Digital Assets and Property Law

Are Digital Assets Property Under American Law?

In the United States, digital assets are property. Both federal and state courts have recognized cryptocurrencies, stablecoins, non-fungible tokens (NFTs), and other digital assets as “the subject matter of property, albeit through disparate analytical approaches that privilege practical outcomes and pragmatism over theoretical clarity and doctrinal coherence.”¹ American courts have, for the most part, “recognized digital assets as personal property almost as a fait accompli based on their widespread commercial circulation and social adoption, focusing primarily on remedial considerations.”² The question is far from academic, for “[e]very day, market participants execute millions of transactions worth billions of dollars, relying on an uncertain patchwork of traditional legal frameworks awkwardly adapted to this new technology.”³

At the federal level, this recognition has emerged primarily through bankruptcy proceedings, enforcement actions, and select agency guidance. Bankruptcy courts have consistently treated digital assets as property of the estate, as in In re Celsius Network LLC, In re Voyager Digital Holdings, Inc., In re Genesis Global Holdco, LLC, and In re FTX Trading Ltd., where the analysis centered on whether customer-deposited cryptocurrencies belonged to the debtor’s estate.⁴ Federal courts have likewise accepted digital assets as property capable of seizure in criminal-forfeiture and regulatory-enforcement contexts, as in the Silk Road prosecution, United States v. Ulbricht.⁵ The Securities and Exchange Commission and the Commodity Futures Trading Commission have cemented this position through enforcement actions in which digital assets have been implicitly recognized as personal property, and the Internal Revenue Service has considered cryptocurrencies a form of property as far back as 2014.⁶

Courts applying state law have generally arrived at the same conclusion, adopting a pragmatic stance that implicitly espouses the functional approach to property. This is particularly evident in cases involving conversion claims, fraudulent transfers, and other traditional property-based causes of action, such as Temurian v. Piccolo, Ox Labs, Inc. v. BitPay, Inc., and Archer v. Coinbase, Inc.⁷ Scholarship addressing the foundational property-law questions rather than regulatory concerns has come from a relatively small cohort, most prominently Juliet Moringiello, Shawn Bayern, and Joshua Fairfield; it has largely converged on the same functional conclusion.⁸

This recognition is not idiosyncratic to the United States. Common-law jurisdictions abroad and civil-law systems alike have confronted the same question, with results surveyed below in How Does American Law Compare to Other Jurisdictions on the Property Question?

What Test Do American Courts Use to Decide Whether a Digital Asset Is Property?

American courts have not adopted a single doctrinal test for the property status of digital assets. As noted above, they have for the most part proceeded pragmatically, treating digital assets as property in order to resolve the dispute before them and espousing, at least implicitly, a functional conception of property that asks how property should be used as a tool of social life, rather than whether a candidate interest exhibits the formal attributes of an established category.⁹

A more principled framework may be extrapolated from Kremen v. Cohen, in which the United States Court of Appeals for the Ninth Circuit confronted the then-novel question of whether a domain name could be property under California law.¹⁰ Drawing on California precedent that property encompasses “every intangible benefit and prerogative susceptible of possession or disposition,” the court articulated a three-part test: there must be an interest capable of precise definition; it must be capable of exclusive possession or control; and the putative owner must have established a legitimate claim to exclusivity.¹¹

Digital assets are particularly amenable to analysis under this framework. As data entries in distributed-ledger systems, each cryptocurrency token, NFT, or stablecoin unit has exact, mathematically defined properties and quantities, satisfying the first requirement. The second requirement of exclusive possession or control is generally achieved through cryptographic mechanisms such as public-private key pairs; in Hohfeldian terms, “this control represents a ‘privilege,’ conferring on keyholders the liberty to use and transfer the asset freely, while creating in all others a correlative ‘no-right’ to prevent or interfere with such activities.”¹² The third requirement of a legitimate claim to exclusivity is established through the technological architecture of the relevant protocol, whether proof-of-work mining, proof-of-stake validation, or smart-contract execution.

The Kremen test echoes the four-part analytical indicia set out by Lord Wilberforce in National Provincial Bank Ltd v. Ainsworth, which have guided property recognition in common-law jurisdictions outside the United States and which ask whether an interest is definable, identifiable by third parties, capable in its nature of assumption by third parties, and possessed of some degree of permanence or stability; the American test, however, is the more functional of the two.¹³

Are Digital Assets Choses in Possession, Choses in Action, or Something Else?

In American law, this classification question is of limited consequence. In common-law systems, personal property has traditionally been framed by the fundamental division between choses in possession and choses in action: the former are rights in things susceptible of physical possession, the latter rights enforceable only by action and not by taking physical possession.¹⁴ In the United States, however, this dichotomy has receded from prominence, and there has been no real debate over whether digital assets are choses in possession or choses in action. In American law, “the crucial question is whether and how digital assets fit within these UCC categories and their corresponding regimes”; common-law classification “plays a consequential but residual role confined to those areas where the UCC does not apply.”¹⁵

Three developments produced this American distinctiveness. First, American property scholarship has traditionally centered on land and real estate, leaving personal property comparatively understudied as a theoretical matter.¹⁶ Second, when American scholars have engaged personal property, they have concentrated on the practical incidents of ownership, possession, control, transfer, and use as collateral, rather than on taxonomy; the last sustained scholarship on the chose in action as a category dates to the first quarter of the twentieth century.¹⁷ Third, and most importantly, this functional orientation has produced a new body of personal-property categories that has structurally curtailed the significance of the summa divisio. That new taxonomy finds its primary expression in the UCC, which organizes commercial personal property into a distinctly American framework that includes goods, accounts, payment intangibles, investment property, and intangibles. Though this framework bears traces of the traditional chose distinction, it operates largely independently of it.¹⁸

The contrast with other common-law jurisdictions is instructive. Outside the United States, the property question for digital assets has been litigated and debated almost entirely within the chose framework. Courts and commentators agree that digital assets cannot be choses in possession, because possession requires tangibility; they divide over whether digital assets are choses in action or instead belong to a tertium quid, a third category of personal property for data objects that exist independently of the law.¹⁹ What is striking about this debate is that the preoccupation with classification “has stifled deeper investigations into the legal rules governing their commercial circulation,” leaving such fundamental questions as the precise subject matter of the property right, the mechanisms by which title passes, and the application of the nemo dat (no one gives what he does not have) rule and its exceptions largely unexplored.²⁰ Having set the classification question aside, American law proceeded directly to those questions of commercial circulation.

How Did American Law Treat Digital Assets Before the 2022 UCC Amendments?

Before the 2022 Amendments, American law addressed the voluntary transfer of digital assets, their use as collateral in secured transactions, and their tokenization by applying existing state common-law principles and UCC categories to a novel kind of intangible. In each of these three domains, the inherited framework proved poorly suited to the realities of digital asset markets.²¹

The threshold question for the voluntary transfer of ownership was whether a transfer of digital assets is a sale of “goods” governed by Article 2 of the UCC. Article 2 defines goods as things that are movable at the time of identification to the contract and that are not money, investment securities, or things in action.²² Digital assets are readily movable, but the better view was that transfers of digital assets fall outside Article 2 altogether, and practitioners generally proceeded on that footing. “Article 2 is founded on the premise that sales contracts concern corporeal things over which a person can have physical dominion”; its cornerstone concepts of possession and delivery assume tangible things, and that premise makes the statute inherently incompatible with intangibles, whether or not digital assets are characterized as things in action.²³ Transfers were therefore governed by state common-law rules on the assignment of intangibles, a body of law that presented three grave problems. First, those rules varied considerably from state to state, imposing legal fragmentation on markets that are borderless and depend on velocity.²⁴ Second, they had been developed to govern receivables, and a body of law built for monetary claims against identifiable persons was conceptually ill-suited to digital assets that exist as self-contained data objects rather than as claims against anyone.²⁵ Third, and most problematic, they applied the nemo dat principle strictly, with no good-faith purchaser exception, so that a buyer could never be certain of acquiring an asset free of competing claims without verifying the entire chain of title back to the creation of the asset, a practical impossibility given the pseudonymity and velocity of these markets.²⁶

The use of digital assets as collateral in secured transactions fared better. UCC Article 9 supplied a framework that was, in principle, clear. Because Article 9 contained no category designed for digital assets, they fell within the residual catchall of “general intangibles.”²⁷ A security interest attached once the secured party gave value, the debtor had rights in the collateral, and the parties authenticated a security agreement describing it; perfection could be achieved by only one method, filing a financing statement; and priority followed the general first-to-file-or-perfect rule.²⁸ This regime, however, was fundamentally misaligned with the workings of digital asset markets. As the sole method of perfection, filing was particularly problematic. The time lag of public registration stood in stark contrast to the near-instantaneous execution of digital asset transfers, and the pseudonymity of distributed-ledger networks impaired both the identification of the correct filing jurisdiction and the searching of existing liens.²⁹ Tellingly, an arrangement emerged that ignored Article 9 entirely. Debtors transferred their digital assets directly to lenders as security, with minimal documentation and no public filings, reflecting a preference for taking control of the collateral rather than relying on registries. By the early 2020s, scholars and practitioners alike had come to regard this state of affairs as unsustainable.³⁰

Pre-2022 attempts at tokenization were almost invariably flawed. The practice uses a digital asset to represent ownership of another asset or a right enforceable against a person, and the token employed is typically an NFT.³¹ Representing rights in one thing through another is an ancient feature of commercial law: negotiable instruments carry payment rights, share certificates represent fractional corporate ownership, deeds embody interests in land, and bills of lading and warehouse receipts confer title to goods in transit and storage. Yet property law constrains these devices in two respects that private parties cannot override. First, it limits which kinds of property may be tokenized and through which instrument; these categories are fixed by law rather than left to private discretion. Second, it prescribes the mandatory form and substance of each instrument.³² These constraints express the numerus clausus (closed number) principle, under which “parties cannot create new forms of property rights beyond the finite list of standardized categories recognized by law, ensuring property interests remain uniform and readily comprehensible.”³³ Proponents of tokenization embraced the flexibility of contract law but disregarded these property-law strictures, mistakenly assuming that the capabilities of the technology would resolve the private-law questions that arise whenever one thing is said to represent rights in another.³⁴ The defect is easily illustrated. If a seller issues an NFT purporting to represent ownership of an off-chain asset and transfers it to a buyer, but then transfers the actual asset to a third party who takes delivery or a valid assignment, the NFT holder cannot claim the underlying asset; the token does not legally embody ownership of it, and the buyer is left only with a claim for damages against the seller.³⁵

What Did the 2022 UCC Amendments Change?

By 2018, it had become apparent that the Uniform Commercial Code was struggling to accommodate distributed-ledger technology and digital assets. In response, the Uniform Law Commission and the American Law Institute convened a joint committee, consulted widely with industry participants, practitioners, and scholars, and concluded that substantial revisions were required. The American Law Institute approved the resulting 2022 Amendments in May 2022, and the Uniform Law Commission followed two months later; as of early 2026 they have been enacted in thirty-three states and the District of Columbia, including New York, Delaware, California, and Florida, and continue to progress toward nationwide adoption.³⁶

At the heart of the 2022 Amendments lies a new article of the Code, Article 12, which establishes a bespoke private-law framework for a new category of personal property: the controllable electronic record (CER). “CERs are a new category of personal property, distinctly grounded in American functional pragmatism and without a parallel in traditional common law taxonomies”; rather than asking whether digital assets are choses in action or some third category, Article 12 defines its category by reference to the functional attribute of control and supplies rules for the transfer of CERs, their use as collateral in secured transactions, and the limited circumstances in which they may carry rights in other property.³⁷

This is a deliberately American solution with a precedent in the Code itself. In the 1950s, Article 9 of the Code marked an American departure from the way other common-law jurisdictions governed secured transactions, replacing inherited categories with an array of functionally defined collateral types. Article 12 repeats that choice for digital assets, forging a new personal-property category around operational characteristics rather than doctrinal purity, and in doing so aligning the law with commercial practice and market expectations.³⁸ The doctrinal mechanics of Article 12 are treated in detail on the companion research page on UCC Article 12 and Controllable Electronic Records; the sections that follow address the framework only to the extent that it answers the property question.

What Is a Controllable Electronic Record?

A controllable electronic record is defined as “a record stored in an electronic medium that can be subjected to control under Section 12-105.” Section 12-102(a)(1) excludes records governed by other regimes, including electronic documents of title, electronic money, and investment property.³⁹ The definition has three components. A record is information stored in a medium and retrievable in perceivable form. Electronic encompasses any technology with electrical, digital, magnetic, wireless, optical, electromagnetic, or similar capabilities. With those two elements alone the category would be vast, reaching virtually all digital content; the third element is control, which delimits the category and gives it its character.⁴⁰

Control under Article 12 requires that a person hold three powers over the electronic record. The first is the power to avail oneself of substantially all the benefit from the record. The second is the power to prevent others from doing so; this is a factual question of exclusivity rather than an assessment of legal title, and the statute is careful to preserve it even where system-level protocols may modify the asset or where the power is shared, so that control remains workable for multi-signature arrangements and protocol-governed systems. The third is the exclusive power to transfer control to another person or cause that person to obtain control of another CER as a result of the transfer. The record, an associated record, or the system must also enable the person readily to identify itself as holding these powers, including through a cryptographic key. Shared powers qualify only within the limits prescribed by Section 12-105(b)–(c).⁴¹

What control captures is the defining characteristic of these novel assets, namely “their ability to be used and enjoyed without relying on intermediaries.”⁴² Bitcoin can be spent by its holder directly and is a CER; an email account or a photograph held on a social-media platform can be reached only through the provider and is not. This attribute distinguishes digital assets such as cryptocurrencies, stablecoins, and NFTs from the broader universe of electronic records; drawing that line is the property-classification work that Article 12 performs. Control receives its full doctrinal treatment on the companion research page on UCC Article 12.

Why “Control” Instead of “Possession”?

The 2022 Amendments built their framework around the new concept of control rather than extending the familiar concept of possession to digital assets. The choice was deliberate, and it rests on a considered judgment about what possession can and cannot do.⁴³

Since Roman law, and across the common-law tradition from Blackstone to Pollock and Holmes, possession has been understood to require physical dominion over a corporeal thing; the Restatement (Fourth) of Property states that intangible property is not subject to possession except perhaps in a metaphorical sense. Its doctrinal apparatus was developed and refined over centuries with tangible things in mind: what counts as sufficient possession, how possession transfers, what duties possessors owe, and how possession bears on priority among competing claimants. One response to digital assets came from the late Professor Moringiello, who maintained that possession had always been a flexible concept and that “tangibility as tangibility has no relevance to property law.”⁴⁴ The demonstration was historical. In Pierson v. Post, mortal wounding and continued pursuit counted as possession of a fox never physically grasped; in Ghen v. Rich, the custom of the Provincetown whaling industry made the killing of a whale a sufficient act of appropriation; in gift law, branding constituted delivery of cattle roaming an open range. Across these contexts, “courts focused on control and the ability to transfer, not on whether someone had physical dominion over an asset,” and on this view possession could be expanded once more to cover electronic assets.⁴⁵

The alternative view prevailed in the 2022 Amendments, and it departs from that solution while sharing its premise. Extending possession to electronic assets “would remove a foundational element of the doctrine as understood across legal traditions for over two millennia,” and the likely result would be an internal splitting of possession into one body of rules for tangible things and another for digital assets; “[t]he simplicity of extension would be purchased at the cost of complexity and legal uncertainty.”⁴⁶ The 2022 Amendments instead created a new and autonomous concept. Control draws on the core structure of possession as a state of fact and carries forward the same functional commitments of dominion, excludability, and transferability, while accommodating features that the law of possession was never designed to handle: exclusivity achieved through cryptographic keys, system protocols, or registration rather than physical dominion, and shared control through multi-signature arrangements, digital escrows, and protocol-mediated governance. The disagreement with Moringiello is not foundational; both positions reject tangibility as the touchstone and insist on the same functional attributes. The framework that emerged in Article 12 reflects a convergence of the two, and Moringiello helped to shape it as vice chair of the drafting committee.⁴⁷

The result is a single, technology-neutral concept that applies to any electronic record capable of being controlled, regardless of the underlying technology. Within the secured-transactions framework, the central policy choice of the 2022 Amendments is that “control of a CER is accorded similar legal significance and effects as the possession of a tangible good,” so that the rules governing attachment, perfection, and priority can operate for digital assets as they have long operated for goods.⁴⁸ Control is the functional successor to possession for assets that have no physical existence to possess.

How Do You Transfer Ownership of a Digital Asset Under American Law?

The 2022 Amendments accept that controllable electronic records are objects of commerce. They do not lay down a comprehensive code of transfer rules of the kind Article 2 supplies for the sale of goods; instead, they refer most questions to other applicable law and establish two cardinal tenets that apply across all voluntary transactions in a CER, whether by sale, lease, gift, the grant of a security interest, or any other transaction creating an interest in property.⁴⁹

The first tenet is the security of property principle, providing that “[a] purchaser of a [CER] acquires all rights . . . the transferor had or had power to transfer.” This principle is foundational to the conveyancing framework of the UCC. Its corollary is the shelter principle, under which a transferor with clear title can pass it on and shield the transferee from competing claims; its other face is a modern incarnation of nemo dat, since a transferee cannot acquire rights greater than those the transferor had.⁵⁰

The second tenet is the take-free rule, which “significantly moderates the rigidity of the security of property principle.”⁵¹ A person who “obtains control of [a CER] for value, in good faith, and without notice of [conflicting claims]” is a qualifying purchaser, and acquires the CER free of any property claim a third party may have, shielded against actions “whether . . . framed in conversion, replevin, constructive trust, equitable lien, or other theory.”⁵² The effect is to make CERs negotiable, in the manner of checks, promissory notes, and investment securities. If a thief misappropriates a CER and transfers control to a qualifying purchaser, the original owner cannot recover the asset and is left only with a damages claim against the thief, exactly as with a stolen negotiable instrument acquired by a holder in due course.⁵³ The combined effect of the two tenets is a transfer regime that “reduces title inquiry burdens, facilitates transactional certainty, and minimizes ownership disputes through its embrace of negotiability.”⁵⁴ Where the 2022 Amendments have not been enacted, transfers remain governed by the prior assignment-of-intangibles regime described above. The doctrinal detail is set out on the companion research page on UCC Article 12.

Can Digital Assets Be Used as Collateral in Secured Transactions?

Digital assets can serve as collateral, and the 2022 Amendments establish a dedicated regime for using CERs in that role. The overarching approach is to classify CERs as general intangibles under Article 9, subjecting them to the general framework of that article while adding asset-specific rules; the central policy choice is that control of a CER is given legal significance and effects similar to those that possession carries for a tangible good.⁵⁵

For attachment, there are two pathways. The first is the traditional route: a security agreement signed by the debtor and adequately describing the CER collateral. The second is an innovation. Where the secured party has acquired control of the CER pursuant to the debtor’s security agreement, control substitutes for the signed record. Both routes also require that value has been given and that the debtor has rights in the collateral or the power to transfer rights to a secured party.⁵⁶ Perfection likewise admits of two methods. A secured party may file a financing statement, as with any general intangible, or may instead perfect by taking control of the CER, directly or through another person acting on its behalf.⁵⁷

Control-based perfection “has substantial advantages”; it is “faster and more streamlined than filing” and “eliminates the jurisdictional complexities associated with determining the correct filing location based on the debtor’s physical presence,” “a particularly thorny issue in the context of digital assets.”⁵⁸ It also confers a decisive protection. A secured party that acquires control for value, in good faith, and without notice of competing claims becomes a qualifying purchaser, so that prior property claims to the CER cannot be asserted against its interest. This advantage is reinforced by the non-temporal priority rule of the new section 9-326A, under which a secured party that perfects by control has priority over a conflicting security interest held by a secured party that does not have control, even one perfected earlier by filing. That rule is an exception to the general first-to-file-or-perfect hierarchy of Article 9.⁵⁹ On floating liens over crypto held as inventory, see the essay Floating Liens Over Crypto-in-Commerce; on the full mechanics of the regime, the companion research page on UCC Article 12.

Can an NFT Legally Represent Ownership of an Off-Chain Asset?

In the ordinary case, an NFT carries no legal title to the thing it names. The 2022 Amendments take a deliberately restrained approach to tokenization. They “largely preserve the status quo with Article 12 explicitly stating that rights in property evidenced by a CER are governed by ‘law other than this article.’”⁶⁰ If an NFT is said to embody rights in a vintage car, a painting, or a digital image, that underlying asset remains subject to whatever law ordinarily governs it; issuing the token does not bring the car, painting, or image within the CER regime. The drafters recognized that comprehensively reconfiguring property rights across every category of asset would exceed the proper scope of commercial-law codification, and they left in place the numerus clausus constraint that defeated most pre-2022 tokenizations.⁶¹

The law supplies specific mechanisms for linking electronic records to rights in other property. Under Article 7, electronic documents of title can represent rights in goods in transit or storage. They remain governed by Article 7 and are expressly excluded from the definition of a CER, even though amended Section 7-106 supplies a control mechanism modeled on Article 12.⁶² Article 12 applies to two bespoke payment-rights categories. The controllable account builds on the Article 9 definition of an account, a right to payment for goods or services; the controllable payment intangible builds on the payment intangible. What makes each “controllable” is that it is evidenced by a CER and the account debtor has agreed to pay whoever has control of that CER.⁶³

These two categories benefit from the same take-free rule as CERs themselves, so that a qualifying purchaser takes the tokenized payment right free of competing claims; and the account debtor may agree not to assert claims or defenses against subsequent transferees. “[T]he 2022 Amendments create what are effectively electronic negotiable instruments by enabling payment obligations to be evidenced by CERs with comparable legal protections for good faith purchasers.” Any waiver of defenses remains subject to the conditions and exceptions in Section 9-403. This filled a gap long lamented by market participants, since the UCC had previously recognized the tokenization of payment obligations only in paper form, through negotiable instruments such as promissory notes and checks.⁶⁴ For the application of controllable accounts and controllable payment intangibles to claims trading, see Debt Tokens.

How Does American Law Compare to Other Jurisdictions on the Property Question?

Many jurisdictions recognize proprietary interests in digital assets, but the scope and doctrinal basis of that recognition differ. Where systems diverge is on the questions that follow: how to classify digital assets, and what rules govern their circulation. Common-law jurisdictions outside the United States have recognized digital assets as property but have remained entangled in a theoretical debate about their classification; this “preoccupation with classification has impeded the development of a coherent framework for how property law applies to these assets in commercial transactions.” The United States, by contrast, set the classification question aside and built a functional commercial-law regime.⁶⁵

In England and Wales, the courts had already recognized that cryptoassets can be the object of property rights, applying the attributes test associated with National Provincial Bank Ltd v. Ainsworth. The remaining question was classification, and the debate there ran between treating digital assets as things in action and recognizing a third category of personal property, a tertium quid for “data objects” existing independently of the law. That debate culminated in statute. After the Law Commission published its Digital Assets: Final Report in 2023, the Property (Digital Assets etc) Act 2025 confirmed that a thing is not prevented from being the object of personal property rights merely because it is neither a thing in possession nor a thing in action.⁶⁶ Other common-law jurisdictions reached the property-recognition conclusion through case law: New Zealand and Australia held that cryptocurrencies satisfy the Ainsworth indicia, Singapore extended the analysis to NFTs and to a stablecoin, and Hong Kong held that cryptocurrency can be held on trust. Canadian courts have been more cautious; in Cicada 137 LLC v. Medjedovic, the Ontario Superior Court of Justice declined to classify digital assets as property, reserving that question for future development.⁶⁷

Civil-law systems have engaged the question through their own categories, and they divide along the familiar fault line between the Francophone and the Germanic traditions. In France, the civil code organizes property around the capacious notion of bien, a category embracing incorporeal things. Within that framework, the Conseil d’État characterized Bitcoin as bien meuble incorporel (incorporeal movable property), and the Commercial Court of Nanterre applied the ownership rules of the Civil Code to award a borrower of bitcoins the bitcoin cash generated by a hard fork as its “fruit.”⁶⁸ The German Pandectist tradition is more restrictive. Under section 90 of the Bürgerliches Gesetzbuch, a “thing” (Sache) must be corporeal; cryptoassets therefore cannot be objects of ownership and are instead treated as intangible assets transferable by assignment.⁶⁹ Japan follows the Pandectist pattern and illustrates its consequences. In litigation arising from the collapse of the Mt. Gox exchange, the Tokyo District Court held that Bitcoin could not be the object of ownership under the Civil Code because it is neither corporeal nor subject to exclusive control; the legislature later addressed the resulting gap by granting exchange customers a statutory priority under the Payment Services Act.⁷⁰ At the regulatory level, the European Union’s Markets in Crypto-Assets Regulation defines a crypto-asset as a digital representation of value or rights, language that speaks of rights rather than of things.⁷¹

These differences in classification have practical consequences for the rights available to holders and the rules governing commercial circulation. The American approach is distinctive because Article 12 supplies a functional regime for transfers and tokenized payment rights, complemented by the collateral rules of Article 9. Its operation does not depend on resolving the inherited dichotomy between things in possession and things in action. Whether other jurisdictions should follow that approach remains contested.⁷²

Is There a Serious Argument That Digital Assets Are Not Property?

The most prominent academic dissent comes from Professor Robert Stevens of Oxford, who argues in Crypto Is Not Property (2023) and the follow-up Should Crypto Be Property? (2025) that cryptoassets are not property at all.⁷³

The argument rests on the distinction between rights and things. On this view, property is not the thing but the legally recognized right in relation to it; Stevens contends that the holder of a cryptoasset has no such right that the common law recognizes. The common law protects property in tangible things through the torts of trespass, conversion, and detinue, and each presupposes a physical thing capable of possession; the holder has no contractual right, because the system is designed to eliminate the intermediaries who might owe one; and there is no statutory right of the kind that defines intellectual property, nor any transferable privilege or immunity. On Stevens’s account, a cryptoasset is therefore valuable information rather than property. He concludes that cryptoassets are not property unless the purpose of a statute requires that they be treated as such.⁷⁴

Two considerations bound the reach of this dissent. First, Stevens writes from within English law and the wider common-law Anglosphere. The torts he canvasses are the English torts of trespass, conversion, and detinue, and his target is the digital-assets project of the English Law Commission and the Property (Digital Assets etc) Act 2025. He does not engage the American functional approach, the Kremen line of authority, or UCC Article 12. His is an influential voice within a particular, formalist strand of common-law property theory, not a statement of American law. Second, Stevens concedes in the 2025 essay that statutory recognition would present no conceptual difficulty, since the law could impose on everyone duties not to interfere with the holder’s exclusive ability to transfer a cryptoasset.⁷⁵

Stevens’s objection is that no jural relation has been identified for the holder of a cryptoasset; his concession is that a statute could supply one. Article 12 gives legal consequences to control, but control itself remains a factual condition rather than proof of ownership. Section 12-104(c) leaves the acquisition of rights to other law, while Section 12-104(e) protects a qualifying purchaser from competing property claims. The American response therefore rests on the interaction between ordinary property law and statutory transfer protections, rather than on an inference that cryptographic power alone creates a legal entitlement.⁷⁶

A mirror-image challenge comes from scholarship contending that code alone can generate ownership, positing that smart contracts can “transmute contractual promises into ‘de facto property rights’ through technical enforcement alone, thereby bypassing the numerus clausus principle.” Such accounts “fundamentally misconceive the nature of legal validity, erroneously assuming that technical mimicry of a property rule is equivalent to the normative investiture provided by a secondary rule of recognition.”⁷⁷ The code-based claim confuses technical power with legal entitlement. In the United States, Article 12 forms part of the applicable law alongside the property rules that it expressly leaves to other sources. The code-based challenge is examined in full on the companion research page on Code is Not Law.

What Does the American Approach Mean for the Future of Property Law?

The American treatment of digital assets offers a template for the continued integration of new intangible asset classes into the property framework. The regime is organized around the functional attributes of control, excludability, transferability, and negotiability, not around the placement of a candidate asset within the inherited dichotomy of choses in possession and choses in action. “Just as Article 12 accommodates digital assets by focusing on their functional attributes, this approach can be extended to emerging asset classes, such as carbon credits, that may appear distinct but pose analogous legal challenges.”⁷⁸ A planned companion research page will examine Verified Carbon Credits and Private Law.

American law answered the property question by supplying a functional commercial-law framework that gives market participants clear rules for transferring digital assets and using them as collateral, rather than by resolving whether digital assets belong to a third genus or to an expanded chose in action; the underlying jurisprudential questions remain largely open. This approach may lack the theoretical elegance of jurisdictions that have undertaken to classify digital assets within their property taxonomies, and what it supplies instead is legal certainty, conformity with the expectations of stakeholders, and room for innovation.⁷⁹ As these markets continue to evolve, other jurisdictions may find that “the American pragmatic functionalism, despite its theoretical untidiness, provides an effective model for adapting private law to emerging technologies in the digital age.”⁸⁰ What remains open is which functional attributes a novel intangible must exhibit before the machinery of Article 12 can be extended to it.