C. HOW TO RECONCILE DLT AND PRIVATE LAW
The law that applies to blockchain transfers and the resulting positions presents a conundrum. In the following, a proposal will be made. Before doing so, this article will explain the outer constraints that every proposal must respect regarding the application of private law to DLT.
1. UNDERPINNINGS OF THE PROPOSAL
Any suggestion for combining the blockchain with private law must take into consideration all three problems that have been identified in the preceding section: the autonomy of DLT, the immutability of transfers, and the a-national character of the blockchain. What is needed is a mechanism that respects the
OBLIGATIONS IN EUROPE 217, 217 (Franco Ferrari & Stefan Leible eds., 2009) (discussing the impact of Rome I on the law applicable to assignment).
106. See Fairfield, supra note 3, at 826–27.
107. Id. at 842–63.
108. See Bayern, supra note 36, at 29.
result of bitcoin transactions—in particular, one that does not try to reverse them and press them into the Procrustes bed of national law—while at the same time responds to the requirements of private justice. In addition, such an approach should not require the elaboration of uniform global rules, which at the moment seem elusive. Instead, it should be fully compatible with the division in national laws that currently exists.
The forthcoming proposal respects all four conditions. It suggests an application of the law that respects the autonomy of DLT, the immutability of transfers and abstains from imposing one national law on the whole blockchain, all without requiring the development of new global rules. Even though this may seem like a perfect solution, the proposal risks coming under fire from both the proponents of the technology as well as from lawyers, because it is based on certain underpinnings that either of them may dislike. To reduce this risk, these fundamental underpinning shall be disclosed in the following. Basically, the proposal is driven by two convictions for which it should not be attacked.
The first conviction is that the blockchain is a useful innovation that can yield significant societal benefits and should therefore be allowed to continue to flourish.109 DLT provides a stable, nonrepudiable and largely tamper-proof mechanism to transfer assets around the world. In the great majority of cases, and provided it is not abused for illegal purposes, it works perfectly without the law.110 This is an advantage that should be maintained. The attractiveness of DLT would greatly suffer if lawyers tried to change the code. Even indirect changes should be avoided, such as a requirement to include a choice of law in the blockchain, for these changes would gravely compromise the functioning of DLT.
The second conviction is that code is not law and that the positions obtained on the blockchain cannot be the end point of
109. See, e.g., Fairfield, supra note 3, at 874 (characterizing DLT as trustless ledgers tracking transactions in real time at comparatively low cost).
110. At the end of the third quarter of 2019, a total number of 311,396 Bitcoin daily transactions were recorded world-wide. See Number of Daily Bitcoin Transactions from 1st Quarter 2016 to 3rd Quarter 2019, STATISTA, https://www.statista.com/statistics/730806/daily-number-of-bitcointransactions/ (last visited Feb. 2, 2019). This number contrasts with the very few instances in which legal problems or disputes have arisen.
ownership analysis but instead require supplementation and additions. Though it works in the majority of cases, in exceptional situations the law must correct the result achieved by the use of DLT. This article has identified above the instances of mistake, fraud, and improper threat, but also those of theft, bankruptcy and succession. From a legal point of view, all of these circumstances require a solution different from that of the blockchain. As the technology does not provide it, the law must step in. It should do so not by invalidating the transfer— something which would be technologically unfeasible. Instead, another means must be established to achieve a balanced and just result.
In sum, there is undeniably a tension between the law and the blockchain. Nevertheless, they must be reconciled if one shares the two convictions just outlined. The thesis of the following proposal is that the blockchain and private law are not mutually exclusive but can exist beside each other. Law and technology must neither ignore nor fight one other. They should live in a symbiosis with each leaving to the other its own field of competence.
2. ACCEPT DLT AS A FACT
The first step of the new solution is that the law should not interfere with the blockchain. The technology should essentially continue to function as it currently does without the law and without the intervention of lawyers. Transfers should be done on the basis of private and public keys only. Any introduction of legal conditions or requirements should be omitted.
This means that the law should not question the validity of blockchain transactions. This would be a hopeless enterprise anyway. The power of the holder of bitcoin resides in his knowledge of the private key. This and the public key of the recipient is all that is needed to initiate a transfer. To call such a transfer “invalid” from a legal point of view would not change the factual power of the private key’s holder to initiate a new transfer, which will then result in a corresponding power of the recipient, and so on. Importantly, this result comes about by technology, not by the law. The legal system is unable to avoid the passing on of crypto assets, and it should not try to inhibit it.111
Instead, the immutability of the transfer from a technical point of view is a fact that lawyers must accept. Choosing to ignore it would come at the cost of failing to provide a solution that is workable in real life.
One may compare the situation to that of a cash payment: The transfer in this case comes about by a factual element, the delivery of one or more banknotes or coins. The law accepts and confirms the transfer because the transferee becomes the owner of the banknote from a legal perspective. This is not the case where an agreement supporting the transfer is lacking, e.g., because the money has been stolen. Yet even a thief can provide title to cash to a bona fide creditor.112 The fact that he possesses the notes or coins allows him to transfer title to a transferee in good faith. The original owner keeps his title and can demand the cash back only as long as the illegal possessor has not spent the money.113
A similar type of legal analysis should also be applied to DLT. The entry into the blockchain is a fact that reveals the current holder of the crypto asset. This position allows him factually and legally to procure title to another recipient. In order to determine this power, it is unnecessary to investigate the validity of the previous transactions recorded on the blockchain. Specifically, one should not go back in time by conducting a “title search” to find out whether the transferor had a position she can transfer, and her predecessor, and so on. As in the case of cash, such a title search is counterproductive because of the fungibility of coins and their function as means of
111. See WRIGHT & DE FILIPPI, supra note 2 at 184 (stressing that any attempt by the government to introduce a technological backdoor or other access control on both hardware and software makes the technology weaker).
112. See Miller v. Race, (1758) 97 Eng. Rep. 398, 401 (“[I]n the case of money stolen, the true owner can not recover it, after it has been paid away fairly and honestly upon a valuable and bona fide consideration . . . ”); see also Transamerica Insurance Co. v. Long, 318 F. Supp. 156 (W.D. Pa. 1970) (denying restitution of money that a bank robber had paid to tax authorities); Atlantic Cotton Mills v. Indian Orchard Mills, 17 N.E. 496, 501 (Mass. 1888) (“There is no doubt that a thief may use stolen money . . . to pay his debts, and in such case an innocent creditor may retain the payment.”). See generally Andrew Kull, Defenses to Restitution: The Bona Fide Creditor, 81 B.U. L. REV. 919, 937 (2001) (providing further cases and commentary on stolen money used for debts).
113. Miller v. Race, (1758) 97 Eng. Rep. 398, 401 (“[B]ut before money has passed in currency, an action may be brought for the money itself . . . . ”).
payment. Lawyers should not second-guess the blockchain by controlling each and every transfer, either giving it their stamp of approval or denying its validity. This approach would make DLT essentially useless; it would become an expensive record system without any practical value. One should therefore accept the record on the blockchain as a fact which creates the power to transfer. This also means that those that have obtained a private key via DLT should, without any showing to the contrary, be seen as the legitimate holders of the crypto asset. As such, their position deserves to be protected by the law.114
An exception should apply only where it can be proven that the crypto asset has been obtained illegally, in particular by hacking, blackmailing, or fraud. In these cases, the presumption of a legal effect is rebutted. The exceptional situation is similar to that of a stolen banknote and will be dealt with in more detail later.115 Apart from such an event, a transfer on the blockchain should be accepted as such.
3. FOCUS ON THE REVERSE TRANSACTION
The fact that transfers recorded on the blockchain cannot be undone does not mean, however, that one would have to consider the situation as presented by the blockchain as final.116 Though it is impossible to delete a block once it has been added to the chain, the law can reverse the effects of such transfer. The means for doing this is ordering a reverse transfer.117 For instance, though the record of a transfer of bitcoin cannot be undone and deleted from the blockchain, the recipient of an erroneous transfer can be obliged to transfer the cryptocurrency back to the sender. The same obligation can be imposed on the party that has not effectuated its counter-performance under a transaction.
114. See infra Section D1.
115. See discussion infra Section D1.
116. See Nakamoto, supra note 16, at 1, 8 (indicating that transactions will be computationally irreversible but that subsequent transactions can occur provided they satisfy the consensus mechanism of the DLT network).
117. For example, such reverse transfer may take the form of a judgment ordering replevin of stolen or fraudulently conveyed cryptocurrency. See Angela Morris, Judge Orders $ 30 Million in Bitcoin to Be Returned in Cryptocurrency Class Action, MIAMI DAILY BUS. REV. (Aug. 3, 2017) (summarizing a default judgment against Project Investors Inc. in a class action case regarding stolen cryptocurrency); see also Leidel v. Project Inv’rs, No. 9:16-cv-80060-MARRA (S.D. Fla. July 27, 2017), ECF No. 123 (ordering a default judgment of 11,325.0961 BTC against Project Investors Inc.).
Even in the case of hacking, blackmailing, or fraud, it makes sense to force the tortfeasor to return the illegally obtained assets because the ineffectiveness of the transfer from a legal point of view does not bestow a private key to the victim. The reverse transfer restores the parties to the same positions they had been in before the transfer. For all practical purposes, it cancels the effects of the first transfer.
It is important in this context to note a certain ambiguity of the term “reversible.” Insofar as it means annulling a transfer as if it had never happened, it is not a workable option for most DLT networks. But insofar as it refers to a reverse transfer as a result of which a new private key is created for the victim, it is certainly feasible with the help of the law. The law cannot undo a fact, but it can provide remedies aiming to reverse the situation that had been achieved. What comes to the fore here is the difference between a set of facts and a normative order. The law as a normative order cannot undo a fact, e.g., a tort that has been committed, a document that has been handed over, or work that has been performed. Yet it can remedy the consequences of these facts rectroactively. Just as the effects of an unjust enrichment can be compensated by a restitution claim, the law can impose an obligation on the recipient of virtual assets recorded on the blockchain to return what has been received.118
The idea of a reverse obligation to correct legally incorrect transfers marries the dominant features of the technology, its autonomy, and nonrepudiability, with the practical need for correcting unjust and societally unbearable results. This is achieved by imposing an obligation to return, which can be complied with by using the methods of DLT. In this way, the blockchain is not “invalidated” but supplemented with an additional reverse transfer. The reversal takes place in the form that the DLT provides and thus does not create any contradiction or upheaval. The law is adapted to the particularity of the technology to achieve its aims.
Yet, there is a catch. The actual performance of the reverse transfer depends on the will of the recipient.119 He must make
118. See RESTATEMENT (THIRD) OF RESTITUTION & UNJUST ENRICHMENT § 54(2)(a) (2008) (“Rescission requires a mutual restoration and accounting in which each party restores property received from the other, to the extent such restoration is feasible . . . .”).
119. See Andrew W. Balthazor, Comment, The Challenges of Cryptocurrency Asset Recovery, 13 FIU L. Rev. 1207 (explaining the challenges of recovering cryptocurrency even when ordered by a court). See also Max I. Raskin, Realm of the Coin: Bitcoin and Civil Procedure, 20 FORDHAM J. CORP. & FIN. L. 969, 975 (2015) (explaining how the possessor of a cryptocurrency account’s private key has total control over the account).
use of his private key to send the crypto assets back to the sender. It is by no means sure that he will comply with his obligation.120
But this peculiarity does not make the reverse transfer improbable or unlikely.121 The legal system has mechanisms to force the use of keys or any other human action. Examples include a court order and the obligation to pay a fine in case of its violation for “contempt of court.”122 Of course, these legal mechanisms are not as certain to succeed as would be the technical deletion of the transfer, which would restore the transferred asset directly to the former holder. Yet such a deletion is not possible, or only possible at a high cost.123 Moreover, the obligation to use a private key to retransfer assets is not very different from other court orders, say, to restore a physical asset or perform another act, e.g., providing testimony as a witness. Legal enforcement works at least in many, if not in most, cases. The undeniable truth that the law is sometimes broken or disobeyed does not mean that it is useless to impose a normative order.124
The consequences of the “reverse transfer approach” shall be illustrated using a practical example: Let us imagine that A wants to exchange a bitcoin in U.S. dollars and enters into an online transaction with B, who is a fraudster. A transfers the bitcoin via the blockchain to B, but B never transfers U.S. dollars. A court of law would order B to transfer the bitcoin back to A. If B does not comply, he will be in contempt of court and
120. See Balthazor, supra note 119, at 1219 (describing methods of enforcing judgements against defendants that hold cryptocurrency).
121. See id. at 1235 (“Cryptocurrency asset recovery poses challenges surmountable under the right conditions.”).
122. See id. at 1226 (“Cryptocurrency may be seized, pursuant to a levy or writ of replevin.”); see also FED. R. CIV. P. 70(e) (“The court may also hold the disobedient party in contempt.”).
123. See Nakamoto, supra note 16, at 8 (showing that for the bitcoin blockchain, it is computationally impractical to reverse a transaction without controlling the majority of computing power in the network).
124. See HANS KELSEN, PURE THEORY OF LAW 113 (2005) (explaining that a law or norm is not invalidated by the existence of contrary behavior); see also H. L. A. HART, THE CONCEPT OF LAW 84 (2012) (distinguishing the normative rules of law from predictive language).
ordered to pay a fine. The same obligation to retransfer could be imposed on the recipient of cryptocurrency from a transferor who subsequently is declared bankrupt. If the transfer is done during the suspect period, the assets would have to be restored to the bankruptcy estate through a new transfer.
4. STOP THINKING ABOUT PROPERTY TRANSFERS
The essence of the proposal made here is to substitute a conceptualization of the transfer in terms of property law by an analysis that is based on remedies under the law of obligations. No longer is it necessary to enquire into the ownership of bitcoin or other cryptocurrencies. For the vast majority, the law accepts and protects the results produced by the blockchain.
The abandonment of a property law analysis of the transfer has two main advantages. The first is that it is no longer necessary to probe and second-guess the validity of every DLT transfer. The distributed ledger is accepted as is. This not only allows the technology to operate without disturbances, it also spares the useless effort to “correct” the blockchain.
The second advantage is that the holder of a private key whose bitcoin has been hacked or stolen can rely on the law’s protection. She is not obliged to prove title to the bitcoin by relying on circumstances outside of the blockchain, specifically that the person she obtained the coins from was the “owner” who legitimately obtained it from the former “owner” and so on. Such a parallel “title search” would indeed be impossible given the decentralized and pseudonymous working of DLT. The blockchain itself is the ledger that confers legitimacy.
A further advantage becomes visible at the international level. Excluding a property analysis dispenses with the need to look for the one national law that governs the transfer. As has been shown above, it is impossible to identify such a law for completely distributed ledgers.125 In addition, it is also a futile analysis, as the law cannot in any sense “validate” a blockchain transfer. The “validity” is certified by technology. Its result cannot be annulled or voided by law.126 It is thus not only impossible, but also useless to search for the law that “governs” a transfer on the blockchain. There is no need for such law, as DLT is a factual and global process.
125. See supra Section A.3.
126. See supra Section A.3.
The many legal questions raised by such transfers cannot be answered by one legal system, but only by a plurality of different laws. These laws concern, for instance, the right of the victim of a fraud or theft to have the assets returned, the obligation to restore assets transferred by mistake, or the fate of crypto assets in the event of the death or bankruptcy of their holder. Why should one national law govern all of these questions? It conforms much more to the current reality of split legal systems to answer these questions by simultaneously applying different national laws.
Take the example of an agreement for the transfer of bitcoin. Such an obligation will usually only be undertaken against some consideration. The transfer is thus part of the performance of a contract. It is important to pay attention to the precise wording of the previous statement: The bitcoin transfer is not a contract but the performance of a contract. The transfer serves to fulfill an obligation arising under a contract concluded outside the blockchain, such as a sales contract for some object that is paid for in bitcoin. This contract is submitted to some national law in accordance with the ordinary rules of conflict of laws. The law governing the contract determines whether the agreement is invalid, e.g., in case of mistake. The same law will also determine the consequences if the transfer made in the contract’s execution has to be returned.127 Since the agreement is concluded outside the blockchain, it is not difficult to determine the contract’s governing law. This law is identified by the usual rules of private international law: in the case of a sale, for instance, the parties can agree to the applicable law to their contract;128 in the absence of a choice by the parties, many tribunals would apply the law of the habitual residence of the seller.129
127. It is generally agreed that the law applicable to a contract also governs the consequences of its invalidity. See Rome I, supra note 93, at art. 12(1)(e) (“(1) The law applicable to a contract . . . shall govern in particular: . . . (e) the consequences of the nullity of the contract.”); RESTATEMENT (SECOND) OF CONFLICT OF LAWS § 221(1) (1971) (stating that rights and liabilities of the parties to a contract in actions for restitution are governed by the state law which “has the most significant relationship to the occurrence and the parties under [choice-of-law principles].”); HAY ET AL., supra note 14, at 1218–22 (providing a more nuanced discussion of the Second Restatement § 221).
128. See sources cited supra note 119 and the accompanying text.
129. See, e.g., Zhonghua Renmin Gongheheguo Shewai Minshi Falvguanxi Shiyongfa (中华人民共和国涉外民事关系法律适用法) [Laws Applicable to Foreign-Related Civil Relations] (promulgated by the Standing Comm. Nat’l People’s Cong., Oct. 28, 2010, effective April 1, 2001) art. 41 (China) (“Where the parties do not so select [the law applicable to a contract], the law of the habitual residence of the party whose performance of contractual obligations can most reflect the characteristics of the contract [shall govern].”); Hō no Tekiyō ni Kansuru Tsūsokuhō [Act on the General Rules for Application of Laws], Act No. 78 of 2006, art. 8(2), (translated in (Japanese Law Translation [JLT DS]), http://www.japaneselawtranslation.go.jp/law/detail/?id=1970 (“In the case [where the parties of a contract do not choose a governing law], if only one of the parties is to provide a characteristic performance involved in a juridical act, the law of the habitual residence of the party providing said performance . . . shall be presumed to be the law of the place with which the act is most closely connected.”); GRAZHDANSKII KODESK ROSSIISKOI FEDERASTII [GK RF] [Civil Code] art. 1211(1) (Russ.) (stating that the governing law is the law of the country in which the primary activity occurs or where the principle actor is located); Swiss PILA, supra note 93, at art. 117(1)–117(2) (“(1) Failing a choice of law, contracts are governed by the law of the state with which they have the closest connection [such as where the performing party habitually resides].”); Rome I, supra note 93, at art. 4(2) (“Where the contract [has not established jurisdiction through art. 4(1) or is covered by multiple parts of art. 4(1)], the contract shall be governed by the law of the country where the party required to effect the characteristic performance of the contract has his habitual residence.”).
It is thus both easy and appropriate to apply the law governing a contract (if there is any) to the obligation to restore the crypto assets in case of nullity of that contract. Determining this law is easy because the contract is a phenomenon outside the blockchain. One can rely on the connecting factors supplied by the usual conflicts rules that point to circumstances beyond the chain, e.g., the choice of law by the parties or the habitual residence of one of them, to determine the law that governs the reversal obligation. It is not necessary to identify a law governing the blockchain as such.
If there is no contract because, for instance, the transferor has been blackmailed into making the transfer, then the conflict rules for torts apply. Most legal systems in the world refer to the law in force at the place of the tort, the so-called lex loci delicti.130
130. See, e.g., Zhonghua Renmin Gongheheguo Shewai Minshi Falvguanxi Shiyongfa (中华人民共和国涉外民事关系法律适用法) [Laws Applicable to Foreign-Related Civil Relations] (promulgated by the Standing Comm. Nat’l People’s Cong., Oct. 28, 2010, effective Apr. 1, 2011) art. 44 (China) (“Tort liability shall be governed by the law of the place where the tort occurs.”); Hō no Tekiyō ni Kansuru Tsūsokuhō [Act on General Rules for Application of Laws], Law No. 78 of 2006, art. 17, translated in (Japanese Law Translation [JLT DS]), http://www.japaneselawtranslation.go.jp (Japan); GRAZHDANSKII KODESK ROSSIISKOI FEDERASTII [GK RF] [Civil Code] art. 1219(1) (Russ.) (stating that the law of the country where the action occurred applies); Swiss PILA, supra note 93, at art. 133(2) (describing how the law of the state where the tort occurred generally applies); Commission Regulation 864/2007, On the Law Applicable to Non-Contractual Obligations (Rome II), pmbl. (15), 2007 O.J. (L 199) 40 [hereinafter Rome II] (“The principle of the lex loci delicti commissi is the basic solution for non-contractual obligations.”).
Challenging for this approach are cross-border torts, in which the damage and the harmful conduct occur in different countries. Some states give prominence to the place of damage.131 Others consider the place of conduct as more important, but make an exception where the tortfeasor could foresee that the conduct would have harmful effects in another country; in this case, they equally follow the law of the place of damage.132 A good case could be made that such damage occurs at the place of the victim’s domicile. The same result may be obtained using the governmental interest analysis that is followed by many states in the United States because arguably the country of residence of the victim has the strongest interest in regulating this tort.133 The blackmailer would therefore be subject to the law of the victim’s country, which would oblige him to restore the crypto assets.
In sum, it is unnecessary to analyze DLT transfers in terms of property law. Rather, the entries on the blockchain should be accepted as they are. This does not mean that they are conclusive with regard to the final distribution of crypto assets. Where they are the result of a void contract or a tort, the crypto assets must be restored under the applicable contract or tort law. The advantage of such an approach can hardly be overestimated. Not only does it avoid the need for title search for crypto assets and the factually impossible deletion of a transfer from the ledger, it also spares the vain search for the law applicable to the blockchain as such because it accepts the ledger for what it is:
131. See, e.g., Hō no Tekiyō ni Kansuru Tsūsokuhō [Act on General Rules for Application of Laws], Law No. 78 of 2006, art. 17, translated in (Japanese Law Translation [JLT DS]), http://www.japaneselawtranslation.go.jp (Japan) (“The formation and effect of a claim arising from a tort shall be governed by the law of the place where the result of the wrongful act occurred.”); Rome II, supra note 130, at art. 4(1) (“[T]he law applicable to a non-contractual obligation arising out of a tort/delict shall be the law of the country in which the damage occurs.”).
132. See GRAZHDANSKII KODESK ROSSIISKOI FEDERASTII [GK RF] [Civil Code] art. 1219(1) (Russ.) (“In cases when the action or other circumstances caused harm in another country, the law of that country may be applied if the person causing the harm foresaw or should have foreseen the onset of the harm in that country.”); Swiss PILA, art. 133(2) (“However, if the result occurred in another state, the law of such state applies if the tortfeasor should have foreseen that the result would occur there.”).
133. See HAY ET AL., supra note 14, at 808–22 (discussing governmental interest analysis and torts).
an autonomous, self-contained, global transfer mechanism. The technology is allowed to flourish and any doubling with a legal analysis is avoided. Consequently, no national law governs blockchain transfers, but rather the autonomous rules of the protocol, if need be, are supplemented with a remedy under an easily identifiable national law.