The previous chapter asked whether marginal reasoning can navigate moral terrain. This chapter asks something stranger: whether marginal reasoning can be mechanized — whether the margin itself can be built into a machine.
In 1843, a young woman of twenty-seven sat at her desk composing what would become the most prophetic footnote in the history of science. Augusta Ada King, Countess of Lovelace — daughter of Lord Byron, student of mathematics, and collaborator of Charles Babbage — was translating an Italian paper about Babbage's proposed Analytical Engine. The translation itself was routine work. What made it extraordinary were the notes she appended, which ran to three times the length of the original article. In her famous Note G, Lovelace wrote that the Analytical Engine "might compose elaborate and scientific pieces of music of any degree of complexity or extent." She saw, with eerie clarity, that a machine built from brass gears and punched cards need not confine itself to arithmetic. It could manipulate any system of symbols — musical notes, algebraic expressions, logical propositions — provided someone first understood the rules governing them.[C1]
Lovelace also saw the limits, or what she believed were limits. "The Analytical Engine has no pretensions whatever to originate anything," she cautioned. "It can do whatever we know how to order it to perform." This was both a sober technical assessment and, perhaps, a prayer — a hope that there remained some irreducible spark of human creativity no mechanism could touch.[C2]
Twenty years later, in a workshop in London, a man who had read Charles Babbage with great interest was building a machine of his own. It was not made of brass gears but of wooden keys and wires, and it did not calculate numbers. It reasoned. William Stanley Jevons, then in his late twenties, had constructed what he called a "Logical Abacus" — and later, a more refined "Logical Piano" — capable of working through the combinations of Boolean logic mechanically, arriving at valid conclusions faster than a trained logician could manage with pen and paper.[C3] When Jevons demonstrated the device before the Royal Society in 1870, he made a claim that went further than anything Lovelace had dared to assert. "Material machinery," he wrote, "is capable, in theory at least, of rivalling the labours of the most practiced mathematicians or even to surpass them."
Rivalling. Not merely assisting. Not just computing what we already know how to compute. Rivalling.
Two Victorians, separated by two decades, had stared into the same abyss. Ada Lovelace saw the machine and insisted it could not originate. Jevons saw the machine and suspected it could rival. Between those two words — originate and rival — lies a chasm that the twenty-first century is still trying to measure.
The Unitarian Inheritance and Early Formation
How a dissenting intellectual culture and family of polymathic achievers shaped Jevons's approach to knowledge.
But we are getting ahead of ourselves. To understand why Jevons could see what Lovelace denied — that a machine might rival the mind that built it — you need to understand what shaped him. To do that, we need to go back to Liverpool in the 1830s.
William Stanley Jevons was born on September 1, 1835, into a Unitarian family of iron merchants and inventors. The Unitarians occupied a peculiar position in Victorian England — dissenting Protestants who rejected the Trinity, prized reason, and produced a staggering number of scientists, reformers, and industrialists. Josiah Wedgwood was a Unitarian. So was Joseph Priestley. The culture valued inquiry over orthodoxy, evidence over tradition.[C4] It was, in other words, exactly the right soil for growing a polymath.
His father, Thomas Jevons, was an iron merchant with genuine intellectual interests and a talent for invention — he wrote on legal and economic topics and even dabbled in engineering. His mother, Mary Anne Roscoe, came from a distinguished Liverpool family of considerable cultural achievement; her father, William Roscoe, was a historian, art collector, and abolitionist whose personal library was legendary in the city.[C5] From both sides, Stanley — as the family called him — inherited the conviction that the world was something to be studied, catalogued, and improved.
His mother died when he was ten. This is the kind of biographical fact that biographers tend to handle with either too much weight or too little. We do not know exactly how it shaped him. We know that he became, in the years that followed, an unusually serious and inward child. We know that he threw himself into collecting — specimens, data, observations of every kind — with an intensity that strikes the modern reader as familiar. John Maynard Keynes, who admired Jevons enormously, described him as "strongly introverted," a man for whom social interaction was effortful in a way that intellectual work never was.[C6] Whether or not one reaches for a contemporary diagnostic label — and several historians have suggested that Jevons may have been on the autistic spectrum — the pattern is clear enough. Here was a boy, and then a man, who found the world of things and numbers and systems more navigable than the world of people.
He was a born collector — not merely of objects, but of facts and patterns everywhere he looked.[C7]
The range of his interests defies easy summary. Over the course of a life that lasted only forty-six years, Jevons made serious contributions to economics, formal logic, statistics, scientific methodology, and public policy. He wrote on chemistry, metallurgy, and music theory. He studied what was then called pedesis — the erratic motion of small particles suspended in fluid, what we now call Brownian motion — and while he got the mechanism wrong (he attributed it to causes that turned out not to be primary), he was right to see it as a phenomenon worth rigorous investigation, decades before Einstein's famous 1905 paper provided the correct explanation.[C8] He built machines. He wrote textbooks that remained in use for generations. He was, in the fullest sense, a Victorian polymath — though the Victorians would not have used that word. They would have simply called him a man of science.
Sydney and the Empiricist's Discipline
How five years as an assayer in Australia trained Jevons to value precise measurement above all theory.
The story of how Jevons found his vocation begins, improbably, at the bottom of the world. In 1854, at the age of eighteen, he left England for Sydney, Australia to take up a position as assayer at the new Royal Mint there. It was a practical appointment for a young man who had studied chemistry at University College London but whose family's financial difficulties — his father's business had failed — made completing a full degree a luxury he could not afford.[C9]
The Sydney years were formative in ways that no one, least of all Jevons, could have predicted. As an assayer, his job was measurement — precise, exacting measurement of the gold content in the metal that poured in from the Australian gold rush. Day after day, he weighed, he tested, he quantified. The work was not intellectually glamorous, but it trained him in something that would prove essential: a deep, almost physical respect for data. Jevons did not theorize in the abstract and then look for confirming evidence. He measured first and theorized second. This instinct would set him apart from nearly every other economist of his generation.
He also found, in Australia, the space to think. Far from the intellectual bustle of London, surrounded by a colonial society that had little use for his wider interests, Jevons read voraciously and began to sketch out ideas that had nothing to do with metallurgy. He studied the social conditions of Sydney and accumulated observations.[C10] He observed. He accumulated. And somewhere in the quiet of those antipodean evenings, the shape of a revolution began to form in his mind.
The pivotal moment came in 1860, after Jevons had returned to England and was finishing his studies. In February of that year, he wrote a letter to his brother Herbert that vibrates with barely contained excitement. He had discovered, he told Herbert, "the true Theory of Economy." The theory was mathematical. It was based on the idea that value derived not from the total quantity of a good but from the usefulness of the last increment — what would come to be called marginal utility. And Jevons knew, with the peculiar certainty of a man who has seen something clearly for the first time, that he was right.
He also knew — and this is the part that breaks your heart a little — that no one was going to care. "I am very hopeful about my Economy theory," he wrote, with that mix of confidence and resignation that marks so many letters from ahead-of-their-time thinkers, but he harbored no illusions about his immediate prospects of being heard. He was young, obscure, lacked a prestigious post, and was proposing to overturn the foundations of political economy using mathematics — a tool that most political economists of 1860 regarded with suspicion bordering on hostility.
He was right on both counts. The theory was correct. And for years, almost no one cared.[C11]
The Coal Question and the Paradox of Efficiency
A pessimistic tract about resource depletion that revealed an unintuitive truth about consumption and innovation.
Before the world took notice of his economics, Jevons published The Coal Question (1865), a pessimistic tract about Britain's coal reserves that attracted the attention of Gladstone himself.[C12] His prediction of imminent depletion proved wrong, but buried in Chapter VII was an insight that would outlast all his other work: that making a resource cheaper to use leads to more consumption of it, not less. Better steam engines burned more coal, not less, because the efficiency gains made coal-powered industry profitable in applications where it had previously been too expensive. This "Jevons Paradox" remains one of the most important and least intuitive ideas in economics — and it is uniquely the product of a polymathic mind.[C13] Only someone trained in both chemistry (he understood industrial processes) and economics (he understood incentive structures) could have seen it. The paradox, the coal book's success, and his growing reputation in logic helped secure Jevons a professorship at Owens College, Manchester, when he was just thirty years old.[C14]
Machine Reasoning and the Question of Origination
The Logical Piano revived Lovelace's ancient question: can machines truly create, or only execute?
And yet it is worth pausing here to note something strange about Jevons's reputation, both during his life and after. He was doing at least three things at once — economics, logic, and scientific methodology — and the world could never quite decide which one mattered most. His contemporaries valued his logic above all. When he presented the Logical Piano to the Royal Society, it caused a genuine sensation. Here was a machine that could think — or at least perform operations that, when done by humans, we would unhesitatingly call thinking. The implications were obvious and unsettling.[C15]
This is where Jevons and Lovelace converge most powerfully. Lovelace, writing about Babbage's far more ambitious but never-completed Analytical Engine, had insisted on a clear boundary: the machine could process, but it could not create. It could execute instructions of any complexity, but the originating intelligence — the intention — had to come from a human mind. Jevons, holding a much simpler device in his hands, a device that merely worked through the combinations of Boolean syllogisms, arrived at a more radical conclusion. "Mind thus seems able to create its own rival," he mused.[C16a]
The sentence deserves careful attention. Jevons did not say that his Logical Piano was a rival to the human mind. He was not that foolish. He said that the mind seems able to create such a rival — that there was no obvious theoretical barrier, that the road from his wooden keys and wires to something genuinely mind-like was, in principle, open. It was a philosophical leap that his Logical Piano, considered purely as a piece of engineering, did not warrant. But Jevons was not reasoning from the engineering. He was reasoning from the logic. If thought could be formalized, and if formal operations could be mechanized, then where, exactly, was the stopping point?
Lovelace had answered: the stopping point is origination. The machine cannot want anything. It cannot surprise us except through our own errors.
Jevons's implicit answer was: maybe there is no stopping point.
Consider what this means in practice. Show a machine the rules of chess, and it will play chess — because you have told it the goal and the rules. But it did not choose to play chess. It did not decide that chess was worth playing. The human mind did that. Lovelace would say: there is the line. The machine executes; the human originates the purpose. But Jevons might ask: if you can formalize what it means to "want" something — if you can mechanize enough layers of intention — at what point does the distinction between executing and originating collapse? When a machine writes a poem that moves you, and you could not have predicted that specific poem before the machine generated it, in what sense is the machine merely executing rather than creating something genuinely novel?
It is a disagreement that remains unresolved. The entire field of artificial intelligence, from Alan Turing's 1950 paper onward, can be read as an extended argument between the Lovelace position and the Jevons position.[C16b] Can machines originate, or can they only process? When a large language model writes a sonnet that moves you, has it originated something, or has it merely executed a very complicated instruction? Lovelace and Jevons, working with technologies that seem laughably primitive by modern standards, had already framed the essential question.[C16]
The Power of Boundary-Crossing Insight
How Jevons's mastery of multiple fields allowed him to see what specialists missed.
There is a temptation, when writing about polymaths, to make their breadth sound easy — to describe their movements between fields as the graceful leaps of a mind unbound by disciplinary walls. The reality, in Jevons's case, was messier. He worked obsessively. He suffered from poor health and chronic anxiety. His journals reveal a man who drove himself relentlessly and worried constantly that he was not accomplishing enough. The collecting — the data, the books, the observations — was not a hobby. It was a compulsion. Keynes's "strongly introverted" is a polite way of describing someone who found sustained human contact draining in a way that sustained intellectual labor was not.
But the breadth was real, and it mattered. Jevons could build the Logical Piano because he understood both formal logic and mechanical engineering. He could write The Coal Question because he combined economic theory with an empiricist's obsession with real-world data on coal production, consumption, and pricing. He could see marginal utility — the insight that would reshape economics — because he came to the discipline not as an apprentice of Ricardo or Mill but as an outsider trained in chemistry and physics, fields where measurement and incremental change were second nature. His polymathy was not decorative. It was structural. Each field he worked in gave him tools and intuitions that he carried into the next.
This is a pattern worth naming, because it recurs throughout the history of ideas. The most transformative insights often come not from the deepest specialist but from the person who carries knowledge across a boundary. Jevons carried measurement from chemistry into economics. He carried formalization from logic into economic theory. He carried the mechanist's confidence — the conviction that if you can describe a process precisely, you can build a machine to execute it — from engineering into philosophy of mind.
Legacy and the Unfinished Question
Jevons's death and the curious reversal of how history would judge his contributions.
On August 13, 1882, Jevons went swimming near Hastings, on the southern coast of England. He was forty-six years old. He drowned. The exact circumstances remain unclear — he was known to be a strong swimmer, and various explanations have been proposed, from a heart condition to simple misadventure in the currents.[C17]
The obituaries that followed praised his work in logic first and economics second. This was, from the perspective of posterity, exactly backwards. Jevons's logic was important but incremental — he refined and popularized Boolean methods, and the Logical Piano was a brilliant curiosity, but formal logic would have advanced along roughly similar lines without him. His economics, by contrast, was foundational. The marginal revolution that he helped launch — independently of Léon Walras in France and Carl Menger in Austria, all three arriving at similar ideas at roughly the same time — would entirely remake the discipline.[C18]
But perhaps the obituarists were responding to something real, even if they got the ranking wrong. The Logical Piano was visible in a way that marginal utility theory was not. You could see it, touch it, watch it work. It was a physical object that embodied an intellectual claim: that the operations of reason could be mechanized. And that claim — Jevons's claim, building on Babbage, going beyond Lovelace — turned out to be more consequential than anyone in 1882 could have imagined.
Today, the machines have gotten rather better. They have moved from Boolean syllogisms to natural language, from wooden keys to silicon, from the parlor demonstrations of the Royal Society to the server farms that power modern artificial intelligence. The question Jevons raised — whether "material machinery" could rival human thought — is no longer theoretical. It is industrial. It is economic. It is, in a way Jevons would have appreciated, amenable to measurement.
He would have wanted to measure it. He would have collected the data, built a table, looked for the pattern. That is what Jevons did. That is who Jevons was: a man who believed that if something was real, it could be quantified, and if it could be quantified, it could be understood. Whether he was right about that — whether understanding and quantification are truly the same thing — is yet another question he bequeathed to us, along with the paradox, the piano, and the unshakable suspicion that the mind can build its own rival.