(c) Philipp Paech 2017
Full version with further references see (2017) 80(6)MLR1073–1110
One of the original traits at the basis of blockchain-based networks is that there is no trusted third person to effect and record transactions between nodes (see the relevant earlier post). In the world of the blockchain, trust, which in the ‘real’ world is typically afforded to public authorities (such as the land register) or certain private parties (such as a bank or a notary), is replaced by reliance on software. For the blockchain nodes to be able to rely on their network’s software, they must be convinced of its soundness, ie they must be confident that it allocates rights according to internal rules to which they agreed upon joining the network. At the same time, this principle entails that the process, once initiated, must be resistant to alteration and beyond human control. Otherwise, again, parties would need to trust the person controlling the process. As a logical consequence, the allocation of rights in blockchain financial networks must be unstoppable and irreversible (as discussed here).
However, the idea of such unstoppable and irreversible allocation in practice of individual rights to users creates tensions not only with the regulatory regime (see the various posts on regulatory issues) but also with the private law framework.
- In this post, I will discuss the first complexity caused by the above tension: the parties as well as the competent courts and regulators will lose authority over the enforceability of transactions or smart contracts once they were recorded on the blockchain.
- A separate post addresses the effects outside the parties’ relationship: the parties’ (consensual) dealings may cause adverse externalities with regard to unrelated, third parties and the market as a whole, notably in case of insolvency (see here).
- A third post turns towards a potential solution of these difficulties: the authority to attribute rights in blockchain financial systems must ultimately be connected to the private law order through the key concept of ‘finality’ (see here).
Absolute authority of the ‘internal rules’
A key component of blockchain is that the process of disposition and acquisition of assets and the execution of smart contracts is determined solely by the internal rules. These rules are the result of the consensus typically established when nodes join the network and thereby adhere to the rules determining the acquisition and disposition of assets and the execution of smart contracts on the network. These rules are laid down directly in the form of a computer code. There are no ‘bylaws’ or similar documents in human language. The algorithms directly produce the relevant effects.
In this process, the rules are constantly called upon to ‘decide’ whether or not a certain transfer will be executed or whether the right of a party arising under a smart contract will be automatically enforced. However, in taking this ‘decision’, the software typically attributes rights to one party that it takes away from the other party. For instance, a smart derivatives contract might include a functionality causing it to terminate itself upon default or other types of termination event, automatically calculating and enforcing the amount still due from one party to the other, while at the same time transferring the associated collateral to the party that is ‘in the money’.
The situation in un-permissioned, anonymous networks
Where, in my example, the other party feels that the conditions for termination were not actually met it may decide to go to court, as may any other party in case it feels its rights have been overridden by a change of the blockchain record. However, where transactions are near-anonymous, as they are in first-generation blockchain applications such as Bitcoin, the story typically ends here as there is no way de facto of suing the other party for damages in kind or in money.
Still, the parties may have previously agreed to an internal dispute settlement mechanism (which does exist, for instance, for online acquisitions paid with bitcoins). This mechanism applies the rules of the system, but not general private law, and necessarily results in an outcome compatible with the logic of blockchain networks—which is built on the logic that validated transactions and the execution of smart contracts cannot be undone on the record.
The situation in permissioned networks
Even in permissioned systems, where the identity of users is known, court decisions do not exert the same authority as in the traditional context of financial market transactions. Should a party claim that a transaction or smart contract that was executed under the internal rules of the network was unenforceable, the court will first consider whether the parties have agreed to the application of the internal rules to their dealings, as an expression of party autonomy. In that case, the code, or rather how it is understood in human language, would be the law, and every subsequent transaction would occur in accordance with it. In other cases, the court may hold that as a matter of general private law the transaction was unenforceable, eg because there was no valid agreement on the internal rules, or in case of a lacuna.
Rectifying the transaction record or ordering damages?
However, even presupposing that there was a trusted entity controlling the network (as will be most probably the case with all third generation applications, in particular in finance) to whom the relevant court order could be addressed, the court will still be unable to order a rectification of the blockchain, as the record cannot be changed subsequently without destroying the logic of the trust-less network itself.
Hence, the only remaining possibility is to claim damages from the transferee, in particular in kind. In that case, the court would order the initiation of a new, reverse transaction. The other possibility is to order damages to be paid in money. However, as is the case in traditional financial markets, claiming damages will often frustrate the transferor whose interests were overridden. The most relevant case is that of the insolvency of the original transferee. Where insolvency has kicked in between the point in time of the original transaction and the order to pay damages, the original transferor will be in a very weak position and basically referred to the quota. It would be much more favourable for the original transferor to have an interest in the specific assets which had been invalidly transferred. Yet, as opposed to the traditional environment, where registers can be ‘incorrect’ (ie the appearance of the register does not reflect the true legal attribution of rights) it is impossible, in a blockchain environment to correct the register and make it reflect legal reality. It is also impossible to operationally reverse transactions in certain cases. As a consequence, the blockchain environment offers damages as the only remedy. This is the within both types of networds, unpermissioned and permissioned ones. The logic in both cases invariably subjects the original transferor to the risk of insolvency of the transferee.
Adjusting contractual rights and obligations?
This issue is also relevant in relation to transactions, in particular, smart contracts that are still open, where one party claims that the contractual duties should be adapted in response to new circumstances not previously considered by the parties, ie in case of a lacuna. In the first-generation blockchain setting, there is no way of changing the record, and thereby the contract, even in cases where both parties agree to the change.
To revert to our earlier example: in the event of default in the context of a derivatives contract, the non-defaulting party too may often prefer not to terminate the contract and instead choose implicitly or expressly to adjust it, in particular by granting a grace period. Again, for lack of a trusted entity with the authority to change the record according to the parties’ agreement, the terms of the smart contract cannot be changed and its execution cannot be halted.
A subsequent ‘reversal’ of the termination, by entering into a new contract, as a form of damages in kind, will often not be possible. This is because the economic circumstances have typically changed in the meantime, in particular where one of the parties has become insolvent or where market conditions have undergone considerable change.
The limited value of ex-post review
Thus, any kind of ex-post review is limited to the potentially unsatisfactory possibility of claiming damages in court. Here, blockchain-held assets differ markedly from assets held in more traditional, account-based structures. Current financial market infrastructures, such as clearing and settlement systems, also use computer programmes to prioritise their users’ interests on the basis of their internal rules, as described above. In the EU, this principle is enshrined in the Settlement Finality Directive. However, outcomes can still be changed by the infrastructure operator, honouring the agreement of the parties or court orders, or simply correcting operational failures.
The increased precision of the programming language in blockchain environments may remove some of the potential need for ex-post review, as the internal rules and in particular warranties and conditions can be formulated with much greater accuracy. However, while this may indeed remove linguistic ambiguity, the greater precision is of little help in relation to issues such as changing circumstances, lacunae or even, depending on the applicable law, questions of equity or good faith.
Rather, these issues could be addressed by leaving certain parts of the agreement outside the blockchain record as a ‘non-smart’ and thus modifiable contract, whereas other parts might ‘go smart’ and be self-executory and immutable. Thereby, some flexibility would be allowed into the relevant agreement (which is the strategy envisaged, for example, for derivatives governed by ISDA master agreements).
Alternatively, the code of the smart contract could become more granular, in an attempt to address all potential future circumstances that may have an impact on the contractual duties. In the ideal case, no flexibility would ever be needed—an approach which, of course, may come close to, but ultimately will never achieve, perfection.
The above findings are, in principle, acceptable in so far as the parties to a blockchain-based transaction and other users of that blockchain network are concerned. By adhering to the network, they have, implicitly or explicitly, agreed to operate in a technical, trustless environment, which only relies on maths and cryptography. At the same time, they have accepted that the internal rules alone determine the outcomes of their dealings. The limited ability of courts to adjudicate disputes and enforce rights between them can be seen as an expression of their party autonomy.
However, while the parties themselves are not obliged to consider the rights of third parties and the market at large, the law and the courts are concerned with exaclty that. The adverse externalities of the parties using a blockchain set up on these third-party actors will be the subject of a separate analysis.