To hear enthusiasts for “Smart Contracts”, they are magic. The meaning of the contract is enshrined in code, and executed by computers. Cryptographic signatures and blockchain protocols assure that the contract is executed correctly and honestly. Once written, no human intervention is needed or, indeed, possible.
Entire businesses are created on this basis, so called Distributed Autonomous Organizations. Once created, these DAOs chunk along mechanically, executing business “autonomously”. No one disagrees about the results, mistakes and conflict are not possible.
This is better than magic. It’s the magic of capitalism raised to the power of magic!
In the very drafty basement of this castle in the air lies the claim that these executable contracts are not only always and completely correct, but also accurately and unambiguously express the intentions of the humans involved.
The former would be an historic first in the history of software, and the latter would be an historic first in the history of human thought.
You don’t have to take my word for it.
This month, the International Swaps and Derivatives Association (ISDA) issued a whitepaper, “Smart Contracts and Distributed Ledger – A Legal Perspective” 
The ISDA is a group that publishes standards for contract language for derivative contracts. These people define what “is” is, and what “means” means.
With all the nitter-natter about doing derivatives trading using executable “contracts” on a blockchain, the ISDA has taken up the question of just hos “contract-y” these so-called contracts may be.
The report is rather long and dry, and generally extremely well thought out.
The key point probably is:
“Certain operational clauses within legal contracts lend themselves to being automated. Other non-operational clauses – for instance, the governing law of a contract – are less susceptible to being expressed in machine-readable code. Some legal clauses are subjective or require interpretation, which also creates challenges.” (p. 3)
Basically, some “smart contracts” are simple bits of code that do something. But an actual derivative contract has a lot more in it that “operational” clauses, and you can’t leave them out. Furthermore, it’s those “non-operational” parts that are the subject of interpretation and dispute. Very few law suits are about account numbers or dollar amounts, they are always about whether and how rules apply.
The bottom line is that “smart contracts” will be subject to interpretation and dispute, period. The question is how to make them work well.
It is important to note that the ISDA report is talking about contracts in the legal sense of the word, an agreement recognized by law. While enthusiastic techies may imagine that they can declare their code to be outside any conventional legal system, it is generally the case that judges will decide what they have jurisdiction over. Code that isn’t recognized in a jurisdiction is probably not a contract in that jurisdiction, no matter how cunning it is.
Which means that the ISDA’s opinion is relevant, to say the least.
The “non-operational” language includes common phrases such as “good faith”, and “ordinary practice”. The report points out that these terms are intended to be subject to interpretation, if only because it is never possible to state all possible future conditions. They also point out that these terms may be interpreted differently by different authorities, which is why it is important to specify which authority will rule.
The report suggests hybrid contracts, part of which are machine executable, and part of which are interpreted by humans. This will require standardization of executable contract code, so the contracts will work everywhere. In short, the report concludes that ISDA has a critical role to play.
- International Swaps and Derivatives Association, Smart Contracts and Distributed Ledger – A Legal Perspective. 2017. http://www2.isda.org/attachment/OTU3MQ==/Smart%20Contracts%20and%20Distributed%20Ledger%20%20A%20Legal%20Perspective.pdf