standards for Web3 adoption. The issue is that it’s difficult to think of what those standards would be as new use cases for buying and selling music NFTs continually emerge in real time. Music/Web3 veteran George Howard — who, among other things, worked on The Song that Owns Itself — tells us in an interview that he strongly believes that “standards follow transactions … What gives anyone the hubris to think that they know ways in which customers will search for and utilize these assets, particularly in a completely new dimension?”

There is, perhaps, a difference here in who needs to claim responsibility for setting standards. MODA or Royal, for example, have a different place in the music NFT landscape than a single independent artist doing an NFT drop.

However, all of them look to benefit from the additional layer of ersatz ownership offered through NFTs, which we’re seeing get conflated time and time again with ownership in the analog world. Instead, we should push to more formally establish ersatz ownership as a layer of potential revenue generation, both for the creator and the owner of the NFT. To make music/Web3 experiences work long-term, we need to establish this new ownership layer as a core part of the legal fabric for the future of the music business.

Contributors

Cherie Hu (A, B, D)
Maarten Walraven (A, B, D)
Yung Spielburg (A, B, C)
Levi Downey (B, C)
Jonathan Larr (B, C)
Jillian Jones (B)
Brandon Landowski (B)
Brodie Conley (C)
Joshua Glazer (D)
Dan Smith (E)
Ana Carolina Laurindo (F)
Jack Spallone (G)

(A) Research project leads
(B) Writers/editors
(C) Core contract annotators
(D) Interviewers
(E) Other contract annotators
(F) Visualization
(G) Member sources

Appendix

Water & Music’s contract annotations:

3LAU’s “Worst Case” NFT contract on Royal
Lil Pump’s “Mona Lisa” S-NFT contract on Republic
Eugy’s “Your Touch” NFT contract on Serenade
Spottie Wifi x Bun B’s “All Time High” NFT contract

Historical initiatives and whitepapers on music metadata/contract standards:

#MTFLabs: Blockchain (2016)
Watermarking technologies and blockchain in the music industry (Digimarc, 2017)
Towards an Open and Scalable Music Metadata Layer (Berklee/MIT, 2019)
Fair Music: Transparency and Payment Flows in the Music Industry (Rethink Music, 2019)
The Genesis of Music3 (MODA DAO, 2021)
A better music NFT metadata standard (Garrett Hughes/crowdsourced, 2021)