Do androids recognise electric sheep? South Africa's recognition of blockchain arbitration agreements

​​​​​​​​The enforceability of (foreign) blockchain arbitrations under South Africa's international arbitration regime might be difficult but there are possible solutions. As technology advances, a recurring tension is the ability of legal systems to cater for or incorporate such developments. Some aspects of technology may simply be unregulated without discrete legal development, whereas other aspects may fall – with varying degrees of ease – within existing frameworks.

With the advent of blockchain technology, the question arises whether South Africa's legal environment allows for the recognition and enforceability of foreign arbitration agreements that use this technology.

And, in the best tradition of lawyers, the answer is… perhaps.

At the outset, the intersection of blockchain technology with arbitrations covers expansive terrain. Blockchain technology, also referred to as distributed ledger technology, is (broadly) an immutable, digital database which uses coding to generate what is essentially a ledger. The digital ledger records and tracks transactions, generally in real time, and can be accessed according to the relevant permissions. Each transaction is recorded as a "block" of data in the ledger, linked to the preceding and successive blocks, creating a digital chain of interconnected, contemporaneously successive data blocks.

This digital ledger can intersect with arbitration agreements in several ways. At one end of the spectrum are self-executing smart contracts, which would theoretically never (practically) require recognition or indeed even human intervention. These smart contracts would automatically trigger enforcement events when an arbitral award is uploaded. By way of example, if monies (forming the subject matter of the dispute) are held by a facility or a bank which is party to the relevant blockchain, these funds could automatically be released once a digital instruction to do so is generated by a favourable arbitral award.

This also applies to the transfer of ownership where ownership is evidenced electronically. If proof of ownership existed in a digital space, this proof could automatically be updated once a final arbitral award to this end is uploaded (share certificates, deeds, registration of intellectual property, etc., if these were recorded electronically). Of course, this would require the relevant registries or facilitators to be party to the blockchain and accept the automated "smart" rules which would apply. In theory, the act of external enforcement could fall away as this becomes an automatic incidence of an arbitral award. This enforcement could span jurisdictions and legal regimes, provided that the relevant actors are party to the blockchain (and the electronic recording of ownership is recognised in the relevant jurisdiction).

We may be some way from a homogenous digital and legal environment which would cater for this, however.^1​

At the other end of the spectrum are arbitration agreements which still require human recognition and enforcement but have (with the relevant arbitral award) been recorded as data blocks in a digital ledger. Given the benefits of blockchain technology, including real time uploads, transparency of the chain of transactions, the permanent recording of each transaction, the chronological sequencing of data blocks which makes tampering difficult and readily identifiable, etc., the immediate reaction may be that there is no discernable reason why awards recorded only on blockchain technology should not be recognised. Alas, the law is not without complications.

South Africa enforces foreign arbitration awards through the International Arbitration Act, 15 of 2017 (the IA Act). The IA Act – which adopts the UNCITRAL Model Law on International Commercial Arbitration (as modified) - provides for recognition and enforcement of awards as required by the New York Convention on the Enforcement of Foreign Arbitral Awards, 1958, subject to some additional requirements.

The IA Act requires the original arbitral award and original arbitration agreement "authenticated in a manner in which foreign documents must be authenticated to enable them to be produced in any court", alternatively a certified copy of the award and agreement, to be produced for enforcement proceedings. A certified copy, however, means "a copy authenticated in a manner in which foreign documents must be authenticated to enable them to be produced in any court."

Under South Africa's Uniform Rules governing court processes, Rule 63 provides for the authentication of foreign documents (excluding affidavits), with authentication serving the purpose of verifying signatures on them. Broadly, this requires the signatory to appear before a specified officer, identify him- or herself and sign the relevant document before the officer, who then appends a certificate or the like confirming the fact of signature by the identified signatory. (As an aside, it is unclear how a copy could ever be authenticated.)

Although Rule 63(4) does provide for a "catch-all" provision where a court "may accept as sufficiently authenticated any document which is shown to the satisfaction of such court or the office, to have been actually signed by the person purporting to have signed such document", this does require some form of signature.

A potential difficulty thus arises. If the original arbitral agreement or award is generated in a virtual environment on the blockchain, it may not have signatures at all or in the traditional sense that are capable of authentication. Moreover, it would be rare for an arbitrator / arbitral panel to sign an award before an officer nominated under Rule 63, solely for purposes of external confirmation of who signed the award. So apart from blockchain complications, even traditionally signed arbitral awards may face difficulties when it comes to authentication. Several potential solutions could be devised within the existing framework to cater for this difficulty regarding signature and authentication:

• the Electronic Communications and Transactions Act, 25 of 2002 (ECTA), provides for the use of electronic and advanced electronic signatures, which may suffice to meet Rule 63(4), supplemented by additional evidence, if necessary (although where an arbitration agreement is silent on the signature requirements, statutorily only the more securely digitally signed method of signature would be accepted. In other words, it is likely that the court would only accept an advanced electronic signature which meets the requirements of ECTA);
• if authentication is impossible, given the nature of the award or agreement, then it could be argued that Rule 63 (and the IA authentication / certification requirement) could never be met, permitting the Court to adopt a purposive approach to ensure that the safeguards performed by these requirements are satisfied. It may be a stretch, however, to argue that the maxim "lex non cogit ad impossibilia" truly applies, given that the parties were free to sign the agreement in the traditional manner;
• if the parties have agreed to the award and the agreement being located in a virtual environment, then doctrines of waiver and estoppel may prevent the losing party objecting to recognition and enforcement on the basis discussed above. This does not, however, solve the difficulty that the Court would still itself have to apply the IA Act, even without any opposition.

Parties seeking to have foreign arbitral awards recognised in South Africa may first want to have the underlying agreements signed traditionally, before an officer with authentication powers, before uploading it to the blockchain. That would render the storage location irrelevant.

Ultimately, the IA Act has cumbersome requirements which may not have been stress-tested practically. It remains to be seen how these will be applied by the Courts when faced with the prospect of recognising and enforcing a foreign arbitral award located on the blockchain. Clear legislative development to address this potential issue would be welcomed. It is hoped that our courts will adapt to ensure that foreign arbitral awards making use of new technologies are not defeated by formalistic points raised by opportunistic defendants.

Although it was stated years ago, in the context of jurisdictional challenges, the following still rings true today:

"We live in a time of rapidly growing commercial and financial sophistication and it behoves the courts to adapt their practices to meet the current wiles of those defendants who are prepared to devote as much energy to making themselves immune to the courts’ orders as to resisting the making of such orders on the merits of their case." ² 

​

Read the original publication at Webber Wentzel.