The law presents two faces to the world. One is the face of an ideal: a clear, common framework that allows society to function, a set of rules accessible to all, ensuring fairness and predictability in human affairs. This is the law as a public good, the bedrock of the social contract. The other face is the one most people encounter in reality: an intimidating edifice of dense, convoluted language, a cryptic domain understandable only by a small cadre of specialists. This is the law as a private language, a fortress of jargon that excludes the very people it is meant to govern. For centuries, these two faces have existed in an uneasy tension. The law has always been a form of "code"—a symbolic system for structuring human behavior—but one written in a language that, for most of history, few could read.
This chapter traces the long and often winding journey of that code. It is a story of translation, a perpetual effort to render the abstract principles of social order into a more concrete and unambiguous form. This journey began with the translation of spoken custom into written edict, progressed through the turbulent translation from one human language to another in the wake of conquest, and has now arrived at a remarkable threshold: the translation of human language into machine language. This latest phase, driven by the rise of computation, programmable logic, and artificial intelligence, promises something revolutionary. It offers the possibility of creating a legal code that is not only precise and systematic but also executable, explainable, and ultimately, more accessible than ever before. It is a quest to finally resolve the fundamental paradox of the law, to merge its two faces, and in doing so, to reshape our understanding of governance, community, and the rules that bind us together.
The complexity of legal language is not a modern invention, nor is it merely the result of poor writing. It is a deep-rooted cultural artifact, a linguistic tapestry woven from the threads of history, professional habit, and the very psychology of authority. To understand the push toward a future of programmable law, one must first appreciate the tangled legacy of the language it seeks to replace. This language, often dismissed as "legalese," is the product of centuries of accretion, where historical accident was hardened into professional necessity.
The story of modern legal English begins not with clarity, but with conquest. Before 1066, the law of the Anglo-Saxons was conducted largely in Old English, the vernacular of the people.1 The Norman Conquest, however, shattered this linguistic unity. The new ruling class, led by William the Conqueror, spoke Anglo-Norman French, which swiftly became the language of power, administration, and the courts.1 English was relegated to the language of the conquered lower classes.5 For formal written records, such as statutes, the scholarly language of Latin was retained, creating a complex and often confusing trilingual legal environment.2
In this chaotic linguistic landscape, a practical problem arose: how could a lawyer ensure that a legal document's meaning was understood and would hold up in court, regardless of whether it was read by an English-speaking clerk, a French-speaking judge, or recorded in Latin? The solution was a form of redundancy. Lawyers began using "doublets" and "triplets"—groupings of synonyms or near-synonyms drawn from the different languages. Phrases like "last will and testament" (English, French, Latin roots), "breaking and entering" (English, French), and "free and clear" (English, French) became common.1 A drafter might use the term "give, devise and bequeath" to cover all possible linguistic and legal bases.5 What began as a sensible strategy to avoid ambiguity in a multilingual world soon became a stylistic habit, adding layers of verbosity that would persist for centuries.
This linguistic foundation was then set in stone by the nature of the English common law system. Unlike the civil law tradition of continental Europe, which relied on comprehensive legal codes, the common law evolved through precedent—the decisions made by judges in previous cases.1 Legal certainty depended on using the exact words and phrases that had been tested and validated in court, sometimes decades or even centuries earlier.1 Lawyers became understandably conservative, loath to alter a single word of a time-tested formula for fear of introducing an unforeseen loophole or ambiguity.7 This created a powerful inertia, a "linguistic fossilisation" where antiquated terms like "aforesaid" and "witnesseth," and complex sentence structures were preserved not out of a desire to confuse, but out of a professional aversion to risk.1
Over time, this complexity acquired a new, more subtle function. It began to act as a signal of authority. Researchers have proposed a "magic spell" hypothesis to explain the persistence of convoluted legal language.8 Much like a ritual incantation, the purpose of legal language is often "performative"—it does not simply describe the world, but actively changes it. A judicial ruling, a deed of property, or a marriage certificate alters the legal status of people and things.8 According to this view, difficult linguistic structures and archaic jargon serve as a marker, a cue that the text is not mere description but an act of power. The very impenetrability of the language lends it an air of gravitas and authority, much as a magic spell might use archaic roots or rhyme to set it apart from ordinary speech.8
This association between complexity and authority is so deeply ingrained that it has created a self-perpetuating cycle. Studies have shown that even non-lawyers, when asked to draft a contract, will instinctively resort to legalese.7 They do so out of a belief that the document needs to "sound official" to be taken seriously and a fear that simple, direct language might be less robust or legally binding.7 As more people encounter and replicate this style, it becomes further entrenched as the perceived "correct" way to write a legal document, making it increasingly difficult to break the tradition.7
For centuries, this state of affairs was largely accepted. However, beginning in the 20th century, a significant counter-current emerged: the Plain English Movement. This was a loosely organized campaign, driven by consumer advocates and legal reformers, to encourage writers to communicate clearly and concisely, without legalese, while preserving accuracy.10 It grew from the notion that people should be able to understand the documents that govern their lives, from credit agreements to insurance policies.5 This movement gained significant traction, leading to milestones like U.S. President’s declaration that federal regulations should be written in "layman's terms" and, more formally, the Plain Writing Act of 2010, which requires U.S. federal agencies to use clear communication that the public can understand and use.11
The long-running battle between the dense fog of legalese and the push for plain language is more than a debate over style. It represents a fundamental conflict over the nature of legal authority itself. Does the law derive its power from the specialized knowledge of an exclusive priesthood, whose authority is signaled by a cryptic and performative language? Or does its legitimacy spring from the informed consent of the governed, which requires a language that is broadly understood? This unresolved tension between exclusivity and accessibility set the stage for a technological intervention, one that promised to deliver both the precision lawyers crave and the clarity the public deserves.
The desire to make law clear, systematic, and unambiguous is not new. Long before the first computer, civilizations have been driven by a powerful impulse to codify their rules. This ancient quest to create a perfect, comprehensive statement of law is the direct intellectual ancestor of the modern concept of programmable law. It represents a continuous historical thread, moving from the act of writing rules down to the act of making them executable.
This drive for written codification is a recurring theme in human history. The oldest known surviving civil code, the Code of Ur-Nammu, dates back over 4,000 years in ancient Sumer.15 The most famous ancient example, the Babylonian Code of Hammurabi, sought to establish a systematic and written set of laws for all to see.6 Later, the Roman Empire produced monumental works like the Law of the Twelve Tables and, much later, the Corpus Juris Civilis, or Code of Justinian.6 In the modern era, this movement culminated in the great national codifications of the 18th and 19th centuries, most notably the Napoleonic Code of 1804, which aimed to create a rational, unified legal system and has served as a model for civil law countries around the world.6 The goal in every case was the same: to replace the ambiguity of unwritten custom and the scattered nature of judicial precedent with a single, authoritative, and logically organized text.
Today, this ancient dream of a perfectly systematic code has found a new expression in the digital realm. The modern inheritor of this tradition is the "smart contract," a concept that takes codification to its logical conclusion. A smart contract is best understood not as a legal document in the traditional sense, but as a piece of computer code that automatically executes all or part of an agreement.16 This code is typically stored on a blockchain, a decentralized and immutable digital ledger, which ensures that once the contract is in place, it cannot be easily altered or stopped.17
The operating principle of a smart contract is remarkably similar to that of an industrial Programmable Logic Controller (PLC), the simple, robust computers that automate factory floors and power plants.18 A PLC operates in a continuous loop: it monitors inputs (e.g., the temperature from a sensor), executes a program based on those inputs (e.g., "if temperature is above X, then..."), and changes its outputs accordingly (e.g., "turn on cooling fan").18 A smart contract applies this same input-program-output logic to human agreements. It monitors for a digital input (a "triggering event"), such as the receipt of a payment or a data feed from an external source, and automatically executes the programmed output, such as transferring ownership of a digital asset or releasing funds from an escrow account.16 Its primary applications are in automating objective, binary tasks: ensuring payment of funds when certain conditions are met, and imposing financial penalties if they are not.17
While revolutionary in concept, the first generation of smart contracts has revealed significant cracks in the code. Their great strength—relentless, automated execution—is also their greatest weakness. The logic of a computer is binary and unforgiving, whereas human and commercial relationships are built on nuance and flexibility. A smart contract cannot, for instance, distinguish between a valued, long-term customer who is a day late on a payment and a fraudulent actor. It will automatically impose the programmed penalty without regard for context or the preservation of a business relationship, something a human might choose to overlook.17
Furthermore, smart contracts face a fundamental logistical challenge known as the "oracle problem." By design, a blockchain is a closed, self-referential system; it cannot directly access external, real-world information like weather data, stock prices, or the status of a physical shipment.17 To get this information, the smart contract must rely on a trusted third-party service called an "oracle" to retrieve the data and feed it into the blockchain. This reintroduces a point of trust and potential failure into a system designed to be trustless. If the oracle is compromised, provides incorrect data, or simply goes offline, the smart contract can fail or execute improperly.17