Thread by @grimmelm

I tweeted briefly yesterday about property and copyright in NFTs in response to the story that Seth Green's planned White Horse Tavern show might be on hold due to the theft of the Bored Ape NFT it was based on.

Today, I want to spell out the issues in a little more detail. 1/

The inciting incident is this story from BuzzFeed News. In brief, Green owned the NFT for Bored Ape #8398. He licensed it for a series, White Horse Tavern. Then someone phished Green, transferred the NFT, and sold it to "DarkWing84" for $200,000. 2/

buzzfeednews.com/article/sarahe…

I said that Green "owned" the Bored Ape #8398 NFT, and I think that term is accurate. NFTs are property, both in the colloquial sense and in the legal sense. 3/

The important part is that someone (e.g., Green) can have control over them, via control of a private key. Whoever knows that private key has the exclusive power to decide what happens to the NFT. Exclusive controllability is the essential prerequisite for property. 4/

Some people might object that an NFT is just some made-up collective hallucination on a blockchain, not an actual thing like a car or a book that can be possesed and owned. But lots of other made-up things can be owned, like patents and corporate shares. 5/

We've been through this before. In Kremen v. Cohen, the court held that a domain name could be property. Whoever controls the registration controls how the domain name is used and transferred. The same basic ideas apply to blockchain-based assets. 6/

scholar.google.com/scholar_case?c…

The next important point is that it is possible to coherently say that the hacker "stole" Bored Ape #8398 and that Green remained the "owner" when this happened, even though the hacker now had control of the NFT and Green didn't. 7/

This is also a familiar distinction in property law: between ownership and possession. If you pick up my computer and walk away with it, I have ownership and you have possession. I can sue you to get it back. If you intended to keep it, you can be arrested for theft. 8/

In short, one of the most basic jobs of property law is to restore *possession* to the rightful *owner* of property when the two have come apart. That is why property law exists. 9/

But just as some changes in possession do not transfer ownership, some changes do. If I *give* you my computer, you become the new owner upon delivery, i.e., when I give you possession. I cannot now sue you to get it back, because you are now the owner. 10/

Thus, another basic job of property law is to sort out "title": who is (and who is not) the owner of a given piece of property. Gift transfers title; theft does not. 11/

So to say that the hacker "stole" Bored Ape #8398 is to say that there was a change in possession but not a change in ownership. Seth Green still has title; the hacker does not. If the hacker still had it (and could be located), Green could sue to get it back. 12/

Some people argue that title to blockchain-based assets is (or should be) determined exclusively by the blockchain. Whoever knows the private key to the address associated with an NFT has title. Either the legal system stays out, or it conclusively defers to the blockchain. 13/

Without passing on whether this is even theoretically possible, I just want to say that this is not something that just happens. Once the legal system decides that something is property (and I think NFTs clearly are), it will treat them as such unless clearly kept out. 14/

So keeping property law away from NFTs would require some kind of contract or other agreement saying that all rights in them are determined exclusively by the blockchain … and there is no such contract anywhere in sight when it comes to Bored Apes and other familiar NFTs. 15/

(Actually writing such a contract, and getting it to bind everyone, is an extremely hard problem for a public blockchain — so hard that I think it is impossible without some kind of enabling legislation. Someone could just *use* the blockchain without agreeing to your terms.) 16/

The next question is who actually has title to Bored Ape #8398. Property law recognizes some changes of possession as sufficient to transfer title; others leave title where it was. 17/

For example, sales and gifts transfer title. So does inheritance. On the other hand, theft does not transfer title. Neither does robbery (in which the the robber forces the owner to "voluntarily" hand over the item). Property law has to sort out which actually happened. 18/

What about phishing? A typical phishing scheme involves what would be called "fraud in the factum": the owner is tricked into signing a document by lying to them about what the document actually says. Such transactions are ... you guessed it ... void. 19/

Another complication arises when the hacker resells the NFT to DarkWing84. This is the problem of the "good faith purchaser." It is not always obviously fair to let the owner reclaim their property from someone further down the line — that person may be just as innocent! 20/

The person in the middle — the thief, fraudster, or hacker — is at fault. But the thief may have absconded with the cash, which means that either the owner or the buyer will end up bearing the loss. 21/

Because there is no one obviously right answer here, different legal systems use different tests to decide whether the good-faith purchaser should get to keep the property, or whether they need to return it to the owner. 22/

One widely agreed-upon rule is that only a *good-faith* purchaser is protected. A buyer who knows about the theft has "notice" that the thief lacks good title to the property, and therefore doesn't get good title, either. 23/

In the context of blockchains, this can be tricky! For example, OpenSea flags and blocks listings for NFTs that it regards as stolen. There is a plausible argument that this creates notice as to those NFTs, such that even a non-OpenSea sale of one is not in good faith. 24/

Even without this issue, jurisdictions disagree on the rule! Without getting too far into the weeds, there's a strong argument that some countries' laws protect the good-faith purchaser of a stolen NFT, and some do not. 25/

https://twitter.com/grimmelm/status/1478973225744875520

https://twitter.com/grimmelm/status/1485720863899107330

And where countries' property laws disagree, there is a difficult choice-of-law problem as to which one applies to a worldwide blockchain. The traditional property law rule is that the law of the "situs" controls: the place where the property is located. 26/

But on a blockchain, the location of an asset is … ??? Neither the NFT nor the blockchain itself has any particular location. The owner, the thief, and the buyer could all be in different places! So this is just an additional layer of uncertainty. 27/

And now we come to copyright! How does the BAYC license attempt to deal with all of this uncertainty about who "owns" an NFT? It doesn't. Not at all. Not in the slightest. 28/

boredapeyachtclub.com/#/terms

The BAYC license says "When you purchase an NFT, you own the underlying Bored Ape, the Art, completely. Ownership of the NFT is mediated entirely by the Smart Contract and the Ethereum Network:" 29/

I literally have NO IDEA whether this license attempts to say that the "owner" of an NFT is the property-law owner, the blockchain-possessor, or something else. It does not appear that they even considered there was any ambiguity about this. 30/

So to the extent that the BAYC license gives a copyright license to whoever owns the Bored Ape #8398 NFT, it is not clear on who that person is. Green? Maybe. DarkWing84? Maybe. So the license is not clear on who can grant a license for a show like White Horse Tavern. 31/

Next problem! What happens to the copyright license granted by Green if ownership of the NFT has passed to DarkWing84? Does it terminate? Continue? If it continues, to whom are royalties payable? Can DarkWing84 exercise Green's powers under the license? 32/

The answers to these questions depend on the BAYC license — which says absolutely nothing about them, even though the idea of derivative work sublicenses is utterly central to the supposed advantages of the BAYC license. 33/

The answers to these questions also depend on the license Green and his collaborators entered into. Guessing from his initial panicked response, it would appear that they did not think through and deal with those issues, either. 34/

In particular, the fact that Green is now on Twitter making arguments about "stolen art" and "exploitation of the underlying IP" show such confusion about the NFT/copyright/property relationship that I am inclined to think that the license does not deal with them coherently. 35/

In particular, Green seems to be saying that there could be a sale of the "art" that does come with rights to the "underlying IP" and I have … questions … about how he and his legal team think this relates to the NFT and the BAYC license. 36/

To sum up, an NFT copyright license needs to deal with the possibility of theft. There are numerous different ways that a license could try to salvage the situation. But the BAYC license doesn't even try. To repeat myself, it is not fit for purpose. 37/