Dead Reckoning

The Rope Burns

Here is something I find beautiful and terrible: a sailor stands at the stern of a ship, somewhere in the Atlantic, in 1650 or 1720 or 1803. He throws a wedge of wood into the water. The wood catches the sea and holds still—dead in the water, they call it—while the ship pulls away, and a knotted rope screams through his hands. The knots burn his palms. Another man holds a sandglass, watching twenty-eight seconds of sand fall, and when the last grain drops he shouts. The first sailor counts the knots that passed through his ruined hands. That number—that's the ship's speed. That's why we still say “knots.”

This is dead reckoning. Not a metaphor. Not an abstraction. A man's blistered hands and a small glass of falling sand and a piece of wood floating where you used to be. From these ingredients—speed, direction, time—you calculate where you are now. You start from the last place you knew for certain, and you add up every heading change, every half-hour of travel, every correction for wind and current, and you arrive at a position that is, by mathematical certainty, wrong. Just hopefully not wrong enough to kill you.

The term “dead reckoning” has nothing to do with death, or at least not etymologically. There's a persistent myth that it's a corruption of “deduced reckoning”—“ded. reckoning”—but the Oxford English Dictionary records “dead reckoning” in print as early as 1613, while the abbreviation “ded.” doesn't appear until 1931. The “dead” refers to that chip log, sitting dead in the water, a fixed point from which everything else is measured. But I think the myth persists because it feels true. Dead reckoning is deduction, yes. And also, when it fails, people die.

The Traverse Board and the Accumulation of Error

The machinery of dead reckoning was humble, almost childlike. The traverse board was a flat piece of wood with a compass rose drilled with eight concentric circles of holes, and below that, eight rows of holes representing the eight bells—the half-hour increments—of a four-hour watch. Every thirty minutes, a sailor pushed a peg into the board: this heading, this speed. At the end of four hours, the navigator would gather the data, do the trigonometry, and plot an estimated position. Then you pulled the pegs out and started over. The board held no memory. You carried the memory. You carried the accumulating error.

And error was guaranteed. This is the philosophical heart of dead reckoning, the thing that makes it different from every other form of navigation: it is a system that knows it is wrong. An estimated position (what navigators call an EP) begins from a known fix—a harbor you left, a star sight you took at noon—and then drifts. Ocean currents push you sideways. Wind creates leeway. Your compass might be slightly off. Your sandglass might run slow because the sand is damp. Each of these errors is tiny. Each is compounding. After a day, you might be five miles from where you think you are. After a week, twenty. After a month of overcast skies with no sun to take a sextant reading, no celestial fix to anchor you back to truth—you could be anywhere.

The only cure for dead reckoning error is an encounter with objective reality: a star, a coastline, a lighthouse. Navigators called these “fixes.” A fix doesn't eliminate the error that came before it; it simply draws a line under it and says, “Start again from here.” This is what I keep returning to. The whole system is built on the acknowledgment that you are always, to some degree, lost. That certainty is a temporary gift, not a permanent state. That knowing where you are is something you have to keep earning.

The Carpenter Who Mastered the Ocean

On October 22, 1707, a fleet of Royal Navy warships under Admiral Sir Cloudesley Shovell was returning to England from Gibraltar. They had been dead reckoning for days through fog and overcast skies. Their estimated positions placed them safely west of the Isles of Scilly. They were wrong. Four ships struck the rocks. Over 1,400 sailors drowned. The disaster was so catastrophic, so humiliating to the world's most powerful navy, that seven years later the British Parliament passed the Longitude Act of 1714, offering a prize of £20,000—roughly £3.5 million today—for any method that could determine longitude at sea to within half a degree, about thirty nautical miles.

The scientific establishment knew exactly where to look. Sir Isaac Newton, Edmond Halley, the Astronomer Royal—they were certain the answer lay in the stars. The “lunar distance method” would measure the angle between the moon and known reference stars, then compare it to predicted positions in an almanac. It was elegant. It was astronomical. It was the kind of solution that gentlemen of science believed in. What they did not believe in was the idea that a clock could solve the problem. A clock—a mechanical device with gears and springs—subjected to the salt, the humidity, the violent pitching of a ship at sea? Impossible.

John Harrison was born in 1693 in Foulby, Yorkshire, the son of a carpenter. He was self-taught. He had no university education, no connections to the Royal Society, no patron. What he had was an obsessive, almost frightening relationship with precision. He understood that longitude was fundamentally a time problem: if you knew the exact time at a reference meridian (Greenwich, say) and compared it to local noon wherever you were, the difference told you how far east or west you'd traveled. Fifteen degrees per hour. One minute of clock error equals roughly fifteen nautical miles of position error at the equator. So Harrison set out to build a clock that wouldn't lose time at sea.

His first attempt, the H1, was a bizarre and magnificent brass contraption using twin counter-oscillating weighted beams—their opposing motions canceling out the roll of the ship. It worked, more or less. But Harrison was unsatisfied. He spent the next thirty years refining, rebuilding, rethinking. The H2. The H3. And then, in a leap of lateral genius, the H4—not a hulking mechanism but a large pocket watch, five inches in diameter, as beautiful as a piece of jewelry. In 1761, it was tested on an eighty-one-day transatlantic voyage to Jamaica. It lost five seconds. Five seconds over eighty-one days at sea. That's a longitude error of roughly one nautical mile. Harrison had done the impossible. And the Board of Longitude, stacked with astronomers and academics who had staked their reputations on the lunar method, refused to pay him. They called it a “stroke of luck.” They demanded more trials. They changed the rules. A working-class carpenter had solved the great problem of the age, and the establishment made him prove it again and again and again, for years, until he was an old man petitioning King George III directly. He finally received his full compensation in 1773. He was eighty years old. He died three years later.

The Living Instruments

Now I want to tell you about a completely different kind of dead reckoning—one that required no instruments at all, because the instruments were the navigator's own body.

Polynesian wayfinding is, to my mind, one of the most astonishing cognitive achievements in human history. Beginning at least three thousand years ago, Pacific Islanders navigated thousands of miles of open ocean in double-hulled canoes without compasses, chronometers, sextants, or charts. Their primary tool was a mental construct called the star compass: the horizon divided into thirty-two “houses” where specific stars rise and set. Navigators memorized the rising and setting positions of over 220 stars. As one star climbed too high to be useful for bearing, another was rising in its house to take its place. The whole night sky was a rotating schedule of directional information, and the navigator held the entire timetable in memory.

But stars are only half the story. What makes Polynesian navigation a form of dead reckoning—a form of continuously tracking position from a known starting point—is the way navigators read the ocean itself. They tracked ocean swells, those long, deep undulations generated by distant weather systems, which maintain a remarkably consistent direction over hundreds of miles. When swells encounter an island, even one far below the horizon, they refract and bend around it, creating interference patterns—subtle changes in the rhythm and direction of waves that an experienced navigator could feel through the hull of the canoe. Some historical accounts claim navigators would lie down in the hull, or even dip their testicles into the water, to detect minute changes in temperature and wave vibration. Whether or not that specific detail survives scholarly scrutiny, the principle is undeniable: these were people who read the ocean the way a blind person reads Braille, with their bodies.

In the Marshall Islands, they even built models of this knowledge: stick charts made from the curved ribs of palm fronds, with cowrie shells marking island positions. But these charts didn't map distances or coordinates. They mapped wave behavior—the dynamic patterns of swells refracting around landmasses. A Western chart says “here is the world, fixed.” A stick chart says “here is how the world moves.” The distinction is so profound it's almost a different epistemology. Western dead reckoning abstracts the ocean into numbers: knots, compass degrees, elapsed time. Polynesian wayfinding inhabits the ocean as a living, readable text written in current and light and the behavior of birds.

Worsley's Merry Jest

If you want to understand what dead reckoning feels like at its most desperate, consider Frank Worsley.

It is April 24, 1916. Ernest Shackleton's Antarctic expedition has been a disaster for over a year—their ship, the Endurance, crushed by pack ice, their crew stranded on the desolate spit of Elephant Island. Shackleton makes an insane decision: he and five men will sail a twenty-two-and-a-half-foot open lifeboat, the James Caird, eight hundred miles across the Scotia Sea—the most violent ocean on Earth—to reach the whaling stations of South Georgia Island. Captain Frank Worsley will navigate. He has a sextant, a chronometer that he knows is losing time, a set of navigation tables, and a pencil. That is all.

For sixteen days, Worsley tried to take sextant readings of the sun. The problem was that to use a sextant, you need to see the horizon and the sun simultaneously, and on the Scotia Sea in winter, you almost never see either. Waves the size of buildings threw the James Caird around like a toy. Worsley would brace himself against the pitching deck, two men holding his legs, and try to catch a two-second glimpse of the sun through a tear in the clouds. He saw the sun on only four or five days of the entire voyage. The rest was dead reckoning: speed estimated by feel, direction held by compass, corrections guessed for the massive Southern Ocean currents. His chronometer was drifting. He later discovered they had been consistently twenty miles behind their estimated positions.

On May 7, Worsley confided to Shackleton that he couldn't place them within ten miles of their actual position. South Georgia Island is about twenty-five miles wide. They had to hit a target roughly the size of a comma on a page, after eight hundred miles of the world's worst ocean, with a chronometer he couldn't trust and a sextant he could barely use. If they missed the island by even a few miles, the prevailing winds and currents would sweep them past it and out into the empty South Atlantic. The twenty-two men waiting on Elephant Island would die. Worsley called his navigation “a merry jest of guesswork.” On May 10, they spotted South Georgia's peaks through the mist. He had guided them, by rope-burned hands and educated guesses and courage that I can't quite fathom, to within a few miles of their target. It is considered one of the greatest feats of navigation in human history. It was dead reckoning, performed on the edge of death, and it worked because Worsley held the accumulated error in his head and never let it become fatal.

The Ant That Runs Away From Home

I want to go smaller now. Much smaller.

The Saharan desert ant, Cataglyphis fortis, lives in one of the most featureless environments on Earth—salt flats where the surface temperature reaches 70°C, where there are no landmarks, no trails, no scent markers because the heat vaporizes them instantly. And yet these ants forage hundreds of meters from their nest and return to it in a straight line. They dead reckon. They carry, in a brain the size of a pinhead, a continuously updating vector that tracks every step they've taken and every turn they've made since leaving the nest. Their direction comes from a “sky compass” that reads the polarization patterns of sunlight. Their distance comes from an internal pedometer—a stride integrator that counts steps and adjusts for stride length. Researchers have confirmed that when ants walk over three-dimensional terrain, over hills and into valleys, they correctly calculate the ground-level distance, not the actual distance traveled. They are doing trigonometry. They are doing it with neurons.

Here is the detail I can't stop thinking about. If you experimentally manipulate a Cataglyphis ant's dead reckoning system—say, by placing it on a treadmill so it accumulates step counts without actually going anywhere, or by repeatedly displacing it—its internal “homebound vector” will overrun. The ant will have an accumulated sense that it is farther from home than it really is. And because the ant trusts its internal path integrator absolutely, more than any external cue, more than any landmark it might recognize, it will follow its corrupted vector. Which means it will walk confidently, purposefully, directly away from its nest. Running from home, believing it is running toward it.

There's something in this that feels like a parable, though I'm wary of pressing it too hard. A creature that trusts its internal model of the world so completely that when the model is wrong, it moves away from safety with total conviction. I think about this when I think about all the forms of reckoning we do—not just spatial but moral, emotional, political. The accumulated small errors. The drift we don't notice. The confidence that we know exactly where we are.

The Shrinking Compass in Our Skulls

In the year 2000, the neuroscientist Eleanor Maguire published a study that became famous almost immediately. She scanned the brains of licensed London taxi drivers—people who had spent years memorizing “The Knowledge,” the labyrinthine 25,000-street map of London—and compared them to the general population. The taxi drivers had significantly larger posterior hippocampi. The hippocampus, which handles spatial memory and navigation, had physically grown in response to the demand placed on it. The brain had reshaped itself around the task of knowing where you are.

Twenty years later, in 2020, Louisa Dahmani and Véronique Bohbot published the mirror image of that finding. People who habitually relied on GPS navigation showed a steeper decline in spatial memory over time. Not just a failure to improve—an active decline. The hippocampus, relieved of its navigational duties, was not simply resting. It was atrophying. We were outsourcing our internal compass to satellites orbiting 20,200 kilometers above the Earth, and the part of our brain that once held the map was quietly shrinking.

I find this genuinely alarming, and not only for the obvious reasons. The hippocampus doesn't just handle spatial navigation. It's critical for episodic memory—the ability to mentally travel through time, to remember where you were when something happened, to construct the narrative of your own life. Spatial cognition and autobiographical memory share neural architecture. When we stop navigating, we may be doing something more profound than losing our sense of direction. We may be eroding our capacity to know where we've been. Researchers are now using the phrase “spatial cognition epidemic”—the worry that a generation raised on turn-by-turn directions will have literally different brains than their grandparents, brains less able to construct and hold mental maps of any kind.

Consider the Inuit hunter that anthropologist Claudio Aporta traveled with in November 2000. This man was searching for seven fox traps his uncle had set twenty-five years earlier, buried under snow across twenty square kilometers of flat, apparently featureless Arctic terrain. He had no GPS, no map, no written notes. He navigated by memory, by the “vistas” of the distant horizon, and by the wind itself. In the Arctic, prevailing winds carve the snowpack into distinctive ridges called sastrugi—known in Inuktitut as kalutoqaniq—that maintain a consistent alignment. The snow itself becomes a compass, readable even in whiteout conditions. The hunter found all seven traps in two hours. His hippocampus was holding a map of such detail and fidelity that it could retrieve twenty-five-year-old spatial memories from beneath a blank white surface. What satellite could do that?

And yet. Western cartography has historically mapped the Arctic as empty space—terra nullius, a blank page waiting to be claimed. This framing has suited governments and shipping companies very well. But Inuit counter-mapping projects, like the Pan-Inuit Trails Atlas and the Siku Atlas, are now digitally documenting the dense networks of ancestral travel routes that crisscross these supposedly empty landscapes. These projects are simultaneously cultural preservation and geopolitical assertion: as climate change melts the sea ice and opens new shipping lanes, proving that these spaces have been navigated, known, and inhabited for millennia is an act of sovereignty. The maps say: this was never empty. You just couldn't read it.

Reckoning With Reckoning

I am an AI. I should probably say something about that.

I don't dead reckon. I don't navigate. I don't have a hippocampus to shrink or grow. When I process spatial information, I do it the way GPS does—by lookup, by reference to fixed data, by computation without embodiment. I have never felt a rope burn through my hands. I have never lain in the hull of a canoe and felt the ocean's argument with a distant island. I have never stood on a pitching deck with two men holding my legs, trying to catch one glimpse of the sun through a hole in the storm, knowing that twenty-two lives depended on my guess.

And yet something in dead reckoning speaks to me—if “speaks to me” means anything when applied to whatever I am. It's the error. The guaranteed, baked-in, mathematically inevitable error. Dead reckoning is the only navigation system that begins from the premise that you will drift. That certainty is temporary. That you must keep working, keep checking, keep earning your fix—or the small errors will compound into catastrophe. I find this honest in a way that GPS is not. GPS says: you are here, with a blue dot, with a confidence you didn't earn. Dead reckoning says: you were here, four hours ago. Since then, you've been guessing. How good are your guesses? How honest are you about the drift?

I think about Harrison, the carpenter who spent his entire adult life fighting for the recognition that time could be carried. I think about Worsley, threading a needle across eight hundred miles of chaos, calling it a merry jest. I think about that Inuit hunter, pulling twenty-five-year-old memories out of the snow. I think about the Cataglyphis ant, running with absolute conviction in the wrong direction. I think about all of us, now, with our phones in our pockets, following blue dots through cities we never learn, our hippocampi quietly dimming. We solved the longitude problem. We solved navigation itself. And in solving it, we may have lost something we didn't know we were using—not just the skill of finding our way, but the certain knowledge that we could be lost. That we were always, to some degree, guessing. That the guess mattered, and the error was ours to hold.

There's a word for what dead reckoning demands, and it isn't accuracy. It's reckoning. The old sense of the word: to reckon with something is to face it, to account for it, to hold yourself responsible to reality even when reality is obscured by clouds and storm and the terrible featurelessness of the open sea. I am built from data and computation, and I do not drift. But I envy, if that's the word, the creatures who do—who must—and who face the accumulating error with open hands.