Choice-of-law Problems with Blockchain-based Negotiable Instruments

I have written a paper on choice-of-law problems with blockchain-based negotiable instruments, with particular reference to bills of lading and investment securities. In this paper, I have examined the emerging substantive rules as well as choice-of-law rules of some legal systems and made suggestions for choice-of-law rules. The information is current as of August 2021. This paper is to be published as a chapter from Andrea Bonomi and Matthias Lehmann (eds) Blockchain & Private International Law (Brill). The text, subject to formatting and copy-editing, may be obtained from https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3937664.

Presentation at the conference “UNCITRAL Asia Pacific Day" (4 December 2020 at the University of Macau)

I gave an online presentation on blockchain-based bills of lading under the UNCITRAL Model Law on Electronic Transferable Records in Session III of the conference.

I attach here the slides and the transcript of my presentation as well as the conference timetable. Preparing for this presentation, I built on my Japanese paper given at a symposium in Tokyo one and a half year earlier and published subsequently. This conference was a fitting occasion to convey my thoughts in English and take stock of the latest developments in actual practice.

Dematerialization of Negotiable Instruments

My article, "Legal Issues Arising From the Use of Blockchains for the Dematerialization of Negotiable Instruments: with a Particular Focus on Bills of Lading and the UNCITRAL Model Law" has just been published. It is in (2020) 5 Yearbook of the Japanese Association of International Business Law 24-39 and is attached to this post.

I wrote this article on the basis of a paper I gave one year ago at the symposium "UNCITRAL Model Law, Cryptocurrency and Blockchain" (Waseda University, 16 March 2019). As the presentation was made in Japanese, this article, too, was written in Japanese. The accompanying English abstract reads as follows:

This article considers the legal challenges which will be encountered when blockchains are used to dematerialize negotiable instruments such as bills of lading. The digitization of negotiable instruments yields a lot of benefits to the society but has been hampered by various technological and legal obstacles. On the technological side, this article examines the advantages and drawbacks of the blockchain as a tool for dematerializing negotiable instruments. On the legal side, there is a lot of uncertainty over the permissibility of, and the legal requisites for, dematerializing negotiable instruments. To improve certainty, the UNCITRAL created the Model Law on Electronic Transferable Records in 2017. The Model Law lays down the attributes which an electronic record needs to possess before it is deemed to be functionally equivalent to the corresponding "transferable instrument," a term broadly synonymous with "negotiable instrument." Thus, the electronic record must, by virtue of a reliable method, be identifiable (which implies the consistency of the record) and amenable to exclusive control. This article considers whether blockchain-based electronic records have these attributes. Among such attributes, the reliability of the method poses a particular challenge to the current law since the latter is not accustomed to evaluate the reliability of blockchains. Prior to the emergence of the blockchain technology, electronic records could only be maintained by an administrator of the database. It follows that under the conventional approach, the reliability of the method would be ensured through the regulation and oversight of the administrator. The blockchain, on the other hand, is a trustless technology which is founded on the notion that a system dispensing with the need for administrators is reliable for the very reason that it is distrustful. Whether the law is ready to embrace this notion will be tested if the society is to harness the full potential of the blockchain.

Symposium on MLETR and blockchains

I gave a paper in a symposium on the UNCITRAL MLETR (Model Law on Electronic Transferable Records) and blockchains (Here is link to the program) on 16 March 2019 at the Waseda University in Tokyo.
The proceedings were conducted in Japanese except the presentation by Luca Castellani, the legal officer of the UNCITRAL Secretariat who was responsible for this Model Law. So my powerpoint slides and handout (below) are also in Japanese.
I have underscored the potential of public blockchains for disrupting the society and considered whether they meet the requirements of the MLETR. Throughout the analysis, I have compared public blochains with permissioned blockchains and central registries. Each of the requirements of the MLETR poses an analytical challenge with respect to public blockchains.
I will try to write an English version of the paper once I have cleared a backlog of work on my desk.

Bahraini legislation based on the UNCITRAL MLETR

(This entry was originally posted at https://wordpress.com/view/blockchaincryptolaw.wordpress.com on 12 February 2019.)

Bahrain became the first to enact a statute based on the UNCITRAL Model Law on Electronic Transferable Records. I have been provided with an English translation of the statute by Jameel Al Alawi, Senior Legal Adviser for the Bahrain Economic Development Board, who was in charge of drafting the statute. With his permission, I post it below.

The Model Law sets out the conditions which must be met for an electronic record to be treated as a "transferable document" (Article 10). The latter is defined as a document that entitles the holder to claim the performance of the obligation indicated in the document and to transfer the right to performance by means of the transfer of that document (Article 2). Bills of lading and warehouse receipts, for example, are covered.

The Model Law adheres to the principle of technology neutrality, which means that the law should neither require nor assume the use of a particular technology for communicating or storing information electronically. Thus, the blockchain technology is not excluded from the Model Law's scope of application. It is in fact a technology well suited for creating and managing electronic records which purport to replicate transferable documents because it is capable of guaranteeing that there is a single true version of electronic records. 

The Model Law requires the use of a reliable method to establish an exclusive control of an electronic record that replicates a transferable document (Articles 10(1)(b)(i)(ii) and 11(1)(a)). In my previous work, I noted:

The reliability of the above-mentioned methods will be assessed by adjudicators on an ex post (i.e. after the occurrence of a dispute) basis. It would, however, be unfortunate if there were no foreseeability as to which methods would pass the reliability test since the use of such methods would then be deterred. A thought should, therefore, be given to the possibility of compiling a list of reliable methods on an ex ante basis. Such a list would need to be reviewed from time to time because neither the configuration of a central registry nor the algorithm of a blockchain is permanently fixed.

The Model Law lists a number of circumstances by reference to which to evaluate the reliability of a method, including the existence of a declaration by an accreditation body (Article 12(a)(vi)). But it leaves the details to the national laws.

What is interesting about the Bahraini legislation is that it provides for the accreditation of an "operator", the latter being defined as a person who operates an information system for managing electronic transferable records (Article 1(l)). It sets forth the procedure and conditions for accreditation (Articles 15 and 16), though it delegates to the competent authority to lay out the details of the conditions by means of a regulation. It also provides for the withdrawal of an accreditation. Once an operator is accredited, the reliability of the method used by the operator is to be presumed unless evidence to the contrary is adduced (Article 8(2)). Furthermore, where reliance on an electronic transferable record has caused damage and the electronic record is managed by an accredited operator, it is to be presumed that the damage was due to the operator’s intention or negligence unless otherwise proven (Article 17).

The Bahraini legislation is applicable to electronic transferable records "whether or not an operator is used in respect of these records" (Article 2(1)). Accordingly, it seems applicable to electronic transferable records managed with the use of the blockchain technology. In view of the definition of an "operator" (Article 1(l)), it seems unlikely that any accreditation will be issued with respect to public blockchains. But the administrator of a private blockchain may fall within that definition. It will be interesting to see whether the conditions for accreditation issued by the competent authority will actually cover private blockchains as well as central registries.

Postscript (15 Feb. 2019): Jameel has informed me that the Bahraini statute entered into force on 1 February 2019 but the regulation is still being debated.

KLRI/UNCITRAL Trade Law Forum

I was invited to speak at the KLRI/UNCITRAL Trade Law Forum 2018 (Incheon in South Korea on 10-12 September 2018. For the programme, see a link here). The KLRI, Korea Legislation Research Institute, is a government-financed think tank.
My topic was "Cryptocurrencies entrusted to an exchange provider: Shielded from the provider’s bankruptcy?" I attach my powerpoint file below.
Very many thanks to the organisers for invitation and to the audience for stimulating questions.

"Implications of the Blockchain Technology for the UNCITRAL Works" - published.

It has come to my attention that my article "Implications of the Blockchain Technology for the UNCITRAL Works" had been published from the United Nations in November last year.
The 50th anniversary Congress was a big occasion for UNCITRAL and it was my great honour to be part of it.
Here is my article excerpted from the book.


An additional note: In my original manuscript, there were some references to specific chapter numbers. I have noticed that in the published version, they have been changed to "Ch. 0" due to the editorial work which has removed all the chapter numbers from the headings. To see where those were actually referring to, please consult my original manuscript here.

Implications of the Blockchain Technology for the UNCITRAL Works

I will be presenting my thoughts on the subject above in the upcoming Congress of the UNCITRAL for the celebration of its 50th anniversary (4-6 July 2017).

My paper currently on the Congress website is a version which I sent to the UNCITRAL Secretariat some months ago and which no longer represents my latest thinking in some significant respects. I am asking the Secretariat to replace it with the latest version, to which I make a link from here. 

The paper gives a particular emphasis on the topic of proprietary restitution of blockchain-based tokens as an area which calls for a globally unified solution.

Postscript: As from 16 June, the Congress website carries the latest version. Many thanks to the Secretariat for swiftly acting on my request. 

Relevance of the blockchain technology to the draft Model Law on Electronic Transferable Recores (as acknowledged by the latest official document)

My article, "Blockchain Technology and Electronic Bills of Lading", has examined the draft Model Law on Electronic Transferable Records and the Rotterdam Rules to see whether it is possible to interpret them in a manner compatible with blockchain-based bills of lading. What follows will note how the relevance of the blockchain technology has come to be acknowledged in the latest official document (A/CN.9/920 (2017)) which contains the draft Model Law with the draft explanatory notes.

The draft Model Law accommodates various types of electronic transferable records based on the principle of technology neutrality. The draft explanatory notes explain that reference in the Model Law to electronic transferable record management systems does not imply the existence of a system administrator or other form of centralized control (Para. 167). 

Nothing in the draft Model Law requires a person to use an electronic transferable record without that person’s consent (draft Article 6(2)). The draft explanatory notes state that consent to using distributed ledger based systems may be inferred by circumstances such as the exercise of control on the electronic transferable record or performance of the obligation contained in the electronic transferable record (Para. 48).

The draft Model Law is based on the principle of functional equivalence. Thus, where the applicable law provides for the endorsement of a transferable document, the draft Model Law treats an electronic transferable record as functionally equivalent to a transferable document only if the information required for the endorsement is “included in” the electronic transferable record (draft Article 16 on endorsement). The draft explanatory notes state that the words “included in” have been chosen to encompass instances when the information is logically associated with or otherwise linked to the electronic transferable record (Para. 141). This wide interpretation would accommodate the endorsement of a blockchain-based token through its transfer from one address to another on the blockchain.

Where there is a legal requirement of a signature of a person, an electronic transferable record can meet that requirement only if a reliable method is used to identify that person (draft Article 9 on signature). The draft explanatory notes acknowledge that certain electronic transferable records management systems, such as those based on distributed ledgers, may identify a signatory by referring to a pseudonym rather than a real name (Para. 60). The notes suggest that an identification by a pseudonym and the possibility of linking it to a real name, if need be, would satisfy the requirement to identify a signatory (Para. 60). The explanatory notes further suggest that linking of a pseudonym to a real name may be based on factual elements to be found outside distributed ledger systems (Para. 60). The same interpretation may be given to the notion of "identification" of the person in exclusive control of an electronic transferable record, a requirement which must be met to establish functional equivalence to the possession of a transferable document (draft Article 11(1) on control).

The relevance of the blockchain technology for the UNCITRAL Model Law on Electronic Commerce (1996), the UNCITRAL Model Law on Electronic Signatures (2001) and the UN Convention on the Use of Electronic Communications in International Contracts (2005)

One of the principles guiding UNCITRAL in its works in electronic commerce is the principle of technology neutrality or technological neutrality,[1] which means that the law should neither require nor assume the use of a particular technology for communicating or storing information electronically. The principle helps ensure that the law is able to accommodate future developments. Thus, blockchain technology, though not yet invented when those instruments were created, is not excluded from their scope of application.

It follows that under the Electronic Commerce Model Law, admissibility in evidence or other legal effect may not be denied to information solely on the ground that it is in the form of a data message stored in a blockchain (See Articles 5 and 9). Where only a hash value of information is stored in a blockchain, legal validity may not be denied to the information provided that it is referred to in the data message stored in the blockchain (See Article 5 bis added in 1998). In the context of contracts, an offer and the acceptance of an offer may be expressed by means of data messages on a blockchain (See Article 11 as affirmed by Article 8 of the Electronic Communications Convention). The performance of contractual obligations are also subject to the Model Law and the Convention. This has a particular relevance to a “smart contract,” a contract which can be automatically executed on a blockchain as programmed. Thus, where the communication of data messages such as notices of receipt of services, notices of failure to perform or notices of termination are programmed by a smart contract, they may not be denied legal effect solely on the ground that it is in the form of a data message (See Article 12 of the Model Law; Article 4(a) and Article 8 of the Convention). Article 12 of the Convention only mentions the formation of contract but the enacting States may wish to extend the idea to the phase of performance to cater for a smart contract. Thus, they may stipulate that the performance of a contract by an automated system may not be denied effect on the sole ground that no natural person intervened in each of the individual actions carried out by the automated system.

The principle of technological neutrality does not mean that any technology can create a data message which satisfies the paper-based requirements such as those of writing and signature. Only the technology capable of fulfilling the purposes and functions of the paper-based requirements can create a data message which is deemed to meet those requirements. This is called the principle of functional equivalence, another principle underlying the UNCITRAL works in electronic commerce. Thus, the Electronic Commerce Model Law sets out the conditions which a data message must meet to fulfill the purposes and functions of the paper-based requirements of writing and signature (Articles 6 and 7). The Electronic Signature Model Law elaborates on the conditions for the signature requirement. A data message stored in a blockchain is deemed to meet the requirements of writing and signature if it satisfies the respective conditions. The Electronic Commerce Model Law also provides that there must exist a reliable assurance as to the integrity of information contained in a data message before the information is deemed to satisfy the paper-based requirement that it be presented in its original form (Article 8). The blockchain technology is particularly apt to provide a reliable assurance as to the integrity of information since it is censorship resistant.

[1] See the Guide to Enactment of the Electronic Signatures Model Law (2001) para 5; the preamble of the Electronic Communications Convention. In the context of the Electronic Commerce Model Law, the expression “media-neutral” is used to convey the same idea (See the Guide to Enactment of the Electronic Commerce Model Law (1996) para 24). Only later, has that expression come to be understood as referring to non-discrimination between paper and electronic media (See the Guide to Enactment of the Electronic Signatures Model Law (2001) para 5).

Security rights in cryptocurrency-related assets

Any asset having market value will generate demand for use as a collateral. 

It would be possible to create security rights in a receivable which is denominated in a cryptocurrency in the same way as a receivable denominated in a fiat currency. 

The UNCITRAL Model Law on Secured Transactions (2016) sets forth a special set of rules for bank deposits (Article 25 on effectiveness against third parties and Article 47 on priority). Are they also applicable to a receivable denominated in a cryptocurrency? If the words “authorized deposit taking institution” as contained in Article 2(c) of the Model Law which defines the term “bank account” are interpreted as a generic expression covering any institution authorized to receive deposits, a company which is authorized to provide an online wallet service for the Bitcoin would be an “authorized deposit taking institution.” This interpretation, if adopted, would attract the application of the rules of the Model Law for bank deposits.

Then, would it be possible to create security rights in the units of cryptocurrency? This question concerns cryptocurrency units themselves rather than a receivable denominated in a cryptocurrency. Under the Model Law, any type of movable asset may be encumbered by a security asset (Article 8(a)) and the words “movable asset” is defined broadly as a tangible or intangible asset, other than immovable property (Article 2(u)). So the Model Law seems applicable to cryptocurrency units. 

There are, however, questions of interpretation as to how its rules are to be applied. Thus, a security right in cryptocurrency units would be effective against third parties under Article 18(2) where the secured creditor holds them at his address if it is possible to interpret the possession of a private key as being equivalent to the possession of a tangible asset.

Under the Model Law, the word “money” is defined as currency authorized as legal tender by a State (Article 2(t)). A cryptocurrency would be capable of meeting this definition if any State authorized it as its legal tender. However, money is supposed to be a tangible asset under the Model Law (See Article 2(ll)). Consequently, the special rules for preserving negotiability of money contained in the Model Law (Article 48 on priority) are not applicable to cryptocurrencies. Should it be thought that cryptocurrencies ought to benefit from the same rules, amendments would be needed.

Finally, the Model Law recognizes that the enacting State may specify certain types of asset as being subject to specialized secured transactions under other law (Article 1 (3)(e)) but it recommends such exceptional regimes to be limited (footnote 3).

Blockchain Technology and Electronic Bills of Lading

My article "Blockchain Technology and Electronic Bills of Lading" has come out from the Journal of International Maritime Law ((2016) 22 JIML 202-211). The full text is here.

Blockchain Technology and Electronic Transferable Records

Here is the powerpoint file for my presentation at the seminar "Electronisation of Transferable Documents or Instruments Used in International Trade" (10-11 March 2016) in Singapore (organised by UNCITRAL, Attorney-General’s Chambers of Singapore and the Association of Banks in Singapore (ABS)).

Blockchain-based electronic transferable record's functional equivalence to endorsement of paper-based transferable documents

A chain of transactions of an electronic record on a blockchain looks like a chain of endorsements on a paper-based transferable document except that the transactions are anonymous. Despite the anonymity, the blockchain technology ensures a far greater security than bearer documents. I think, therefore, that the law should be crafted in such a way as to allow documents made to order (such as order bills of lading) as well as bearer documents (such as bearer bills of lading) to be replaced by blockchain-based electronic records.

From that point of view, this post will examine the draft text of the UNCITRAL Model Law on Electronic Transferable Records. I have noted in my earlier post that the work on the Model Law should be pursued with the blockchain technology in mind so as to facilitate its applications to replace paper-based transferable documents. As of the time of writing, the most recent official documents containing a draft text are A/CN.9/WG.IV/WP.135 and A/CN.9/WG.IV/WP.135/Add.1, both dated August 2015. The draft provisions quoted below are taken from them.

Draft article 19. Endorsement
Where the law requires or permits the endorsement in any form of a paper-based transferable document or instrument, that requirement is met with respect to an electronic transferable record if information [relating to the endorsement] [constituting endorsement] [indicating the intention to endorse] is [logically associated or otherwise linked to] [included in] that electronic
transferable record and that information is compliant with the requirements set forth in articles 8 and 9.

While an electronic record on a registry may include information constituting endorsement, in the case of an electronic record on a blockchain, it is the blockchain (or, to be more precise, a chain of transactions contained therein) rather than the record itself which indicates intentions of endorsement. Accordingly, the expressions "indicating the intention to endorse" and "logically associated with" seem better cater for the blockchain technology. The words “logically associated” was indeed retained "to provide for all possible instances and methods for the incorporation of an endorsement in an electronic transferable record." (A/CN.9/828, para. 80).

A more difficult question relates to compliance with the requirement set forth in Article 9.

Draft article 9. Signature
Where the law requires a signature of a person, that requirement is met [with respect to] [in relation to] [by] an electronic transferable record if:
(a) A method is used to identify that person and to indicate that person’s intention in respect of the information contained in the electronic record; and
(b) The method used is either ...

This provision will not pose problem with respect to a blockchain-based electronic record to the extent it applies to signatures to be included in the record itself, e.g. a "for the master" signature on a bill of lading. 

It is also intended, as indicated by Draft Article 19, to cover signatures for endorsements. It would be natural to impose a signature requirement to ensure functional equivalence to endorsement, especially if we envisage a registry-based electronic transferable record.  But the requirement of identification, if it is read as requiring identification by name, would be incompatible with electronic records on an open, permissionless blockchain since the parties are anonymous. Logistically, it seems possible to build a system whereby the performance of the obligation represented by the electronic record (e.g. the delivery of goods) is effected without the name of the holder of the private key having to be revealed to the obligor (e.g. carrier) by, for example, allowing the record to activate a key in the physical world (e.g. the key to the container). In my earlier post, I have suggested that a blockchain-based bill of lading may, for that reason, be seen as functionally equivalent to a bearer bill of lading. But the present post is considering how to make rules enabling the replacement of order bills of lading (and other documents made to order) with a blockchain-based electronic records. 

The word "identify" also appears in the following provision.

Draft article 17. [Possession] [Control]
1. Where the law requires the possession of a paper-based transferable document or instrument, that requirement is met with respect to an electronic transferable record if:
(a) A method is used to establish exclusive control of that electronic transferable record by a person and to reliably [identify] [establish] that person as the person in control; and
(b) The method used is either ...

The Secretariat notes, "identification should not be understood as implying an obligation to name the person in control" (para. 22) (See my earlier post for a comment). It would aid clarity to say so expressly in the provision. The same could be done in draft Article 19 and it would make it friendlier to blockchain-based electronic records. But it might not be satisfactory for register-based electronic records. That is why I call this a "more difficult question."

Different functions of electronic transferable records: UNCITRAL's definition

There are no fewer than three different functions which electronic records transferable on a blockchain may perform with the necessary support of applicable laws. The Model Law on Electronic Transferable Records currently drafted by UNCITRAL is only applicable to a type of electronic transferable record which fulfills one of them.
According to the latest draft published in an official document (A/CN.9/WG.IV/WP.135), for the purpose of the Model Law, an “electronic transferable record” [is an electronic record that contains all of the information that would make a paper-based transferable document or instrument effective and ... "(Draft Article 3). The same article defines “paper-based transferable document or instrument” as "a transferable document or instrument issued on paper that entitles the holder to claim the performance of the obligation [indicated] in the document or instrument and ...". It will be possible to use a blockchain to transfer an electronic record which contains all of the information that would make a paper-based bill of lading, for example, effective. Such a record would would constitute an "electronic transferable record" within the meaning of the draft Model Law.
Cryptocurrencies, on the other hand, do not represent entitlement to claim the performance of any obligation. Unlike the traditional forms of electronic money, there is nobody who owes obligation to the holder of cryptocurrencies. If right conditions exist, the market will find an inherent value in cryptocurrencies (or some of them) and recognises them as substitutes for money. Cryptocurrencies are electronic records and transferable on a blockchain. But they do not fall within the definition of the Model Law.
A blockchain may also be used to transfer an electronic record indicating proprietary interests (such as security interests) in tangible or intangible properties. My suggestion for substituting a blockchain-based ledger for the registry of the Cape Town Convention (See my earlier post) relates to this type of usage. Such electronic transferable records do not represent entitlement to claim the performance of any obligation. Hence, they do not come within the definition of the Model Law.
The current Model Law project, therefore, leaves a vast area of use cases of the blockchain technology untouched. I think that UNCITRAL can be instrumental in harnessing the technology in all its applications. 

Work of UNCITRAL on electronic transferable records

Since 2011, the UNCITRAL has been working on legal issues relating to the use of electronic transferable records. From the beginning, it envisages two approaches to establishing the identity of the person to whom an electronic transferable record is issued or transferred, namely the token model which identifies the person in the record itself and the registry model which identifies the person in a separate registry (A/CN.9/WG.IV/WP.115 (hereafter "the 2011 document") at para. 48). Like its previous works on electronic commerce, the UNCITRAL is adhering to the principle of technology neutrality (Id. at para. 35) and nowhere in the official documents published to date could I find any mention of blockchain, cryptocurrency or bitcoin. But I think the work should be pursued with the blockchain technology in mind so as to facilitate its applications to replace paper-based transferable documents such as bills of lading.
In the current draft of a model law (A/CN.9/WG.IV/WP.135/Add.1, August 2015), the exclusive control of an electronic transferable record is treated as functionally equivalent to the possession of a paper-based transferable document. Thus Draft Article 17(1) provides:

Where the law requires the possession of a paper-based transferable document or instrument, that requirement is met with respect to an electronic transferable record if:
(a) A method is used to establish exclusive control of that electronic transferable record by a person and to reliably [identify] [establish] that person as the person in control; and
(b) The method used is either:
(i) As reliable as appropriate for the purpose for which the electronic transferable record was generated, in light of all the relevant circumstances, including any relevant agreement; or
(ii) Proven in fact to have fulfilled the functions described in subparagraph (a) above, by itself or together with further evidence.

In the Remarks which accompany this provision, the Secretariat makes some noteworthy comments. It says, "the electronic transferable record in itself does not necessarily identify the person in control, but rather the method or system employed to establish control as a whole performs that function" (Id. at para. 22.). This understanding may simply be intended to cater for the registry model as described in the 2011 document. But it may also open the door to blockchain-based electronic records which may be seen as ill-fitted with the description in the 2011 document of the token model. 

The Secretariat goes on to say, "identification should not be understood as implying an obligation to name the person in control, as the draft Model Law allows for the issuance of electronic transferable records to bearer, which implies anonymity" (Ibid.). This view will also ease the way for the blockchain technology as the latter permits the holders of electronic records to remain anonymous. 

The Secretariat also notes, "reference to the person in control of the electronic transferable record does not imply that that person is also the rightful person in control of that record as this is for substantive law to determine" (Id. at para. 21). In other words, the exclusive control of an electronic transferable record is only equivalent to the possession of a paper-based transferable document. Put in the context of the blockchain technology, this view seems consistent with my opinion that the ownership of a blockchain-based electronic record cannot be determined simply by reference to who has the exclusive control of it (See my earlier post).

The Secretariat proceeds to say, "reference to the person in control does not exclude the possibility of having more than one person in control." It is not easy to see what this observation means for a blockchain-based electronic record. The latter is under the exclusive control of the person holding the private key for the address in which the record is kept. It is, on the other hand, possible for one private key to be known by a number of persons.

Applicability of CISG

The CISG (UN Convention on Contracts for the International Sale of Goods) does not expressly define what constitutes a "contract of sale." 

There can be little doubt, however, that the Convention will have no application to contracts to buy cryptocurrencies with traditional currencies since the Convention is only applicable to sales of "goods" which are generally understood to mean tangible objects (e.g. Schlechtriem, Commentary on the UN Convention on the International Sale of Goods (1998) p. 23 [Herber]).
A separate question is whether the CISG is applicable to contracts to buy goods with cryptocurrencies. In the language of the Convention, the question is whether payment of cryptocurrencies should be interpreted as constituting "payment of the price." If not, such contracts would be barter contracts and accordingly fall outside the scope of the Convention (For barter contracts, see e.g. Schlechtriem.)
In this connection, it is worth recalling that the CJEU observed in its VAT judgment that bitcoins had no purpose other than to be a means of payment (para. 24). In other legal contexts, other purposes such as speculative investment purposes may be relevant. But in the context of CISG, a better interpretation seems to be that payment of bitcoins or other cryptocurrencies constitutes "payment of the price," since they have no intrinsic value of their own.