Matt Levine at Bloomberg has some fascinating commentary on the idea advanced by some crypto advocates that the blockchain code is the only thing that governs, binds, and controls crypto transactions. On this view, if the code allows you do something, then it is allowed, and if it doesn't, then it isn't allowed, because the code itself defines what is and what is not allowed. On this view, any other text, contract or document is irrelevant, and has no meaningful impact on whether a particular transaction or event is permissible. So if the terms and conditions or the marketing materials on the website conflict with the code, the code rules. This view is heavily grounded in the view that crypto falls outside the reach of government regulation, an idea that is... well, let's say it is not universally accepted. My personal view is that most crypto transactions within the US and the EU are already subject to some fairly significant regulations, or will be in the near future. The absolutist view of the function of the blockchain code is not likely to survive over the long term. But the concept is interesting, and weaker versions of it exist in modern legal reasoning. If you have a written contract with someone that says that you will pay them your rent in US dollars on the 8th day of each month, but what you actually do is pay them in Swiss francs on the 11th day of each month, and you do this consistently for a long time, and your counterparty accepts the payment every time, many legal scholars will say that the behavior of the parties has effectively modified the terms of the agreement. And this argument may be accepted by a court even in cases where the written contract explicitly states that any modification must be in writing. What you actually do, in some contexts and in some cases, may be more important than the written instrument. Courts will not simply ignore a long history of consensual behavior by both parties that conflicts with the terms of a written agreement. Someone could try to argue that a smart contract on the blockchain, by definition, is both the contract and what the parties actually do, and that any other text or document has no effect. This argument amounts to an assertion that a smart contract, by its very design, makes it impossible for the parties to do anything that conflicts with the terms of the agreement, and also forces the parties to do whatever is required by the agreement, because the contract itself is the behavior. This sounds brilliant. But in the real world, mistakes happen. If there is strong evidence that a smart contract is not an accurate reflection of the intent of the parties, a court may look at other things to determine the true intent. This reasoning can become a bit circular. A true believer will argue that the smart contract is the only evidence of the parties' intent, because that's how they work, and that both parties understood that at the time the entered into the agreement. But that's not all that different than pointing to a paragraph in a traditional "analog" contract that says that in can only be modified in writing. The brings us back to the question of whether a smart contract is something more than a contract, i.e., is it also a behavior? Anyway, here's a few interesting articles. This is a column by Matt Levine from June 2016 in which he talks about the DAO hack and some other cases involving similar legal questions. Yes, it's eight years old, but still interesting: https://tinyurl.com/170617BBLevine This one by Levine from September 2024 talks about someone who exploited the code and algorithms at Spotify. The case raises similar legal questions: https://tinyurl.com/240909BBLevine And in this one, published today, Levine talks about a case involving warrants issued by a company to its employees. They are essentially a customized type of call option, and the terms were originally meant to include an adjustment in the strike price associated with dividends. Somehow the final version of certain key legal documents did not include that provision. Employees are now arguing that it is the intent of the parties that really matters, and not the final version of a lengthy document that was never actually read by anyone except a handful of attorneys. For this one, he starts out talking Microstrategy and an unrelated topic. His columns have different sections. One section is titled "Code is law," followed by a section called "Dividend adjustment." https://tinyurl.com/241031BBLevine These are gift links that expire in seven days. And here's a fun paper published by Yale Law School almost two years ago, titled "The Death of Cryptocurrency: The Case for Regulation." https://law.yale.edu/sites/default/...ents/weaver_death_of_cryptocurrency_final.pdf
That's my view. There's little if anything to actually regulate because the rules set forth by the code and the repeated actions taken by those who transact on that code layer have already "laid down the law" so to speak.
as long as the developer produces no bugs and is not malicious all is good. but in the real world ...
All of that. Plus, there is a non-zero possibility that the original specification doesn't really cover the actual use case. Code: A QA engineer walks into a bar. Orders a beer. Orders 0 beers. Orders 99999999999 beers. Orders a lizard. Orders -1 beers. Orders a ueicbksjdhd. First real customer walks in and asks where the bathroom is. The bar bursts into flames, killing everyone.