Blockchain Imaginaries and Their Metaphors: Organising Principles in Decentralised Digital Technologies

ABSTRACT Heralded as revolutionary in their potential to improve efficiency, transparency, and sustainability, blockchain technologies promise new forms of large-scale coordination between actors that do not necessarily trust each other. This paper examines blockchain imaginaries and associated metaphors. Our analysis focuses on bitcoin and ethereum, today’s most prominent blockchains that use the proof-of-work consensus mechanism. We identify three principles that organise blockchain imaginaries: substantial, morphological, and structural. These principles position blockchain as an enabler of economic, political and epistemological practices, respectively. Blockchain infrastructure and protocols rely on substantial metaphors (e.g. gold, gas) to govern resource allocation, morphological metaphors (e.g. work, trust) to generate consensus and structural metaphors (e.g. chain, transaction) to establish shared knowledge. Those imaginaries rely on metaphorical displacements of meaning that make blockchain technology relevant and intelligible while simultaneously shaping the direction of technological development and positing these technologies as new forms of economic, political and epistemological organisation. They are not merely descriptive but performative. We conclude by showing how these principles partially overlap with three symbolically generalized media: money, power and truth. Money organises scarcity within the economic system, power organises consensus within the political system, and truth organises knowledge within the science system.


Introduction
Blockchain is today's most advanced distributed ledger technology (DLT). DLTs are systems that guarantee the integrity of data scattered across remote machines. They differ from distributed databases since DLTs do not require a central administration guaranteeing the logical integration of data stored in different locations. In DLTs, data spreads across peer-to-peer networks, and logical integrationsynchronization and agreement over the ledger's state -is reached through consensus mechanisms. Blockchain was developed as bitcoin's infrastructure, sustaining the first decentralized digital currency, which gathers over 1 trillion US dollars in total market value. The appearance of exchange platforms made it simple to acquire and exchange cryptocurrencies, and promises of quick profits have spotlighted these assets. Bitcoin supports payments and value storage, offering an alternative to the banking system. Blockchain use cases increasingly extend beyond cryptocurrencies, spreading across journalism and news media, research and education, local and central government, archives, property records and asset management, health care, insurance, and the Internet of Things (IoT).
Metaphors affect both source and target domains: the displacement of signifiers makes complexity intelligible and bridges separate domains. The notion of boundary objects encapsulates how material and processual objects cross disciplinary borders, acting as interfaces . Their identity as objects common to different social worlds means there is no consensus over their meaning, although it may be specified within a local context (Star and Griesemer 1989). Boundary objects and concepts are loose and vague but provide cohesion between different spheres and fields (Löwy 1992), allowing the coordination of action without a consensus.
Boundary work coalesces into the programmatic design of technical objects, infrastructures, and standards, but also discourse production around technologies such as blockchain. This paper focuses on metaphors to account for the boundary work that allows blockchain to interface with other spheres of social practice. We followed actors, objects and concepts in webs of practice, which led us to study imaginaries and metaphors specific to different aspects of blockchain technology that work as practice orienting principles.

Blockchain Imaginaries
The adoption of new technologies does not rest solely on technical objects and forms of knowledge since such processes unfold across different political, economic, and cultural dimensions. Vidan and Lehdonvirta (2018) note that claims about the trustworthiness of blockchain technologies are based on technical elements, but such accounts do not suffice during controversies or breakdowns when the discursive construction of trustworthiness becomes more visible. In other words, the performance of trustlessness is not entirely technological -algorithms, networks, CPUs -but is also discursive. In their study of archival imaginaries, Woodall and Ringel (2020) analysed how blockchain figures in discourse about archives and how notions associated with archival work are present in the discourse about blockchain. The relation is bi-directional: archivist organisations strategically use blockchain to support their claims of reliability and archival concepts support claims of blockchain's trustworthiness. These discourses use vague definitions (e.g. blockchain as archive, place, and repository) and draw upon specific imaginaries to make the technology relevant beyond the communities associated with its initial use case application: cryptocurrencies.
Blockchain is a digital infrastructure that supports historical records. Power (2019) analyses blockchain as an exemplary case of infrastructures of traceability that result from different social processes. Such processes include concerns about authenticity, provenance, health, and sustainability that pervade the pharmaceutical and food industries, money and asset traceability requirements for legal purposes, and state border control for tracking bodies in movement. Traces are not simply inscriptions representing something absent: they can acquire an independent ontology with its own organisational dynamics (Power(2019)). Records and traces do not simply represent social realities but are constitutive of those realities. They are not neutral but are designed to generate effects in specific audiences (Van Maanen andPentland 1994, cited in Power 2019) or, as is increasingly the case, in machines. Thus, blockchains should also be considered from the angle of inscription devices. This opens up the possibility of understanding blockchain from the narrative perspective.
Inspired by the philosophers Searle (2006) and Ricoeur (1983Ricoeur ( , 1985Ricoeur ( , 1988, Reijers and Coeckelbergh (2018) approach blockchain as a narrative technology. They argue that, just like institutional facts, events recorded in the blockchain can be considered status functions declarations. These declarations (speech acts) have both locutionary aspects (declarations, propositional structures) and illocutionary aspects (articulations of desires and beliefs), generating new realities by reconfiguring rules, rights, and duties. If computation operates as a metaphor for substitution itself, blockchain operates as a metaphor for fixed, enduring inscription in digital form, allowing the digital equivalent of univocity and immutability. That abstraction relies on different metaphors, boundary concepts and objects. In this paper, we combine technical interrogation with an analysis of how social imaginaries of blockchain are grounded in meanings and practices. Our analysis is structured around three organising principles of said imaginaries, supported by different metaphors. We identify substantial, morphological and structural principles -relating to economic, political, and epistemological spheres. Blockchain is here conceived as a technology for governing resource allocation, generating consensus, and establishing shared knowledge.

The Origins of Blockchain
An unknown individual or collective using the name Nakamoto (2008) laid out the conceptual foundations of bitcoin and the first blockchain. 1 Swartz (2018) traces bitcoin's origins to the pursuit of digital cash by the cypherpunk and crypto anarchy subcultures. May was a prominent figure on the scenes and a proponent of the crypto anarchy ideology, which amounted to cryptographypowered anarcho-capitalist libertarianism. In The Crypto Anarchist Manifesto (1992) he claims computation, encryption and other protocols for authentication and verification allow anonymous interaction free from government regulation. May's ideology was expressed in his email signature: 'encryption, digital money, anonymous networks, digital pseudonyms, zero knowledge, reputations, information markets, black markets, collapse of governments'. Digital money sustained his idealizations of a cyberspace version of Galt's Gulch, the Randian paradise in Atlas Shrugged where people are free from government control (May 1993). Hughes, a leading member of the cypherpunk subculture, also emphasised the importance of systems that protect privacy, anonymity, and freedom of speech: We the Cypherpunks are dedicated to building anonymous systems. We are defending our privacy with cryptography, with anonymous mail forwarding systems, with digital signatures, and with electronic money. Cypherpunks write code. We know that someone has to write software to defend privacy, and since we can't get privacy unless we all do, we're going to write it (Hughes 1993, n.p).
These ideas were influential in the creation of cryptocurrencies. Bitcoin can be considered a translation of these ideologies. For Callon (1990, 143), the notion of translation helps understand how objects can be seen as a materialization of interests. Designers and innovators inscribe their visions and assumptions within the technical contents of new objects. Akrich uses the term inscription to denote how the design of technical objects articulates actors, tastes, competencies, motives, aspirations, and predictions 'that morality, technology, science, and economy will evolve in particular ways' (Akrich 1992, 208).
Bitcoin was the first digital currency to use blockchain but not the first to use cryptologic mechanisms. During the late '70s and early '80s, computer scientist David Chaum developed a distributed public record system that could be instantiated, maintained, and trusted by mutually distrustful parties. Chaum states that computerization is laying the foundation for a 'dossier society, in which individuals' lifestyles, habits, whereabouts, and associations' could be inferred from consumer transactions (Chaum 1985(Chaum , 1030. The paper presented an anonymous, cryptography-based transaction system that prevents state and corporate surveillance: the eCash electronic money system. Its centralized nature meant it depended on Chaum's company, and only a single US bank used eCash for about three years. However, it became an influential experiment -May (1993) mentions Chaum's digital money as a pillar of his crypto anarchist goals. In 1997, Adam Back invented hashcash, a proof-of-work algorithm used to mitigate denial-of-service and spam attacks. This algorithm required the performance of parametrisable computational work and was employed in email spam prevention. This strategy became known as proof-of-work (PoW), whose goal was to make it trivial for a system to send a single email but expensive to send many. Hashcash inspired the development of digital currencies, namely Finney's reusable proof-of-work (RPOW), created in 2004. RPOW was based on hashcash tokens (whose value reflected the difficulty involved in their calculation) that could be reused without repeating the work required to generate them. Despite the adherence to security, transparency, and verifiability principles, the project depended on secure but centralized servers and was never adopted in economic activities.
The original bitcoin paper (Nakamoto 2008) built upon these works, detailing a solution for a problem in distributed computing systems known as the Byzantine Agreement Problem. This problem refers to a theoretical state of distributed computing systems in which individual components may malfunction and information about errors is imperfect -(i.e. Byzantine) faults may present different symptoms to different observers. When lacking centralized control, distributed systems can store different versions of data due to errors or malicious activity. A Byzantine fault-tolerant system is a distributed system that can perform its task as long as the majority of its components do not malfunction. Unlike RPOW, which relied on centralised, secure hardware, there are no formal barriers to participating in bitcoin's blockchain infrastructure (ensuring its consistency, completion, and immutability), requiring only the deployment of a network node. Nakamoto's bitcoin decentralized protocol supported what became known as permissionless blockchain networks, which present no formal restrictions to deploying nodes and participating in maintaining the consistency, completion, and immutability of records. This protocol transparently tracked timestamped coin transfer records in a publicly accessible ledger, solving the so-called double-spending problem, which had hitherto only been tackled by centralized authorities governing digital transactional record systems.
The following sections of this paper explore the three organising principles we identify in blockchain imaginaries by analysing today's two most popular blockchains: bitcoin and ethereum. Our focus is on proof-of-work blockchain technology. While other consensus mechanisms (e.g. proof-of-stake) rely on different metaphors, they fulfil similar functions to those described in the 'Morphological principle' section.

Substantial Principle
Substances and materials have politics, and new substances, whether they are invented or discovered -or actual substances, for that matter -are no different. In this section, we approach Blockchain as a substance-bearing technology. The substantial principle in blockchain imaginaries represents technological affordances by using familiar materials as metaphors, which underpin their capacity to govern scarce resource allocation.

Digital Gold
In the original bitcoin paper, Nakamoto (2008) envisioned a digital currency that would work like gold in several ways. First, it sought to be universally valuable, without usage restrictions: anyone could use it anywhere and for whatever purposes. Second, it would store value: holding the cryptocurrency would, like holding gold, work as a hedge against inflation and currency devaluation. Although gold holds use value beyond exchange value (e.g. jewellery, electronics, and medical devices), the value of both bitcoin and gold largely depends on how much people are willing to pay. Finally, both bitcoins and gold are 'mined'. Mining is the name of the process through which the protocol approves and verifies transactions, appends blocks of transactions to the blockchain, and creates new bitcoins. The creation of bitcoins is based on a 'steady addition of a constant . . . amount of new coins [that] is analogous to gold miners expending resources to add gold to circulation' (Nakamoto 2008, 4). Nodes performing the proof-of-work involved in network maintenance are called miners. Miners are rewarded newly created bitcoins, similar to seigniorage revenue (the difference between the value of newly created currency and the cost to produce it), plus the fees involved in each transaction. In addition, the bitcoin protocol limits the total amount of bitcoins and will stop producing new coins once 21 million are in circulation. The analogies with gold are made quite explicit by Nakamoto himself: indeed there is nobody to act as central bank or federal reserve to adjust the money supply as the population of users grows . . . it's more typical of a precious metal. Instead of the supply changing to keep the value the same, the supply is predetermined and the value changes (Nakamoto 2009, n.p.).
The digital gold metaphor underpins blockchain proof-of-work consensus mechanisms while simultaneously sustaining the trustworthiness and acceptance of bitcoin for economic purposes.

Ether and Gas
Bitcoin was developed for storing value and making payments, but developers quickly started creating new applications. One of these developers was Vitalik Buterin, who conceived ethereum in 2013. Ethereum is the second most valuable cryptocurrency in market price, after bitcoin. Before founding ethereum, Buterin contributed to mastercoin (currently Omni Layer), an overlay protocol that added functional higher-level layers to the bitcoin blockchain network. In a paper entitled The Second Bitcoin Whitepaper, a developer called J. R. Willett (2012) presented the concept of an abstraction layer that encodes advanced transaction types within tiny bitcoin transactions. It supports the registry of assets, the creation of new currencies, and smart contracts. Smart contracts are autonomous programs that execute, monitor, or document the terms of a contract or an agreement. A smart contract can automate buying a currency once it reaches a given price, creating escrow agreements (setting collateral for, e.g. providing protection for car sharing with strangers) or paying royalties to artists each time their song is played. Bitcoin's scripting language supports basic smart contract functionalities like multi-signature accounts, payment channels, escrows, and time locks. Mastercoin sought to expand these possibilities.
Buterin thought mastercoin's approach was unstructured, focusing on discrete features and rules. He wrote a proposal to improve it, entitled Ultimate Scripting, following an open-ended specification of data and operations. The proposal was never adopted, and Buterin then decided to create ethereum. Like bitcoin, ethereum is a permissionless blockchain network. Unlike bitcoin, ethereum envisioned general-purpose computation, adding further levels of abstraction to the blockchain. The Ethereum Virtual Machine (or EVM), a Turing complete decentralised computing platform, lies at its core. Turing completeness refers to an abstract machine able to perform any calculable (computable) task -given enough memory. The EVM executes algorithms written in specific programming languages, supporting decentralised applications. These programs are stored in the public ledger while the parties of the agreements remain anonymous. Buterin's motivation behind the name choice reveals differences between the substantial imaginaries of ethereum and bitcoin: I was browsing a list of elements from science fiction on Wikipedia when I came across the name . . . . I suppose it was the fact that sounded nice and it had the word 'ether', referring to the hypothetical invisible medium that permeates the universe and allows light to travel (Buterin 2014, n.p.).
In contrast to gold and its inert properties, ether is a substantial metaphor that supports the imagination of an active medium that is not inscrutable. The substance represents EVM's ability to execute terms of 'animated' agreements, which are automatically enforced.
In addition to ether, another imaginary substance plays a central role in ethereum. As we have argued, Turing completeness is related to flexibility and abstraction. In computational terms, it refers to functions such as looping and branching (e.g. while and if statements), and the ability to manipulate data. Ethereum's flexibility allows sophisticated smart contract logic. However, generalpurpose computational devices such as the EVM tend to be vulnerable to problematic code, such as infinite loops, which could break or overburden the system. In order to prevent malfunction and attacks, computation in ethereum has a cost. Computational cost is not quoted in ether (the ethereum cryptocurrency) but in gas: a unit for measuring the computational effort of executing on-chain operations such as transactions, simple arithmetic, and checking account balances. The gas fees are quoted in ether and are determined by the demand for network resources and ether market price. Quoting computation prices in gas rather than ether allows stable prices for computation (e.g. an operation will always have the same gas price). It protects the cost of computation from the volatility of the ether market price. Gas fees also affect the speed of program execution. Miner nodes will privilege higher gas priority fees when selecting pending operations. Gas serves the function of allocating scarce resources for the computational steps involved in transfers and smart contracts. It is the ethereum ecosystem's inherent funding mechanism for its required computational power.

Digital Scarcity in the Blockchain
Blockchain technologies enable the emergence of markets and decentralized digital currencies by inscribing scarcity in digital objects and networks. Scarcity at the level of cryptographic tokens is, in the last instance, guaranteed by incentive mechanisms that ensure an abundance of expensive computational hardware and energy expenditure. Bitcoin mining and ethereum's gas are strategies to provide the required incentive structures for the availability of distributed computation power. The first metaphor underpinning substantial imaginaries of blockchain, digital gold, supports the cryptocurrency monetary theory. Maurer and others (2013) highlight the deliberate intention of the mining metaphor and claim bitcoin materialised a digital metallism imaginary: the reference to monetary systems based on precious metals may have contributed to its acceptance. Bitcoin materialised earlier crypto anarchist dreams of stateless cryptological money, but its initial niche adoption was followed by the arrival of expensive specialised hardware and speculative investment.
There are debates on whether cryptocurrencies truly achieve the status of digital gold or money. The economist Krugman (2011) claimed that bitcoin created its very own gold standard by fixing the money supply and the total amount of currency in circulation. However, this claim assumes that the software system has inherent value (Golumbia 2016). The conception of commodity money, which animates cyberlibertarians, crypto anarchists and Nakamoto, dates back to the works of 18th and 19th century economists Adam Smith and Carl Menger. They considered that the division of labour led to the necessity of swapping goods and to the accumulation of items that were generally desirable -these goods eventually became money. The mythical idea that a return to real money based on free trade will free individuals from state oppression underpins cyberlibertarian and crypto anarchist ideologies. Anthropological works have debunked this myth. Graeber (2011) observes how Menger's theory of money is based on the assumption that money had its origins in marketplaces where various products were available for direct barter. Graeber agrees with Humphrey (1985), who claims there is no ethnographic evidence of a pure barter economy in societies that did not use money, much less of the emergence of money from barter. This foundational myth resonates with libertarian views widely spread within the cryptocurrency community, in which good money (i.e. freed from state interference) makes capitalism work for everyone.
However, ethereum's gas indicates how substantial imaginaries and material metaphors in blockchain imaginaries go beyond metallism. This second metaphor indicates the potential of blockchains for creating mechanisms for designing markets. Indeed, market design is one of the most prominent applications of blockchain technology. Computation mediates relations and animates their elements through its generative and performative capabilities. Blockchain-based market design explores such capabilities to grant autonomy to economic agents and create business models based on automation and decentralised contractual relations. All these possibilities are enabled by the resource allocating function sustained by blockchain imaginaries' substantial principle.

Morphological Principle
Blockchain morphological imaginaries and metaphors relate to network forms and dynamics. Its distributed relational model points to the gregarious quality of blockchain infrastructure, operating as a mixture between cooperation and competition. Our focus in this paper is the proof-of-work consensus mechanism. Even though blockchain morphology refers to a topology of networked computers, this morphological principle is associated with metaphors that conflate such technologies with familiar elements of our everyday social practices. The morphological organising principle within blockchain imaginaries refers to consensus generating socio-political dynamics.

Work
The Nakamoto Consensus protocol supports permissionless blockchains: distributed peer-to-peer networks without participation restrictions that are Byzantine Fault Tolerant. It guarantees the authenticity of data and enables a network of mutually distrusting machines -where anonymous nodes may join and leave at will -to reach an agreement. Each node stores a replica of the ledger information required to verify incoming transactions. Nakamoto's achievement articulates peer-to-peer networks, a politics of mutual suspicion and incentivization via a proof-of-work consensus mechanism (mining). Consensus is achieved through the decentralized mining process.
Mining is done by repeatedly applying a cryptographic hash function to the block's header, changing one of its attributes called the nonce (a numerical value) until its output satisfies the target hash (determined by the variable parameter difficulty, which is proportionate to the overall computing power in the network). Hashes are often compared to fingerprints, as hash functions can be applied to data objects of any size and deterministically output a fixed-size string (known as a hash). Block hashing supports the proof-of-work algorithm and the block linkage function for validation (forming a proof-of-work chain). Hashes grant blockchain its quality as a growing list of blocks of records (in an append-only manner, hence its immutability). Mining nodes perform work as they compete to be the first to gather transactions, check their validity and solve the proof-of-work cryptographic puzzle. Miners with greater computational power -more expensive hardware and higher electricity expenditure -have a higher hashrate (hashes/second) and thus better chances of being the first to find a solution. The winner adds a verified block to the chain, being rewarded with newly created bitcoins and the block's transaction fees. Thus, proof-of-work makes participation costly and rewards honest nodes, discouraging malicious activity.

Trust and Consensus
Cryptographic hashing functions involved in proof-of-work consensus algorithms are meant to generate 'trust' in the blockchain content. These functions output a fixed-sized string that can be used as a concise way to represent and verify data. While mining a successful hash when appending a block to the chain is computationally difficult, verifying that it matches the data of that block is not. The result is 'bitcoin's chief technical achievement -a near-immutable, fully consensual record without a central record keeper' (MacKenzie 2019, n.p.). Nakamoto himself disclosed his motivation behind the creation of bitcoin. Writing in the P2P Foundation web forum, Nakamoto (2009) claims that trust in central banks is often breached: their ability to produce money leads to inflation and currency debasement. Nakamoto's critique also targets commercial banks -which people trust to hold their money, execute transfer orders, respect privacy, and offer protection from identity theftfor their role in credit bubbles and their expensive overhead costs. Finally, Nakamoto claims that trust in a central authority is why previous efforts to create digital currencies failed. The Nakamoto consensus protocol is a solution for maintaining trust in a peer-to-peer network of mutually suspicious participants. It bypasses the need for trust, allowing agreement in what became known as trustless environments. Transparency is a central component in such environments. As Power (2019) suggests, traceability draws programmatic power from the transparency myth of modernity and it is a mode of operationalizing that myth. Permissionless blockchain infrastructures are a specific form of operationalizing transparency via traceability: a visible public record shared by a network of mutually suspicious machines engaging in horizontal surveillance.
In Nakamoto's vision, the reliance on institutions and authorities is not to be replaced with personal connections, which is the rationale behind the web of trust model. A web of trust is a system set up by cryptographic communities to verify whether a public key (used in encrypted email communication, for example) truly belongs to someone. It is an alternative to centralized trust authorities known as public key infrastructures (PKIs), built on the assumption that people physically meet to verify their key fingerprints. Any user can sign trusted keys, hence vouching that the keys belong to the owner of that address and not an attacker impersonating that person. According to Vidan and Lehdonvirta (2018), Nakamoto's ideology can be characterized as techno-utopian, seeking to replace trust in individuals and institutions with trust in code. 'Through code', write Vidan and Lehdonvirta, 'Nakamoto expects to reap the benefits of a modern impersonal economy without the cost of having to rely on a centralized power to regulate it' (Vidan and Lehdonvirta 2018, 43).

The Blockchain Democratic Lifeworld
The morphological principle organising blockchain imaginaries is that of an open, permissionless network governed by the notion of consensus. Nakamoto's vision may be seen as a sociotechnical intervention into intersubjectivity, at least in a simplified form. This intuitive approach to intersubjectivity is present in the phenomenological sociology of Schutz (1964), who approaches intersubjectivity as the product of a shared, temporally situated common orientation. The intersubjective synchronization of internal streams of consciousness requires a shared orientation towards an external object. With its temporality or duration, such shared orientation enables the we relation that emerges in practice and characterizes intersubjective processes. Schutz uses the example of two people observing a bird's flight: since we are growing older together during the flight of the bird, and since I have evidence, in my own observations, that you were paying attention to the same event, I may say that we saw a bird in flight (Schutz 1964, 25).
International social forms can be approached from the perspective of intersubjective relations. In a free scaling peer-to-peer network, however, these relations enter cycles of amplification that project them with unprecedented scale, speed, and geographical scope. The Foreign Exchange Market (FOREX) analysis by Knorr Cetina and Bruegger grounds this market's instantiation as a global social form in the face-to-screen situation micto-sociological setting, which is enabled by telecopresence (response presence). They conclude that traders 'provide for the market's existence and process continuity through the intensity of their communication with one another', in 'a large, globally distributed conversation' (Cetina and Bruegger 2002, 914). The network analogy and its relational vocabulary may not be enough, writes Knorr Cetina, who invites us to 'think in terms of reflexive mechanisms of observation and projection': scopic systems that 'collect and focus activities, interests and events, and project them in identical fashion to dispersed audiences', acting as a 'centering and mediating device through which things become assembled and from which they are projected forward' (Cetina 2005, 220-1). Furthermore, forms of social organisation around networked digital technologies and their performative possibilities can be integrated into the systems (Cetina 2003, 8).
Blockchain mimics such intersubjective dynamics in a distributed computational environment. While the centrality of the shared witnessing of an event (transparency, accessible records) remains, that logic is abstracted and integrated into decentralized consensus protocols. The blockchain replaces the centrality of the gaze -which Knorr Cetina associates with scopic systems and screenswith decentralized validation and self-executing code. In doing so, they perform similar mediation roles as enablers of global social forms. Blockchains collect and articulate activities, interests and events, projecting them forward in an identical fashion to a distributed network of machines. As a distributed digital infrastructure, the blockchain acts as a reflexive system based on decentralised records. While the inscribed content's locutionary aspects are interpreted and animated in computational processes, its illocutionary aspects organise material practices.
Blockchain presents itself as a new arena for power struggles (and profit extraction). The arrival of big investments in cryptocurrencies and hardware tightened the interconnections between blockchain and financial institutions. Blockchains have been built to replace rational-legal bureaucratic institutions. They mimic the operation of both markets and states, but their control is displaced from those centres of power to technical elites. The most relevant aspect of blockchain morphological imaginaries is its ability to generate consensus without formal barriers to participation. In this sense, they are sociopolitical machines. Nakamoto's consensus is modelled after simple democratic governance arithmetic. The consensus protocol can be gamed if attackers decide to pay the costs associated with 'dishonest' participation to gain control of over 50% of a blockchain's computational power. This is known as a 51% attack, or majority attack, allowing attackers to block transactions or reverse their own transactions (enabling double-spending). This aspect relates to what Swartz (2018) terms infrastructural mutualism, opposed to what Maurer and others (2013) call digital metallism. For Swartz, infrastructural mutualism in bitcoin points to a cooperativist view of the function of money liberated from intermediaries who control and survey exchange and society. However, this aspect goes beyond cryptocurrencies, as illustrated by the increasingly diverse blockchain-powered applications. The peer-to-peer organisational structures known as decentralized autonomous organisations (DAOs), often conceived as cooperatives encoded in blockchains and smart contracts, are paradigmatic of the sociopolitical implications of decentralized technologies.

Structural Principle
References to the structure of the digital objects in blockchain make this another locus for imaginaries and metaphors shaping this technology. This section will explore the data structures in blockchain as infrastructural affordances. These are the traces that blockchain as an infrastructure produces and maintains. The structural principle of blockchain imaginaries enables the establishment of common knowledge.

Block and Chain
Blockchain is a distributed computational system that relies on a given data structure: linked blocks. Each block includes a header and a body containing executed transactions data (inputs, outputs, and values). Headers are unique to each block but follow a similar structure, which varies between different blockchain protocols. The term chain refers to cryptologic linkage: each new block's header contains the previous block header's cryptographic hash. As we have seen, appending new blocks (mining) is dependent on the proof-of-work consensus algorithm. The result is known as a proof-ofwork chain. Hash functions also support another blockchain header field: the root or top hash of a hash tree, which is used to validate the block's transaction data. These trees represent a block's transaction data via multiple hierarchical hashes, allowing efficient and secure verification. They allow nodes to verify new transactions without hosting a complete copy of every transaction in the proof-of-work chain. Header hashes and hash trees map data to concise outputs that are easy to store, transmit and verify. Together, they allow independent confirmation of on-chain data integrity, all the way back to the genesis block.

Key, Identity, Wallet, Transaction, Coin
Control over blockchain digital assets makes use of public-key cryptography. A public key asymmetric cryptographic system relies on a pair of keys: the public key, which is meant to be disseminated, and the private key, which should be accessible solely to its owner. A private key is simply a massive random number so large it is practically impossible to be duplicated by random number generators. A public key is deterministically derivated from the private key via unidirectional cryptographic trapdoor functions. A private key always generates the same public key, which uniquely corresponds to, but cannot be used to derive, the original private key. Such systems allow the encryption of messages and the ability to digitally sign them without the need for a shared cryptographic key. A sender can add cryptographic signatures to messages using the private key and anyone can verify the origin of the message using the sender's public key. Signatures also ensure message validity and integrity -they can attest that it has not been tampered with after it has been signed. Bitcoin and ethereum use the Elliptic Curve Digital Signature Algorithm (ECDSA) for authentication and identity management.
Individuals interact with the system via addresses in a permissionless blockchain like bitcoin and ethereum. Since no information needs to be recorded in the blockchain, one can easily create many account addresses by generating new cryptographic keys on a personal computer without any associated costs or even an internet connection. Despite the importance of key generation, storage, and usage, keys are usually operated by exchanges and wallet software, away from the user's eyes.
Private keys act as credentials to manage blockchain digital assets. The transactions themselves are simply digitally signed valid protocol messages. Blockchain technologies use this technique to ensure that only the owner of digital assets has control over them.
A bitcoin is not a single object, a discrete thing, but the history of signed transactions that lead back to the genesis block's 50 bitcoins, mined by Nakamoto himself, or to posterior mining rewards (MacKenzie 2019). The digital object supporting bitcoin transactions is the unspent transaction output (UTXO). UTXOs are both the input and the output of a transaction in bitcoin and other cryptocurrencies. The UTXO model associates the remaining numerical output of previous transactions with an address, whose total balance is the sum of what remains unspent in the variable number of UTXOs. The model is often compared to receiving change after using cash: the coins and bills received are the unspent outputs of past transactions, and the total money a wallet carries is the sum of those coins and bills. The address, wallet, transaction, and coin metaphors are ways to make intelligible the way blockchain employs timestamped records and public-key cryptography in its data structures and protocol dynamics.

Cryptological Positivism
Cryptographic signatures and hashes support the blockchain's operation as an infrastructure of traceability. For Power (2019), traceability is an imagined form of knowledge, inspiring infrastructures that produce precise and granular traces. Cryptological knowledge governs blockchain's mode of operationalising transparency and traceability. The structural principle of blockchain imaginaries, represented in the archetypical data structure of bitcoin, is governed by linked cryptographic proofs. The function of blockchain data, like that of organisational records, is never simply to record what happened: they are designed to produce an effect in some kind of audience (Van Maanen and Pentland 1994;cited in Power 2019) or, as in the case of blockchain, in peer-to-peer networked computing infrastructures. As an archive, blockchain records textual traces: linguistic, propositional utterances, such as records, transactions, and headers. Blockchain data fields work as status function declarations, whose locutionary aspect is mostly meant to be interpreted by machines. However, illocutionary aspects point to decentralized multilayered verifiability (e.g. using balance checks, hash trees, and the single shared history of a growing list of blocks using hashing) which articulate people's desires and beliefs in vast sociotechnical networks of material practices.
Blockchain is an organisational technology, and its structural principle refers to cumulative computationally verifiable and easily communicable (concise) 'truths' or 'facts' that are constructed in the exchange. This aspect allows for trust in the propagation and storage across every node of the decentralized network. Blockchain-based global sociotechnical assemblages rely on both the vast solution spaces for cryptographic hashing functions and the materiality of available hardware and power grids involved in mining. They also depend on rational models articulating participants' choices, interests, resources, desires and beliefs. As inscription devices, blockchains allow fixation through what could be considered a distributed set of mirrors and lenses, sensing and projecting forward an algorithmically and mathematically sound reality. Thus, the structural organising principle of blockchain imaginaries points to an epistemological stance we call cryptological positivism.

Conclusion
This paper identified three organising principles behind the meanings and imaginaries surrounding blockchain, and described their metaphorical underpinnings. Blockchain's complexity and noveltytying together knowledge from cryptography, distributed system and economics -explain the identified principles' importance as structuring techno-mediated visions. Our analysis sought to illustrate how those principles are not merely descriptive but performative, guiding technologists, decision-makers, advocates, and individual adopters.
Blockchain imaginaries and metaphors organise social practices around beliefs and expectations surrounding those technologies. Substantial blockchain imaginaries operate at the level of economic value by creating artificial scarcity. Scarcity allows the development of new markets with inbuilt incentive systems, attracting the necessary networked computational power (hardware and electricity) for decentralized social exchange. Although this aspect is related to what became known as 'digital metallism', our analysis indicates that the substantial principle in blockchain imaginaries goes beyond metallism. Morphological blockchain imaginaries refer to the form of blockchain-based social exchange dynamics. They underpin large-scale imagined anarcho-utopian communities, free from traditional social institutions, pre-existing relations, and trust. These imaginaries support visions of cooperation and competition that take networked computational technology to their conceptual and practical limits (e.g. the amount of computation required for proof-of-work). The morphological principle in blockchain imaginaries is deeply connected with what researchers have termed 'infrastructural mutualism'. The structural imaginaries of blockchain are supported by blockchain data structures and refer to computationally verifiable and communicable commonly held 'facts'. These 'facts' are declarations about and produced in the blockchain-supported exchange. The structural principle of blockchain is structured around 'cryptological positivism.' It refers to knowledge production and circulation, as well as cycles of validation, reproduction and amplification. Terms like gold, gas, addresses, coins, consensus, trust, and work are operationalized as boundary concepts and objects that take a concrete form in blockchain technology design. The organising principles of blockchain imaginaries and their metaphors allow the bridging of social practice and technological development. However, they are interdependent, like the Borromean ring, and each of these principles reflect the other two.
Blockchain as a cultural technique further intertwines humans with digital technoscientific mediations in new forms of social organisation. Interesting questions for further research can be derived from our analysis: how blockchain imaginaries address multiple audiences and how these technologies differ from pre-existing media. Our analysis points to a possible connection, which was left unexplored, between the identified principles and social construction processes in a broad sense of the term -specifically, the creation and maintenance of social entities. However, from a systems theory perspective, the three imaginaries seem related to crucial social elements of differentiation and integration theories. Luhmann suggests that the autonomy of sub-systems is achieved through specific codes in their internal communication's symbolic media, which distinguishes them from their environment. However, he associates functionally differentiated societies with generalized symbolic media. Luhmann (1995) claims systems differentiation results from the contingency of new media coming into being. The expanding number and the enlarged scope of projects surrounding blockchain and DLTs indicate an aspiration to universalism. From the Luhmannian systems theory perspective, it can be argued that specific sub-systems 'use symbolically generalized media such as money, power or truth as their code, which gives these systems an inherently universalizing logic' (Bongaerts 2008;cited in Weiß 2020). For Luhmann money is the medium through which the economic system deals with scarcity, power is the medium through which political system deals with consensus, and truth is the medium through which the science system deals with knowledge. Blockchain substances enable forms of money, its morphology represents an arena for power and politics, and its structure points to the social computation of commonly held 'truth'. Blockchain technology expands computation's universalising logic by laying claim to all those elements, and its protocol was designed to be their technical infrastructure.

Notes
1. The word "blockchain" was not used in the paper, which refers to a chain of blocks (Nakamoto 2008, 7). There are claims that, as a cryptographic solution based on verifiable links between historical records, blockchain precedes bitcoin. However, the decentralization principles that characterize blockchain were not present in these solutions.