Blockchain Legal Enforceability Triad
Every agreement requires at least two utterances, an offer and an acceptance. That is the minimum requirement for a ‘meeting of minds’.
It is not necessary that these utterances are directed towards a specific party, but they do need to have a known origin. Ex. I can make an offer to sell my Mars bar to anyone, but someone accepting it cannot answer _as_ humanity in general. They may answer _to_ humanity in general, since this includes me.
The utterances, from now on signals, need to be in an ordering with one another. Even if that’s as simple as a sequence number or a known date. It is not necessary to know the exact or approximate time, or a total ordering of all the signals, but we need to know offer and counter-offer, cause and effect.
And finally, the signals need to be reducible to some new common knowledge, eg. understanding, between the parties. Otherwise no meeting of minds. A meeting requires that there is some overlap of understanding due to a common interpretation of the signals exchanged.
So :-
- Origin
- Ordering
- Interpretation
We can take our framework out for a spin with some examples.
| Example | Origin | Ordering | Interpretation |
|---|---|---|---|
| Exchange of letters | Ink signatures | Postal date stamps | Common language and culture |
| FIX protocol share dealing | Secure sessions | Incremented sequence number | Simultaneous software implementation of FIX spec |
| EDI supply-chain automation | Secure telex lines | Time-stamping provided by a value-adding network operator [VAN] | UN/EDIFACT PASREQ – Travel, tourism and leisure product application status request |
| Oil trade | Yahoo email accounts | Yahoo Messenger’s history (was previously permanent) | Shared abbreviations carried over from telex messaging |
Notice that these three aspects are orthogonal. And there’s not really any difference between this and general group communication.
For any blockchain/DLT system, the origin is determined by digital signatures, the ordering is achieved by a byzantine fault-tolerant global consensus, and the interpretation is folding a reducing function over the ordering.
For example, in Bitcoin, item (1) is deal with using Elliptic-Curve digital signatures, item (2 by Nakamoto consensus (eg. the actual chain), and (3) by Bitcoin’s simple scripting mechanism.
In both Ethereum’s smart contract virtual machine, and in Bitcoin’s scripting language, the interpretation reduces all the signals, eg. transactions, associated with a particular resource to a single current state. Eg. if Alice’s balance is 47, it does not matter to the system’s future behaviour how this balance came about. In mathematics, and in software development, we call systems like this as having memorylessness.
Which brings us to the enforceability of on-chain contracts.
[quote, SEC, Report of Investigation Pursuant to Section 21(a) of the Securities Exchange Act of 1934: The DAO]
____
The automation of certain functions through this technology, “smart contracts,” or computer code, does not remove conduct from the purview of the U.S. federal securities laws.
____
I think the legal validity of a smart contract is relatively simple if we consider each of the three aspects separately. The question of origin is covered by legal frameworks like the EU eIDAS directive (itself based on the UNCITRAL Model Law on Electronic Signatures on which Ian Gregg made some comments that are fascinating in a post-Blockchain world). The question of ordering is analogous to the posting rule of acceptance for contracts. And the question of whether or not the interpretation is valid is similar to questions around whether bank balances, or mortgage payments, are correctly calculated.
We could have a pile of perfectly legitimate signed instructions, but they could not be actioned in a court if it is impossible to tell if, for example, a house purchase came before or after the sale, or even if some other event has invalidated the claim.
We might have a perfect sequence of title transfers, but no means to know who wrote them or if they were authorised to do so. Again that is not useful.
Or we could have a perfect sequence of signed noterised documents detailing h̻̟͎̼̯͔͎̒ͣͣͤͩͥi̼̹̮͖̙̠͑̅̀ͣ̒͐̎s̙͈͎̜̞͚̊̃̍̓̔ͧ͋ ̰̯͎̜̙̦͉ͪ̂̀̾̎̇n̳͍̹͓ͩ͑̒̓ͤ͌ͩ̀ͪa̯̖ͬ̂̓ͮ̿̈͆̑ͨ͌̈m͙̩͇̟̬̫̾̇̈̃̈́͌̉e̘̹̼̳͖͇̫͊ͤ̈́̐̔ͣ ̞͕̪͖͉̰͈͒ͬ̅ͭ͋ͬḫ̥͈̺͎͆ͤ͊̅̇͛̄ͥȋ̜̯̳͖̝̫̩͉̄ͥͯ́s̩̗̞̻͚͖̘̮͎̀͐̐̽ ̬̗̳̮̗̮̳͋̏͊͂͊ͫn̩͚̼̳̣ͭ̍̓͗̇͛̂ͅa͕̗̘͖ͧͣ͑̍̓͗̃͐̊m͍̘͍̬͍̯̏ͦ̉̀̍̇ͥê͓̭̟̯̮͇̑͌̾ͭ̓ͧ ̩͓̙̘̋̓ͣ̃̆ͭ͐̍̅h̤̟̙̲͈̽͆̌ͨ̉́ͤ́i̪͙̠̙̭ͨ̌̑͂͂̇̎ͫs͙͍̳͙̗ͫ͑ͩ͆̓͑̉ͩ ͚̠̱̗̼͕ͭ͊̒͌ͧ̏ͮn̮͔̪̩̜ͥͮ́ͮ͌͗̐̐a̳͉͇̹̝͍ͩ̾̅ͯͭͭͥm̙̮͙ͤͮ̂͐͆̈́̓ͭ̇̂ė̬̟̭͈͓̠̯̂̂ͤͩ̎ which is again not actionable since we do not understand what the followers of the Old Ones intended.
Since blockchains can cover this triad, and cover it to the satisfaction of multiple parties, they allow us to go beyond simply trusting the largest institution’s computer. Something that’s already caused numerous miscarriages of justice.
I doubt any of my thoughts here are unique, but I find this is a useful conversational frame for talking business people though why a blockchain(-ish) system is better for their compliance and legal needs than just using MySQL 🙂
If people are interested I could follow this up with a detailed article for each side of the triangle… Fun with eIDAS, distance contract rules, shrink-wrap licenses and more 🙂
P.S. I’m fine with reliable conventional time-stamping, it’s just not as flexible.