My grandfather kept his shop at the end of Alderton Street, in a narrow building wedged between a bakery and a laundromat, as if the city had simply forgotten to leave him more room. The sign above the door read FELIX MORROW — TIMEPIECES in letters that had once been gold but had faded to the color of old mustard. I spent three summers there as a teenager, and I can still hear the sound of that place: a dense, layered ticking, like a forest where every leaf was counting something.
Grandfather Felix did not think of himself as a repairman. That word, he told me once, implied that things broke through some failure of character. Clocks did not fail, he insisted — they merely paused to reconsider. He said this without irony, without any visible awareness that it was an unusual thing to believe. He would hold a stopped pocket watch to his ear the way a doctor holds a stethoscope to a chest, as though listening for a confession5.
The shop was organized according to a logic I never fully decoded2. Bracket clocks occupied the highest shelves, their glass panels clouded with age. Mantel clocks lined the counters in rows that seemed almost conversational, their faces all turned slightly toward one another. Pocket watches lay in open velvet trays, and grandfather clocks stood along the back wall like a row of patient witnesses. Customers who came in for the first time almost always stopped just inside the doorway and looked around slowly, the way you do when you step into a cathedral and realize the ceiling is higher than you expected.
Grandfather Felix charged very little and worked very slowly. He said that any craftsman who hurried was simply borrowing trouble from a future he hadn't earned yet. He used a loupe that had belonged to his own grandfather3, a brass-rimmed lens worn smooth on one side from decades of being pressed against the same eye socket. When he worked, he became extraordinarily still, and I learned not to speak during those stretches. I would sit on a stool near the window and read, or watch the people passing outside who never seemed to notice the shop at all.
One August afternoon, a woman came in carrying a carriage clock wrapped in a dish towel. She set it on the counter and said her husband had dropped it the morning he died, six years ago, and she had never had the heart to find out if it still worked. Grandfather Felix unwrapped it carefully, turned it over in his hands, and set it down without opening it. He told her to come back on Friday.
On Friday she returned. Grandfather Felix placed the clock on the counter between them. It was running. The woman put both hands over her mouth and stood like that for a long moment, and grandfather Felix looked away toward the window with a studied interest in nothing in particular4. She paid him what he asked, which was almost nothing, and when she left she was walking differently than when she had come in — not faster, not slower, but as if the ground had become slightly more reliable.
I asked him that evening why he had charged her so little. He was closing up the shop, covering the counter clocks with a length of soft flannel the way you might tuck something in for the night.
"She already paid for it," he said. "She paid for it for six years7."
I thought about that for a long time. I think I am still thinking about it.
Grandfather Felix died the winter I turned twenty-two, and the shop was sold and eventually became a cell phone repair store, which is perhaps not so different after all. I kept one thing from the shop: a small brass travel clock that had never worked in all the years I knew it, which he kept not on a shelf with the others but on his desk beside his coffee cup10. When I asked him once why he kept a clock that didn't run, he looked at me as though I had asked something almost too obvious to answer.
"To remind me," he said, "that not everything that has stopped is finished."
For most of the twentieth century, economists operated under a comfortable assumption: people are rational. Given a set of choices, individuals would weigh their options carefully, calculate the costs and benefits, and arrive at the decision that best served their interests. This model, known as the rational actor framework, was elegant and mathematically convenient, which may explain why it persisted for so long12 despite mounting evidence that human beings routinely behave in ways that defy it.
The challenge to this orthodoxy gained serious momentum in the 1970s, when psychologists Daniel Kahneman and Amos Tversky began publishing research demonstrating that people rely on mental shortcuts, which they called heuristics, when making decisions under uncertainty. These shortcuts are efficient in many situations but produce predictable errors in others. In one landmark study, participants were asked to choose between two medical programs to combat a disease expected to kill 600 people. When the options were framed in terms of lives saved, most participants chose the safer program. When the identical outcomes were reframed in terms of lives lost13, most participants chose the riskier option. The objective facts had not changed at all, yet the decisions flipped. This phenomenon, which Kahneman and Tversky called the framing effect, illustrated something profound: the way a choice is presented shapes what people decide, often more powerfully than the content of the choice itself.
This line of research eventually grew into what is now called behavioral economics, a field that integrates psychological insights into economic models. Unlike traditional economics, which describes how people should behave according to theoretical principles, behavioral economics attempts to describe how people actually behave. The distinction sounds simple, but its practical implications have proven enormous.
Consider retirement savings. Classical economic theory predicts that workers will rationally calculate their future needs and save accordingly. In reality, many workers—particularly those in their twenties and thirties—consistently undersave, even when their employers offer matching contributions that amount to free money. Behavioral economists trace this failure partly to present bias, the tendency to value immediate rewards far more heavily than future ones. A dollar today feels more valuable than a dollar next year, even when logic says otherwise.
Rather than lecturing workers about the importance of saving, behavioral economists proposed a different approach: change the default14. Instead of requiring employees to actively enroll in a retirement plan, employers would automatically enroll them and require them to opt out if they wished. Studies of companies that implemented this change found dramatic increases in participation rates, sometimes rising from below 50 percent to above 90 percent. The workers' options had not changed. The information available to them had not changed. Only the default setting had changed—and that was enough.
This concept of structuring choices to steer people toward beneficial outcomes without removing their freedom to choose differently is called libertarian paternalism, a term coined by legal scholar Cass Sunstein and economist Richard Thaler. Critics of the approach argue that it is manipulative, that governments and corporations should not be in the business of nudging citizens toward particular decisions even under the banner of their own good. Proponents counter that choice architecture is unavoidable—any presentation of options reflects some structural assumption15—and that it is more honest to acknowledge this influence and deploy it responsibly than to pretend it does not exist.
Thaler would later receive the Nobel Prize in Economics in 2017, recognition that behavioral economics had moved from the margins of the discipline to its center. Today its applications extend well beyond retirement savings. Public health agencies use default opt-ins to increase organ donation rates16. Energy companies send customers reports comparing their consumption to that of their neighbors, exploiting social norms to reduce energy use. Tax authorities redesign collection letters to emphasize that most citizens pay on time, leveraging conformity pressure rather than fear of penalties.
What unites these interventions is a shared recognition that context is not neutral. The environment in which a decision is made—the order of options, the framing of information, the presence or absence of a default—is as much a part of the choice as the options themselves. Behavioral economics did not merely refine a model; it replaced one picture of human nature with another, one that is less flattering in its acknowledgment of human irrationality but ultimately more useful for the task of designing systems that serve people as they are rather than as theory imagines them to be.
There is a particular kind of arrogance embedded in the act of making a map. To draw a coastline is to insist that you have understood it, that your eye and your instrument have been adequate to its complexity. Yet coastlines shift, inlets fill with sediment, cliffs surrender themselves to the sea. A map, once printed, becomes a historical document almost immediately. This was the central irony that consumed Helena Marsh for most of her career at the Geographical Survey Institute, and it was the irony she finally addressed—publicly, controversially—in the lecture she delivered at the Institute's centennial symposium.
Marsh had spent thirty-one years producing nautical charts for the North Atlantic corridor. She was regarded, even by her detractors, as someone of uncommon precision. She had developed new protocols for depth-sounding verification and had personally revised the standard for shoreline notation that the Institute used across all its publications. Her colleagues described her variously as meticulous, exacting, and—less charitably—impossible to satisfy22. What they rarely said, at least not to her face, was that she was also genuinely haunted by her work.
The haunting had begun early23. When Marsh was twenty-six and still a junior surveyor, she had worked on a revision of charts for a stretch of the Faroese coast. The work was painstaking: weeks aboard a vessel that rolled even in calm weather, constant readings, endless verification. The resulting charts were certified accurate and distributed to commercial shipping interests. Three years later, a fishing trawler ran aground on a rock formation that the charts showed as submerged to a safe depth. No lives were lost, but the incident shook Marsh in a way that never fully resolved. She understood, intellectually, that the rock's apparent rise could be explained by any number of natural processes—sediment shifts, subtle tectonic movement, even measurement error in the original survey. But the intellectual explanation did not quiet what she felt.
For the next two decades, Marsh folded this discomfort into her working habits. She became more rigorous, not less—updating verification cycles, arguing for greater funding for re-survey expeditions, writing internal memos that her supervisors found alarmist. She believed, and said often enough that it became something of an office refrain, that a chart which had not been re-verified within five years should carry an explicit notation of its age24. The Institute's position was that such notations would undermine public confidence in official cartography24. Marsh argued the opposite: that unwarranted confidence was the greater danger.
What Marsh could not bring herself to say plainly—not until the symposium—was that her argument was as much philosophical as it was practical. She had come to believe that the conventions of cartographic authority, the clean bold lines and precise numerical depths, performed a kind of deception. They communicated certainty that the discipline could not honestly claim. The map said: here is the world. What the map could not say, and what Marsh believed it should find some way to say, was: here is the world as we understood it at a particular moment, from a particular vantage, with instruments of particular limitation.
At the symposium, she called this the cartographer's confession—the acknowledgment that every map is also a statement of what the mapmaker did not know, could not see, and could not predict. The audience, composed largely of professional geographers, surveyors, and cartographic historians, received the lecture with what one attendee later described as uncomfortable recognition26. Several colleagues approached her afterward to say privately that they had always felt something similar but had not known how to articulate it. A few were openly dismissive, arguing that Marsh was romanticizing fallibility and that the practical utility of charts depended on readers treating them as authoritative.
Marsh retired from the Institute the following year. She spent the years afterward writing a short book about the history of cartographic error—not a catalogue of disasters, but an examination of what the errors revealed about the assumptions each era brought to the problem of representing place28. She was not interested in blaming the mapmakers. She was interested in what it meant to commit an image of the world to paper, knowing the world would not hold still.
The book was modest in its ambitions and modest in its reception. But it found its readers—students of geography, historians of science, and a surprising number of people who had simply loved maps since childhood and had always sensed, without quite knowing why, that there was something a little sad about them.
On warm summer evenings across much of North America and Asia, the edges of forests and open meadows flicker with cold, greenish-yellow light. Fireflies—beetles of the family Lampyridae—have fascinated observers for millennia, yet the biochemical machinery underlying their glow was not fully decoded until the latter half of the twentieth century. What researchers discovered is a system of extraordinary efficiency, one that has since inspired innovations in medicine, environmental monitoring, and even food safety testing.
The light-producing organ, called the lantern, occupies the underside of a firefly's abdomen. Within specialized cells known as photocytes, a reaction unfolds between a small organic molecule called luciferin and an enzyme called luciferase. When luciferin is oxidized in the presence of luciferase, adenosine triphosphate (ATP), and molecular oxygen, the reaction produces a short-lived intermediate that releases energy almost entirely as visible light rather than heat. Scientists describe this process as bioluminescence, and the firefly version of it is remarkable because it converts roughly ninety to ninety-five percent of its chemical energy directly into light32. By contrast, a standard incandescent light bulb converts only about five percent of its electrical energy into visible light, releasing the rest as heat. Even modern LED bulbs, celebrated for their efficiency, operate at around forty to fifty percent conversion rates. The firefly, shaped by hundreds of millions of years of evolution, remains the most efficient light-emitting system known.
Not all firefly light is identical. Different species produce flashes that vary in duration, rate, and color, ranging from pale yellow-green to orange. These differences arise partly from subtle variations in the luciferase enzyme and partly from the pH level inside the photocyte cells. Laboratory studies have shown that when researchers adjust the acidity of a reaction mixture containing firefly luciferase and luciferin, the color of the emitted light shifts perceptibly. At lower pH—more acidic conditions—the light tends toward orange or red; at higher pH it tilts toward green33. Evolution has apparently tuned each species' internal chemistry to produce a species-specific signal, reducing the likelihood that a female will respond to the flash of the wrong species.
The ecological function of bioluminescence in fireflies is primarily communication for mating. Males typically fly through the air producing a species-specific flash pattern, while females perch on vegetation and respond with their own precisely timed flash. A male reads the delay between his flash and the female's reply—often a fixed interval of two or three seconds for a given species—and steers toward the source. The system is not without hazards, however. Females of the genus Photuris have evolved the ability to mimic the reply flashes of smaller Photinus females. When a Photinus male approaches, the Photuris female seizes and eats him, acquiring defensive chemicals called lucibufagins that make her unpalatable to predators such as spiders and birds34. This phenomenon, sometimes called aggressive mimicry, illustrates how a signaling system designed for one purpose can be co-opted for an entirely different one.
Beyond their ecological roles, firefly proteins have become indispensable research tools. Because luciferase produces light only when ATP is present35, scientists can use the enzyme as an exceptionally sensitive detector of cellular energy. In cancer research, luciferase genes are inserted into tumor cells; researchers then track tumor growth in living mice by detecting the light those cells emit, eliminating the need for repeated surgical biopsies36. In food safety laboratories, luciferase-based tests reveal bacterial contamination within minutes by measuring the ATP released from microbial cells—a process that once took two or more days using traditional culture methods.
Despite their utility and their centuries of cultural prominence—appearing in Japanese poetry, Chinese folklore, and countless summer memories—firefly populations have declined sharply in recent decades. Light pollution disrupts the flash communication that fireflies depend on for reproduction; artificial lights make it difficult for males and females to detect each other's signals. Habitat loss removes the moist, leaf-littered environments where larvae spend