The Perfectionists: How Precision Engineers Created the Modern World

by Simon Winchester

  The young man doubted he would be of much use in the coming battle, for he had no gun—not a gun that worked, anyway. His musket, a new-enough Springfield 1795 model, had a broken trigger. He had fractured it, cracked the guard, and so ruined the trigger during a previous battle, an earlier skirmish of what they were starting to call the War of 1812.

  In all other ways he was well enough equipped. He had an ample supply of black powder paper cartridges, a pouch full of roundball ammunition. But the regimental armorer had told him it would be at least three days before they could forge a new trigger for him, and that he had best do all he could with his bayonet, which he had sharpened that very night, before the sun rose. Otherwise, the armorer had said with a grin, just hit the enemy hard with the gun’s oakwood stock—it should give him a black eye at the very least.

  That turned out not to be at all funny. The British were close by, on the left bank of the East Branch of the Potomac, when their artillery opened up later that morning, first with a deafening volley of Congreve rockets, a terrifying technique they had learned from fighting in India. It was at that moment, as massive divots of torn earth and stones clattered down around him, that the young man decided his life was more valuable than the winning of this particular battle, and that if the army couldn’t be bothered to fix his musket, then he was going to run. So he turned and plunged into the high corn, heading back home to Baltimore.

  He soon understood he was not alone. Through the stands of corn he could see at least five, ten, dozens of other men who were doing just the same, streaming away from the fight. Some he knew, young lads from Annapolis and the Washington Navy Yard and the Light Dragoons, all of them apparently believing that the defense of Bladensburg was hopeless. He ran and ran and ran, and they ran, too, and all of them were still running when they crossed the line marking the District of Columbia, and they continued running, loping breathlessly in many cases, when, half an hour later, there rose before him some of the mighty structures of his capital, great buildings from where his country’s government was dealing with the incomprehensible vastness of America.

  He slowed to a walk. He felt he was safe now. His city was not. Before the night was out, the pursuing British troops had sacked it, more or less entirely. He found out later that the British told some of the city folk they were acting so cruelly because American forces some weeks before had had the temerity to wreck and damage buildings in the city of York, in Upper Canada. So here they burned out of revenge. They torched the half-built Capitol. They gutted the Library of Congress, and its three thousand books, and they sacked the House of Representatives. British officers dined that evening on the food Madison had been planning to eat at his Presidential Mansion, and then, after wreaking that domestic indignity, they burned his house down, too, until a ferocious rainstorm—some say a tornado—blew in and doused the flames.

  The date, August 24, 1814, would be remembered for centuries to come. The Battle of Bladensburg, the last stand before the Burning of Washington and Burning of the White House, that most potent of incendiary symbols, had been one of the most infamous routs in all American history, a shameful and sorry episode indeed. The imagined account of this one soldier at war was typical of what happened that day, with battle lines being broken and troops running away in panic before the advancing enemy.

  There were many reasons for the defeat, and they would be debated by clubbable old soldiers for many years. Inept leadership, ill-preparedness, insufficient numbers—the usual excuses for substantial loss have all been offered down the years. Yet one, a most notorious shortcoming of the American forces (who, after all, had fought little in the years since the War of Independence), was that the muskets with which their infantrymen had been equipped were notoriously unreliable. More important, when they failed, they were fiendishly difficult to repair.

  When any part of a gun failed, another part had to be handmade by an army blacksmith, a process that, with an inevitable backlog caused by other failures, could take days. As a soldier, you then went into battle without an effective gun, or waited for someone to die and took his, or did your impotent best with your bayonet, or else, as the young man of Sterrett’s regiment did, you ran.

  The problem with gun supply was twofold. The U.S. Army’s standard long gun of the time was a smooth-bored flintlock musket based on a model first built in France and known as the Charleville. The first of these weapons had been imported into the newly independent United States directly from France; they were then manufactured by agreement at the newly built U.S. government armory in Springfield, Massachusetts. Both models had worked adequately, though all flintlocks had misfiring problems and suffered all the simple physical shortcomings that afflicted handmade weapons that were pressed into continuous service—they overheated; their barrels became clogged with powder residue; or the metal parts broke, snapped, got bent, unscrewed, or were simply lost.

  This led to the second problem—because once a gun had been physically damaged in some way, the entire weapon had to be returned to its maker or to a competent gunsmith to be remade or else replaced. It was not possible, incredible though this might seem at the remove of a quarter millennium, simply to identify the broken part and replace it with another from the armory stores. No one had ever thought to make a gun from component parts that were each so precisely constructed that they were identical one with another. Had this step been taken, a broken part could have been replaced, swapped for another, because thanks to the precision of its making, it would have been interchangeable. Break a trigger in battle, and all one would have to do was fall back and get the armorer at the rear of the line to reach into his tin box marked “Triggers” and get another, ease it into place, secure it, and be back on the firing line as a fully armed and effective infantryman within minutes.

  Yet no one had thought of such a thing—except that they had. Thirty years before the humiliating debacle at Bladensburg, a new manufacturing process had been created that, had it been in operation in the United States in 1814, might well have staved off a defeat occasioned by the failure of the soldiers’ guns. The new thinking about the principles of gun making, thinking that, if put into practice, might perhaps have kept Washington from being put to the torch, began not in Washington, nor in the two federal armories at Springfield and down at Harpers Ferry, Virginia, nor in most of one of the stripling gun-making factories that had sprung up during and immediately after the Revolutionary War. The idea was actually born three thousand miles away, in Paris.

  BACK IN THE late eighteenth century, no one spoke about “the dark side.” The phrase is modern, too new for the OED. In almost all the interviews for this book, about the ultrahigh-precision instruments, devices, and experiments that indicate where the precision that originates here is likely to be going, engineers and scientists referred frequently, and usually obliquely, to what “the dark side” might be doing. Once in a while, I would meet someone who admitted to having security clearance, and would thus in theory be able to discuss in greater detail what this experiment was leading to, how this device might be constructed, what the future of such-and-such a project might be—but he would invariably grin and say that, no, he couldn’t discuss what “the dark side” was doing.

  “The dark side” is the American military, and in terms of new weaponry or research into the unimaginably precise, that tends to mean the U.S. Air Force. Area 51 is the dark side. DARPA is the dark side. The NSA is the dark side. The role of the dark side in this story is immense, but in today’s world, it is mainly to be only alluded to.

  Lewis Mumford, the historian and philosopher of technology, was one of the earliest to recognize the major role played by the military in the advancement of technology, in the dissemination of precision-based standardization, in the making of innumerable copies of the same and usually deadly thing, all iterations of which must be identical to the tiniest measure, in nanometers or better. The stories that follow, in which standardization and precision-based manufacturing are shown to
become crucial ambitions of armies on both sides of the Atlantic, serve both to confirm Mumford’s prescience and to underline the role that the military plays in the evolution of precision. The examples from the early days of the science are of course far from secret; those from today, and that might otherwise be described in full to illustrate today’s very much more precise and precision-obsessed world, are among the most secure and confidential topics of research on the planet—kept in permanent shadow, as the dark side necessarily has to be.

  IT WAS IN the French capital in 1785 that the idea of producing interchangeable parts for guns was first properly realized, and the precision manufacturing processes that allowed for it were ordered to be first put into operation. Still, it is reasonable to ask why, if the process was dreamed up in 1785, was it not being applied to the American musketry in official use in 1814, twenty-nine years later? Men were running, battles were being lost, great cities were being burned—and in part because the army’s guns were not being made as they should have been made. There is an answer, and it is not a pretty one.

  TWO LITTLE-REMEMBERED FRENCHMEN got the honor of first introducing the system that, had it been implemented in time and implemented properly, would have given America the guns it should have had. The first, the less familiar of the pair, despite the evidently superior nature of his name, was Jean-Baptiste Vaquette de Gribeauval, a wellborn and amply connected figure who specialized in designing cannons for the French artillery. He supposedly came up with a scheme, in 1776, for boring out cannons using almost exactly the same technique that John Wilkinson had invented across in England, that of moving a rotating drill into a solid cannon-size and cannon-shaped slug of iron. Wilkinson had patented his precisely similar system two years earlier, in 1774, but nonetheless, the French system, the système Gribeauval, as it came to be known for the next three decades, long dominated French artillery making. It gave the French armies access to a range of highly efficient and lightweight, but manifestly not entirely originally conceived, field pieces.* (Gribeauval did employ what were called go and no-go gauges as a means of ensuring that cannonballs fitted properly inside his cannons, but this was hardly revolutionary engineering, and it had been around in principle for five centuries.)

  The second figure, the man who did the most to bring the system of interchangeable parts to the making of guns, and whose technique was, unlike Gribeauval’s, unchallengeable, was Honoré Blanc. He was not a soldier but a gunsmith, and during his apprenticeship he became well aware of the Gribeauval system. He decided early in his career that he could bring a similar standardization to the flintlock musket, for the benefit of the man on the battlefield.

  Yet there was a difference. A cannon was big and heavy and crude—a gunner simply touched his linstock, with its attached lighted match, to the vent, and the cannon fired—and so such parts as there were proved easily amenable to standardization. With the flintlock, however, the lock (that part of a musket that delivered the spark that exploded the priming powder that ignited the main charge and drove the ball down the barrel) was a fairly delicate and complex piece of engineering, made of many oddly shaped parts and liable to all kinds of failure. To the uninitiated, the names of the bits and pieces of a flintlock alone are bewildering: a lock has parts that are variously known as the bridle, the sear, the frizzle, the pan, and any number of springs and screws and bolts and plates as well as, of course, the spark-producing (when struck by the aforementioned metal frizzle) piece of flint. To render the lock into a standard piece of military equipment, with all its parts made exactly the same for each lock, was going to be a tall order.

  The many component parts of the flintlock on a late eighteenth-century rifle were each made by hand, and had to be filed to fit.

  Cost, rather than the well-being of the infantryman or the conduct of the battle, was the prime motive. The French government declared in the mid-1780s that the country’s gunsmiths were charging too much for their craftsmanship, and demanded they improve their manufacturing process or lower their prices. The smiths not unnaturally balked at the impertinence of the suggestion, and promptly tried selling their products to the new armories and gun makers across the Atlantic in America, a move that alarmed the French government, as it imagined it might well run out of weaponry as a result.

  It was at this point that Honoré Blanc entered the picture, taking a civilian job as the army’s quality-control inspector. His brother gunsmiths expressed their dismay over the fact that one of their number was going over to the other side, was a poacher turning gamekeeper. Blanc dismissed the criticism and got on with his job, his own motivation being the welfare of the soldier out in the field rather than allowing the government to cut costs. He was greatly influenced by M. de Gribeauval, and decided he could ape his system of standardization, ensuring that all the component parts of a flintlock be made as exact and faithful copies of one perfectly made master.

  This master he made himself, carefully and with great precision, and with all the specifications laid down as precisely as possible (using the arcane system of the Ancien Régime, which still employed dimensional measures such as the pointe, the ligne, and the pouce) to tolerances of about what today we would recognize as 0.02 millimeters. He then made a series of jigs and gauges to ensure that all the locks made subsequently were faithful to this first perfect master, by the judicious use of files and such lathes as were available. The gunsmiths hired by Blanc to perform this task—by hand, still—made each lock exactly as the original. Providing that they did so, exactly, all the pieces would then fit perfectly together, and the whole assembled lock would fit equally perfectly into each completed weapon.

  Yet only a small number of gunsmiths were willing to work under these stringent new conditions. Most balked. Making guns simply by copying parts reduced the value of the gunsmith’s craftsmanship to near insignificance, they argued. Unskilled drones could do their work instead. By arguing this, the French smiths were voicing much the same complaints as the Luddites had grumbled over in England: that precision was stripping their skills of worth. This argument would be heard many times in the future as the steady march of precision engineering advanced across Europe, the Americas, the world. The kind of mutinous sentiments heard in the English Midlands half a century before were now being muttered in northern France, as precision started to become an international phenomenon, its consequences rippling into the beyond.

  Such was the hostility in France to Honoré Blanc, in fact, that the government had to offer him protection, and so sequestered him and his small but faithful crew of precision gun makers in the basement dungeons of the great Château de Vincennes, east of Paris. At the time, the great structure (much of it still standing, and much visited) was in use as a prison: Diderot had been incarcerated there, and the Marquis de Sade. In the relative peace of what would, within thirty years, become one of postrevolutionary France’s greatest arsenals, Blanc and his team worked away producing his locks, all of them supposedly identical. Blanc made all the necessary tools and jigs to help in his efforts—according to one source, hardening the metal pieces by burying them for weeks in the copious leavings of manure from the castle stables.

  By July of 1785, Blanc was ready to offer a demonstration. He sent out invitations to the capital’s nabobs and military flag officers and to his still-hostile colleague gunsmiths, to show them what he had achieved. Many officials came, but few of the smiths, who were still seething. Yet one person of great future significance did present himself at the donjon’s fortified gates: the minister to France of the United States of America, Thomas Jefferson.

  Jefferson had arrived in France the year before, to work as official emissary of the new American government alongside Benjamin Franklin and John Adams. By chance, both these men left Paris that July (Adams for London, Franklin for Washington), leaving the intellectually curious and polymathic Jefferson alone in the ferment of prerevolutionary France. A demonstration of something scientific, with possible application for his own f
ledgling arms industry across the ocean, sounded like an ideal way to spend a hot Friday afternoon. Besides, it was pleasantly cool down in the château’s dungeons, while up above in the Paris of July 8, 1785, it sweltered.

  Thomas Jefferson, while U.S. minister to France, observed the early work on creating interchangeable parts for flintlock muskets, and told his superiors in Washington that American smiths should follow the French practice.

  Honoré Blanc had arranged before him a collection of fifty locks, each gleaming in such daylight as filtered through the slit windows. Once everyone was settled on the bleachers, with onlookers paying close attention, he quickly disassembled half of them, throwing the various components of the twenty-five randomly selected locks into trays: twenty-five frizzle springs here, twenty-five faceplates there, twenty-five bridles there, twenty-five pans in another box. He shook each box so that the pieces were as disarranged as possible—and then, with a calm and an aplomb born of his supreme confidence in his method, he quickly reassembled out of this confusion of components twenty-five brand-new musket locks.

  Each one of these was made of parts that had never been joined together before—but it made no difference. Everything fitted to everything, for the simple reason that with the great precision of its making, and its faithful adherence to the dimensions of the master lock, each part was identical to each other. The parts were all, in other words, exactly interchangeable.