A. The hot iron skelp is hammered into semicylindrical shape in the swage block, and then hammered shut around a mandrel. Skilled hammering at the right heat fuses the iron into a single piece. B. A wooden cutter holder with incised spiral grooves is placed in a toothed collar, with one of the incised grooves fixed in the collar tooth. As the cutter is pushed into the rifle, it traces the incised spiral. At the end of each pass, the cutter holder is repositioned to move up one groove in the collar tooth to cut the next rifle groove. C. With the lock in half-cocked position, the shooter pours powder into the touch hole and the pan. The frizzen is snapped down to protect the powder. Soldiers near an action moved with their muskets primed and in the half-cocked position.

The third major component of a gun is the lock, which presents the most intricate of the gunsmith’s challenges. Loading a flintlock musket or rifle required the soldier to stand with his weapon upright, leaving him dangerously exposed on a field of battle for up to a minute or even more for a green soldier. The ammunition used in the War of 1812 came in paper cartridges that contained both a ball and the right amount of powder. The soldier opened the cartridge with his teeth, poured the powder down the barrel, pinched the cartridge paper around the ball to make the wadding (so explosive energy didn’t leak around the ball), dropped the wad into the muzzle, and ramrodded it tightly in place. Under cover again, he prepared for firing by pulling the hammer into the half-cocked position; then he poured a small amount of additional powder into the pan and snapped down the frizzen to cover the pan. With that, the gun was primed, the pan powder protected, and the hammer in a safe position. Near an action, soldiers moved with half-cocked muskets.

To fire, the hammer was first pulled back to the fully cocked position, which rotated the tumbler backward, compressed the mainspring, and locked the tumbler in place with the sear. When the trigger pull released the sear, the tumbler rotated forward, and the hammer slammed into the frizzen plate, pushing it open and striking sparks. The powder in the pan ignited next to a small touch hole, emitting a large smoke plume into the shooter’s face.^ae The explosion ignited the barrel powder through the touch hole, and the gun discharged. In human time, the trigger pull and discharge are virtually instantaneous.

The lock tumbler encapsulates the lock maker’s challenge. It is a complex shape, subject to great stresses, that must operate within a close tolerance range. The tumbler connects to the hammer by fitting a large square-end spindle into a square opening at the hammer base. If the fit is loose, or the corners of the fitting abrade, the hammer stroke will lose force. The surface of the tumbler’s high-friction points, like the tumbler notches, must be very hard; if they abrade so the sear slips, the lock is inoperable. But if the tumbler is too hard, it will lose tensile strength and be prone to cracking. The steel in the mainspring must be of excellent quality to retain its power. If the pivot fixing the tumbler to the stock is misaligned with the hammer rotation, the torque could twist the lock plate or otherwise interfere with a smooth action of the flint and the frizzen.

A manual craftsman like Bookout starts each lock piece with wrought iron, which is easy to chisel when it’s hot, and cuts it into an approximation of the final shape. It is then forged by hammering to expel remaining impurities and to increase its structural integrity and toughness. The forged piece would be further chiseled and filed into the final shape, largely by eye. The last step before the final grinding and polishing would be case hardening at the main friction points, by careful heating and quenching in a carbon bath.⁴

The armories’ water-powered lathes saved much time in rough cutting and grinding, while powered trip-hammers removed most of the brute work of barrel welding. (A standard sledge hammer suspended on a fulcrum and raised and released by a toothed wheel could achieve hundreds of steady-force hammer strokes a minute.) Otherwise armory guns were made more or less with the methods Bookout uses, although separate groups of craftsmen worked on each major component. The early-period mechanization chatter was mostly focused on the lock, the hardest mechanization challenge of all—which takes us back to Whitney.

Eli Whitney’s Reputational Thrill Ride

For a century and a half after his death, Eli Whitney was virtually canonized as the Father of American Technology. According to the traditional story, Whitney was the inventor of the cotton gin, which transformed the antebellum South (and unfortunately reinvigorated the institution of slavery); he was the first person to machine-produce precisely fitting interchangeable parts for muskets and was the inventor of critical new machine tools, like the celebrated Whitney milling machine.

The Whitney role in military manufacturing came under withering challenge in the 1960s. The revisionists charged that Whitney’s pretension to making arms with interchangeable parts was merely a ploy to justify extensions of his contracts. Indeed, he had little idea of how to manufacture muskets at all, much less how to blaze new trails in making them. He was unconscionably late in fulfilling his arms contract, in part because he spent so much of his time pursuing his cotton gin profits.⁵

That harsh view of Whitney as manufacturer has moderated considerably in recent years. While it’s true that Whitney made few contributions to machining technology, most of the extreme claims for his accomplishments were made by others, often long after his death. The traditional source for the story that he claimed to manufacture interchangeable parts appears to be itself a partial fabrication.⁶ While he did have a rocky start on his first musket contract, so did many other contractors. The current consensus is that Whitney was quite a competent manufacturer and one of the earliest advocates for mass production by machinery, if not expressly for interchangeable parts—in short, a respectable figure, if not the demigod of legend.

My own view is that in his early career Whitney was indeed something of a flimflam man; some recent work even raises doubts as to whether he invented his cotton gin (see Appendix). And I think the record supports the charge that he dangled the promise of machined interchangeable parts to gain extensions on his contracts. But it’s also true that he was a talented artisan and entrepreneur, and once he focused on actually building his weapons—about 1805, when he turned forty—he proved himself to be a good manufacturer and was regarded as such by his peers and armory officials. While it is almost impossible to trace weapon types to specific battles, there is decent circumstantial evidence that Whitney’s muskets were used by a good portion of the troops in some of the hottest infantry engagements of the War of 1812 and that they performed as expected.

There is nothing contradictory in such a portrait. Whitney was a hyper-talented farm boy with a modestly connected father. Older than most of his classmates when he entered Yale, he was a good engineer and metalworker, articulate, a formidable salesman, and desperate to succeed. In the first dozen years or so after graduation, he was very much on the make and sometimes played fast and loose with the truth, but as he matured and focused on a business he was good at, he did well.

Whitney finished Yale in 1792 with a vague idea of becoming a schoolmaster. The president of Yale referred him to a tutoring job on a Southern plantation and introduced him to Phineas Miller, a Yale alumnus of about Whitney’s age who had been tutoring on a plantation for several years. Whitney went south with Miller with the intention of acclimating on Miller’s plantation before taking up his own duties.

The record is mostly silent on Whitney’s first year in the South. But he never made it to his tutoring job, and a year after his arrival, he patented his cotton gin in partnership with Miller. Their business plan was to leverage control over ginning technology to create ginning centers throughout the cotton country, charging 40 percent of the ginned cotton. Whitney returned to New Haven to manufacture the gins, while Miller created and marketed the local centers. Financing came primarily from the mistress of Miller’s plantation, Constance Greene, a widow whom Miller eventually married.

The business was a failure. Their pricing was extortionate, and their gin, while a substantial advance, was relatively easy to replicate. The next few years were a nightmare of endless patent litigation, with few victories, rising indebtedness, and the looming bankruptcy of Mrs. Greene. The French war scare in 1798 came just as Whitney was reaching the end of his rope. As he later wrote to a friend, “Bankruptcy & ruin were constantly staring me in the face.... Loaded with a Debt of 3 or 4000 Dollars . . . I knew not which way to turn.”⁷

Then Whitney came across a federal circular recruiting gun manufacturers. He wrote to the Treasury secretary, Oliver Wolcott, another Yale man, that his gin factory had been idled by disruptions in trade and that he proposed to “undertake to Manufacture Ten or Fifteen Thousand Stand^af of Arms.”⁸ Since the Whitney gin had been widely publicized in the North, Wolcott responded immediately with an invitation to Washington, “knowing your skill in mechanick.”⁹ Barely a month later, Whitney had a contract for 10,000 stand of muskets on a French pattern, to be delivered in stages over the next two years. The total contract price was $134,000, with a $10,000 advance for tooling: it was the largest of the private gun contracts and the first with an advance. The only objection within the administration was that the schedule was unrealistic, which was true.¹⁰ But the schedule was furthest from Whitney’s mind: as he wrote to his friend, in the nick of time he had won a large contract by which “I obtained some thousands of Dollars in advance which has saved me from ruin.”¹¹

Whitney had minimal acquaintance with gun making, and set off on a tour of arms makers to learn more about the craft. He must have realized early that the delivery schedule was impossible, and as he missed his contract dates, year after year, he defended his tardiness by claiming that he was really engaged in a kind of R&D project: “One of my primary objectives is to form the tools so the tools themselves shall fashion the work and give to every part its just proportion—which when accomplished will give expedition, uniformity and exactness to the whole.... In short the tools which I contemplate are similar to an engraving on copper plate from which may be taken a great number of impressions precisely alike.”¹² Each time, he got a pass and sometimes even a further advance.

In Whitney lore, there was a climactic meeting in Washington in 1801 in which Whitney “made a triumphal demonstration of his musket” to a galaxy of top officials including Adams and Jefferson. There, “by allowing the officials to assemble the parts of the locks—selecting the constituent parts at random—he dramatized the concept of interchangeability and made its advantages obvious.”¹³

There was such a meeting, and it was indeed a major event, but there is little firsthand information on its details. Whitney merely reported that he went to Washington and “carried on a musket of my manufacture & several samples of Locks &c.” Whitney’s congressman also attended and wrote that the samples “met universal approbation.”¹⁴ But there is no evidence for the story that officials around the table, or anyone else, reassembled locks.

Whatever did or did not happen in Washington, however, Whitney did dangle the promise of interchangeability. There’s no other way to interpret the “copper plate” letter, and all early accounts cite it in that context. ¹⁵ A few weeks before his Washington meeting, moreover, Whitney asked for a reference letter from his friend Decius Wadsworth, a Yale classmate of Miller and the military’s head arms inspector. Wadsworth opened by pronouncing Whitney’s muskets to be possibly “superior to any muskets for common use ever yet fabricated in any country.” And went on:

[Typically] all the similar parts of different locks are so far unlike that they cannot be mutually substituted in cases of accidents. But where the different parts of the Lock are each formed and fashioned successfully by a proper machine and by the same hand they will be found to differ so insensibly that the similar parts of different locks may be mutually substituted.... [This concept] has been treated and ridiculed as . . . vain and impractical . . . [but Mr. Whitney] has the satisfaction however, now of shewing the practicality of the attempt.

That is a very large claim and, if taken literally, was simply false. Wadsworth, indeed, having made the requested representation, carefully dissociated himself from it. His next sentence is that “I am of the opinion that there is more to please the imagination than of real utility in the plan yet it affords an incontestible proof of [Whitney’s] . . . superior skill as a workman.”¹⁶

Jefferson, in any case, came away with exactly the impression Whitney was hoping for. He later wrote to Madison that Whitney had invented machinery “for making all the pieces of his locks so exactly equal, that take 100 locks to pieces and mingle their parts and the hundred locks may be put together as well by taking the first pieces which come to hand.”¹⁷ In short, something happened in Washington to convince Jefferson that Whitney had achieved effective interchangeability, and once that impression was made, it was amplified on every retelling.

It is also striking how lightly Whitney wore his promises. He had assured Wolcott, who had been embarrassed by his nonperformance, that “nothing shall induce me to shrink from the task or for a moment divert my attention from its final accomplishment.”¹⁸ But he really meant “unless something better comes along.” In the fall of 1801, his new advances in hand, he got word of the possibility of a lucrative settlement on the cotton gin patent. Whitney left for the southlands late in the year, did not return until May (long after the gin business was over), and took similar absences for the same reason over the next several years.¹⁹

A few years after that, it seems, he finally got serious, mostly giving up on the cotton gin litigation and devoting his formidable intelligence and mechanical gifts more or less full-time to the business of producing arms. While there is little evidence of his making any mechanical breakthroughs, he was a good manager and quickly got to a steady-state production of about 2,000 muskets a year. Several musket modifications Whitney made on his own initiative, like making the pan of hardened brass and designing a simplified hammer shape for easier machining, were later incorporated into the standard musket specifications. ²⁰

How did America’s muskets perform on the battlefield? The 1814 theater on the Niagara Peninsula separating Lakes Ontario and Erie was the war’s fiercest. From July through November, British regulars and American ground forces were squared off in the bloodiest and most sustained fighting of the war. The battles of Chippawa and Lundys Lane and the siege of Fort Erie were the most intense set-piece engagements, but there was nearly constant hostile contact in the intervals between. The Americans achieved a glorious tie, winning about as many head-to-head encounters as they lost: a performance that weighed heavily in the British decision to back away from the war.

The military historian Richard Barbuto, who has written the closest-to-the-ground history of the Niagara campaigns, told me that he never saw a report of a battle lost because muskets failed. Soldiers took dreadful care of their weapons, and especially broke bayonets. But while repair logistics were often difficult, the absence of postbattle complaints suggests that the muskets mostly performed as expected.²¹

Were any of the muskets Whitney’s? It seems so, for in the years just before and during the war, Whitney sold 5,000 muskets to the New York state militia, who were deeply involved at Niagara. In the early days of the war, the New Yorkers had performed poorly, but by 1814 they were battle-hardened veterans who bore a heavy share of the fighting. Unusually, the Niagara ground commander over regulars and militia alike was a New York militia general, Jacob Brown. He had earned the post by his aggressiveness in seeking out and confronting the enemy, and he didn’t hesitate to place his own state’s soldiers in the thick of engagements. Daniel Tompkins, the New York governor, was a strong supporter of the war, close to Brown, and well informed on events at the front. Tompkins knew Whitney muskets, and in late August 1814, at the height of the fighting, he intervened forcefully to break a bureaucratically imposed blockade on their sale, taking every one he could get at the state’s expense and without inspection. ^ag There could hardly be a better reference.²²

By 1814, Whitney was an established manufacturer with as much experience in musket making as anyone else in the country. He was turning out inelegant, inexpensive, quasi-mass-produced weapons that worked as they were supposed to. The postwar Whitney correspondence with Wadsworth and other armory and Ordnance officials confirms that he had become one of the half dozen or so private contractors they had come to treat as sound and reliable men. Besides Whitney, there was Simeon North, Lemuel Pomeroy, Asa Waters, Robert Johnson, Nathan Starr, and a few others. Ordnance maintained a constant correspondence with all of them, trading information on machines, waterpowers, iron and steel, and other topics of the trade. Whitney’s letters are terse and businesslike, with none of the high-flown rhetoric that the younger Whitney used to impress his funders.

In the first flush of revisionism on Whitney in the 1960s, the contractor Whitney was often invidiously compared to was Simeon North. North became a great gun maker, but earlier in his career had some misadventures of his own.

Pistol Maker to the Nation

Simeon North was born the same year as Eli Whitney, and in much the same milieu, on a middle-class but still tightly circumstanced farm in Connecticut. He had a standard farm boy’s education—no thoughts of Yale—and was hard at work from an early age. His intelligence is obvious, and from what little else is known of him, he was literate, genial, even-tempered, and taciturn. He enjoyed good health throughout his life and died after a short illness at age eighty-six. In a picture taken late in his life, he looks pure New England oak: erect, strongly built, firm mouth, looking directly at the camera. One vanity was his adoption of the title “Colonel” from a brief, and mostly honorary, stint in the state militia.

North married at twenty-one and took up farming on a sixty-six-acre plot in Berlin, about fifteen miles south of Hartford. He added to his acreage whenever he could, and in 1795, he acquired a tract with a stream, a sawmill, a dam, and waterpower rights. A natural mechanic, he kept the sawmill, added a forge, and started making scythes and other farming hardware for the local market. It is not known when he gave up farming to concentrate entirely on manufacturing.

Simeon North in Old Age.

North himself left few records, and the standard source is a 1913 biography written by two of his great-grandsons that reproduces many of the key documents relating to his arms business. More recently, the late Robert Jeska, a collector of North pistols, spent years tracking down every North reference in national and state archives to put together a near-complete story of his early pistol contracts, which he privately published along with a large trove of important documents. Along the way, Jeska clears up many gaps and errors in the official records.²³

North’s first pistol engagement came in the spring of 1799, the year after Whitney’s first contract. Like Whitney, North had no gun-making experience; he most likely purchased the contract from another craftsman, who was serving as the chief coiner at the Philadelphia mint. The contract was for five hundred cavalry pistols (the 1799s), following a French design, at $6.50 a pistol,^ah with $2,000 in advance.²⁴ That was followed up in February 1800 by a second contract for 1,500 pistols (the 1800s) and an additional $2,000 advance. He did not meet his schedules on either contract, and an unduly large portion of the 1799s were marked as “unserviceable.” ²⁵ The 1800s were completed nearly on schedule, however, and delivered throughout the country.

In June 1808, North made a new contract with the navy for 2,000 naval pistols (the 1808s), which were heavier and of larger caliber than cavalry pistols, with an eighteen-month delivery schedule and an advance of $4,000. The contract correspondence included North’s well-known statement that “by confining a workman to one particular limb of the pistol until he has made the whole two thousand, I save at least one quarter of his labor, to what I should provided I finish them by small quantities, and the work will be much better as it is quicker made.”²⁶

In the near-term, that strategy caused him no end of trouble. North devoted most of the first contract year to erecting a new factory and acquiring sufficient inventory to make all the components before assembly. In February 1809, the navy asked North if he could expedite the deliveries, because they wanted to increase the order. North responded enthusiastically. He would have “the whole of my barrels made for this contract within eight or ten weeks,” and as many as 800 were nearly ready for inspection. ²⁷

Naval procurement officers were delighted and requested 500 pistols as soon as possible. North was horrified and wrote that he didn’t have any pistols, he was still just making barrels. The navy sent Isaac Chauncey, a hard but fair man, to inspect North’s progress. There were indeed no pistols, and the barrels were disappointing—many were of different sizes and most needed to be rebored. Under Chauncey’s pressure, North continued to overpromise and underperform. He did not get his production untracked until 1810, and finally closed out the contract in early 1811.²⁸

It was a black time for North. The navy repeatedly told him that they were “very much disappointed.” Indeed, growing frictions with Great Britain had forced them to scramble up and down the coast scrounging up enough pistols to get the fleet to sea. But as the war scare eased, the navy relented, pronounced themselves happy with North’s pistols, and increased the contract by another five hundred, which were shipped in small batches until 1814. A substantial number of the 1808s went to Chauncey on Ontario, who would not have accepted them if they were defective.

Chauncey’s inspection report also makes clear that North had not yet reached the point of making standard parts. Although he made his lock forgings in bulk, they still required extensive hand-trimming and fitting work. Like the armories and Whitney, North was still at the stage of exploiting division-of-labor efficiencies, without changing basic craft processes.

By this point, North was well established as a reliable arms manufacturer, but for some reason over the next few years, his performance slipped badly. As the production of the 1808s got untracked in 1811, North negotiated a contract for 2,000 horse pistols (the 1811s) with staged deliveries over the next two years. A few months later, the record shows him soliciting the navy for further contracts, claiming he was gearing up to produce 4,000 pistols a year.

With the advent of the war, small-arms purchasing shifted to the Commissary-General of Purchases, a new office in the Treasury filled by Callender Irvine, who launched a vendetta with Whitney^ai ²⁹ and adopted North as a kind of counterexample of a good arms contractor. In the spring of 1813, although only a small number of the 1811s had been delivered, North and Irvine executed a huge contract for 20,000 pistols on a modified design (the 1813s). North got an advance of $20,000 and agreed to complete the contract within five years. He also agreed to a clause requiring that “the component parts of pistols, are to correspond so exactly that any limb or part of one Pistol may be fitted to any other Pistol of the Twenty Thousand.”³⁰

Then, in the midst of a war that was going badly, North’s output nosedived. By the end of 1814, when the war effectively ended, North had produced only 756 of the contracted 2,000 1811s, and none at all against the massive 20,000-pistol contract, despite furious letters from Irvine, who was sorely embarrassed. That was most uncharacteristic of North. While he was often late because of unrealistic contract schedules, he usually worked at a steady pace and stayed within shouting distance of the original deadlines.

There are enough straws of evidence to prompt a guess as to what was happening. North had spent $100,000 on a new factory, a high price for the day. He had also indirectly complained to Irvine that the market price of pistols had risen well above his contract price. And there were also several instances of his selling, or trying to sell, pistols to various state militias. Finally, after the war he wrote to a naval procurement officer that he had 1,400 1811s in inventory that had never been inspected.³¹ Taken together, the evidence suggests that North may have been in financial trouble and was withholding pistols from Irvine in the hope of selling them on the open market, where he could get a better price without a deduction for advances. That wasn’t illegal, but it was the kind of sharp practice not usually associated with North. Nevertheless, as we saw with Connecticut clock salesmen, the rest of the country had learned to be wary when doing business with Yankees.³²

The war ended in early 1815, with the 1811s in limbo, and the 1813s in complete abeyance.^aj Irvine was dismissed and Decius Wadsworth central-ized army weapons procurement within the Department of Ordnance. He renewed North’s big 1813 contract, with some design changes (the 1816s), and approved a $25,000 advance (North had asked for $60,000). Wadsworth cautioned North against accepting any other contracts and warned other branches of the government that North had all he could handle. By mid-1817, North was back on track producing the 1816s, although he did receive one sharp warning from Ordnance in 1818 about “the far inferior workmanship” of some deliveries.³³

Before approving the new contract, Wadsworth asked the superintendents of the Springfield and Harpers Ferry armories to inspect North’s plant, where they suggested that North may have made progress on the famous “interchangeability” clause in the 1813 contract—although neither Irvine nor anyone else seems to have mentioned it again. According to the superintendents, North was using a standard die to make all lock plates the same size, or very nearly so, with the various lock pieces individually fitted to the standard plate. Presumably, if part of a lock broke, the whole lock could be replaced by another of the same plate size, and most of the time it would work. That would have been near-interchangeability at the subassembly level, a very intelligent approach. Modern tests of several 1811s, 1813s, and 1816-model pistols produced in this factory, however, did not find that their locks interchanged. But in the longer run, as we will see, North did play an important role in achieving strict uniformity and interchangeability of firearms parts.³⁴

The Military Thinks Long Term

The 1815 decision to centralize all small-arms procurement in the Ordnance Department was the product of intense personal lobbying by Wadsworth. He and his deputy, Lt. Col. George Bomford, had been at the epicenter of the wartime procurement chaos—the multiple overlapping lines of authority, the turf wars and infighting, the lack of standards. Irvine at Treasury had been a constant disruption. He seemed almost to enjoy overruling military specifications and frequently refused to pay for emergency field purchases.

Wadsworth, forty-seven years old in 1815, was a member of one of Connecticut’s most prominent families, with many high-ranking military men in the family tree. Moderately wealthy, he was a career artillery officer and a creative engineer and inventor. As a young officer, he had worked under the great French artillerist Louis Tousard, the godfather of American artillery practice, author of the American artillerists’ field manual, designer of West Point, and a devotée of French pioneers of interchangeability in arms making.

Bomford was twelve years younger than Wadsworth, of obscure provenance. His mother is unknown to history, while his father was apparently a British artillery officer. A chance acquaintance with an influential West Point graduate who recognized Bomford’s talents opened the door to an appointment at the academy. Wadsworth recruited Bomford ten years after his graduation, when he had already distinguished himself as a fortification engineer and had invented a new type of cannon.³⁵ At Wadsworth’s death in 1821, Bomford replaced him as Ordnance chief, serving until 1842. He married well, gaining the political connections that were essential for success within the tightly knit ruling circles of the early United States.

Wadsworth and Bomford were bound together by their shared experience of the war. One or the other was constantly traveling during the war years, and their extensive correspondence illuminates the frustrations behind their reform agenda.

In Washington, Wadsworth’s desk was awash in matters picayune and crucial alike. He personally ordered a drum, a fife, and dress swords for a recruiting party and then rushed to Albany to inspirit a dithering Dearborn amid the military calamities of the first summer. He created the first working inventory and reporting system for army weapons, oversaw the procurement of artillery carriages at Pittsburgh through a local contractor, intervened with Irvine to get the contractor paid, and then sent Bomford out to customize the carriages. Throughout the war he dispatched a stream of illustrated instructions for younger artillery officers: how to set up a mortar bed, how to organize cannon transport, and much else.³⁶

Bomford’s letters brim with ideas: designing a standard pattern for caissons (munition wagons), creating a new howitzer, proposing standards for shot, for stocks, and for muskets.³⁷ Constantly lamenting the lack of “System,” Bomford worked on specifications for musket balls, tried to grade iron quality by its specific gravity, developed empirical standards for powder charges to extend musket life, and proved that mobility was to be preferred over throw weight in choosing field howitzers. He raged over the lack of professionalism among the quartermasters—“Waggons picked up here and there,” then followed a supply train, and found weapons strewn all over the road. Nearly every letter fumes over one absurdity or another. Repair procedures damaged good weapons. The quality of swords and bayonets was execrable: One sword will “fly into pieces at a blow ... while another bend to a quadrant.” Both he and Wadsworth spent substantial amounts of their own money setting up “laboratories”—repair depots—and tiding over unpaid contractors.³⁸

During the very tough 1814 campaign on the Niagara peninsula, Bomford ran a musket repair operation, driving the craftsmen to keep up with the flow of arms from the front and complaining of how the soldiers misused the equipment.³⁹ That same summer, Wadsworth prepared the defenses on the Chesapeake peninsula and at Baltimore against the raiders who had put the torch to Washington. He was almost certainly involved in arranging the successful artillery defense against a determined British assault on Fort McHenry in Baltimore harbor, the episode immortalized by “The Star-Spangled Banner.” Ordnance files don’t show his precise location at the time, but he was on the peninsula and had long been mentoring the McHenry artillery officers. The British mounted a serious twenty-five-hour attack, and the defense is widely acknowledged as a masterpiece. ⁴⁰

With the war mercifully ended, and procurement authority consolidated, Wadsworth and Bomford set out to fix the mess they had inherited, with the Springfield Armory serving as their primary development laboratory. ⁴¹ The armory is located at about the geographic midpoint of the Connecticut River, near the Massachusetts/ Connecticut border. By 1815, the spinning-mill industry created by Samuel Slater had enjoyed twenty years of development and growth. Lowell’s Waltham mill had opened in late 1814, adding to a glut of new mills hoping to profit from the unavailability of British cloth. The mills, along with Terry’s clocks, had force-fed the development of a native machine tool industry. David Wilkinson, one of Samuel Slater’s first partners, was one of a pantheon of early American machining greats who created the technical substrate for advances in armory and other metal goods production.

That talent pool, the surplus capital in Boston, the fine water sites along the main river and its tributaries, the easy access to the Salisbury iron region, and the convenience of riverboat shipping to and from New York combined to make the Connecticut River Valley a cynosure of an East Coast tradition of advanced precision-machinery manufacturing development.

The armory’s new superintendent, Roswell Lee, became the Ordnance point man in the valley and the third critical player on the Wadsworth-Bomford advanced-manufacturing team. Scion of one of the bluest-blood New England families, he had once been engaged by Whitney to maintain his Southern cotton gin establishments. He had been a high-ranking officer during the war, knew Wadsworth, and had lobbied him hard for a postwar civilian position in armory work.⁴²

As Springfield superintendent—he served from 1815 until his death in 1833—Lee’s wide, active correspondence with the valley’s private contractors made the armory a clearing house for technical information, tools development, and hard-to-find skills. A practical mechanic and a competent manager, he maintained a record of steady productivity improvement and good financial controls, and gradually disciplined the armory’s unruly workforce.

Wadsworth’s and Bomford’s goal at the outset was something well short of strict interchangeability of parts. They both talked in terms of “uniformity,” which probably meant little more than ensuring that all units of a firearm were built to the same design and dimensions with consistent quality and costs. Achieving even that proved far more difficult than expected.

Wadsworth kicked off the uniformity program with a June 1815 planning meeting in New Haven hosted by Whitney, with Wadsworth, Lee, James Stubblefield (superintendent at Harpers Ferry), and Benjamin Prescott (Lee’s predecessor at Springfield) in attendance. After several days of discussions, they agreed on a first step of creating a set of “pattern muskets” for a new model 1816, which incorporated a number of small improvements, including several Whitney had introduced for his militia contracts. (American muskets were based on the French Charleville, a 1763 model supplied to Americans during the Revolutionary War and updated somewhat when production began at Springfield in 1795, becoming the model 1795.) The patterns were to be distributed among the armories and all federal contractors as the template for all future production. As the model number 1816 suggested, they assumed the patterns would be ready the next year.

But it took nearly three years just to produce patterns. The job was first assigned to Harpers Ferry, but Stubblefield botched it, since he had little interest in the idea. Wadsworth reassigned it to Lee, but the squabbling within the armories and with contractors went on for another two years before a design was agreed on.⁴³

It had not yet dawned on Wadsworth and Bomford that they didn’t know how to specify a pattern.^ak ⁴⁴ The modern system of designs on blueprints with precise three-axis dimensioning didn’t exist in 1815. No rule with graduated markings, as opposed to a plain straightedge, has been identified at Springfield before 1848. Reasonably priced vernier calipers, true precision measuring instruments, were not available until the early 1850s and were not used on the Springfield shop floor until the 1870s. Ordnance’s idea of uniformity at this stage was to circulate several pattern muskets for contractors and armory workers to emulate as best they could. None of the patterns would be exactly the same at the outset, and variations would inevitably accumulate and multiply.⁴⁵

As Ordnance focused more explicitly on machine manufacturing and true interchangeability, each step forward revealed yet another abyss to be crossed. Achieving their objective would ultimately require reconceiving specification and production systems all the way down, in every detail. But one of the great advantages of military R&D is that it can take a long view. For more than thirty years, Ordnance stuck to the mission, steadily refined its objectives, and eventually got within reasonable proximity to its goal. Lee was at Springfield for the first eighteen years of the program, and Bomford for nearly the whole span.

The notion of interchangeability was definitely in the air. Jefferson had been an early apostle and was quick to climb on the Whitney bandwagon. ⁴⁶ Callender Irvine, as we have seen, made it an express but not enforced part of North’s 1813 contract. Pressure to satisfy a political audience is the likely explanation for a peculiar 1818 directive from Wadsworth to all armories and contractors that all musket parts must be made “to fit every musket,” which he surely knew was not possible.⁴⁷ When Lee was pressed by a contractor on the meaning of the instruction, he responded only that his “present station” forbade him from making unfavorable comments except to his superiors.⁴⁸

Line officers were less restrained. Major James Dalliba, who inspected the Springfield Armory for Wadsworth in 1819, wrote in his report that pursuing uniformity “precisely to one pattern in all the detail of parts” was neither attainable nor advisable and would incur a large expense. He also poured cold water on the idea that interchangeability would facilitate field repairs, since weapons became deformed with use.⁴⁹

Wadsworth was already terminally ill by the time of Dalliba’s report, but Bomford and Lee never jettisoned the objective of interchangeability, vaguely defined though it was. Their sustained attention, constant cajoling, and alertness in scouting out technologists who might contribute to this or that piece of the puzzle are still characteristic of the kind of successful military development programs that have recently produced the Internet and many developments in semiconductors and other advanced technologies.

A fine illustration of the Bomford-Lee system in action can be seen in the early career of Thomas Blanchard, perhaps the greatest of a handful of outstanding American inventors in the first half of the nineteenth century.

The Machine Geek

Thomas Blanchard was the classic nerd, a technology geek, but since he came of age in the Connecticut River Valley in the early 1800s, he was a machine geek. An indifferent student with limited social graces—he was afflicted with a bad stammer—his father early despaired of turning him into a farmer. As a teenager he was shipped off to work for his eldest brother, who ran a tack factory—and Thomas had found his milieu. His first job was hand-fixing heads on tacks, which he hated. He quickly invented an automatic tack counter to eliminate his record-keeping and then proceeded to eliminate his job by inventing a tack-making machine that turned out five hundred tacks a minute. He patented his tack machine in 1817 and sold the licensing rights for $5,000—a stupendous sum for a young man. That money allowed him to buy his own manufactory in Millbury on a site with waterpower privileges, in an area with some forty water-powered mills and factories already in place.⁵⁰ Blanchard must already have had a local reputation. His first patent, for a wool-cloth shearing machine, had been awarded in 1813 when he was twenty-five, and the year after he opened his shop, he was called for a consultation by Asa Waters, one of the region’s gun-making elite. Waters had several patents, including ones for a trip-hammer barrel welder^al ⁵¹ and a lathe to produce a tapered barrel. He was struggling, however, with the challenge of creating a lathe that could machine the “flats and ovals” required for a musket’s breech end.

According to Waters’s son, who was later an important manufacturer in his own right, Blanchard listened to the problem and then “glanced his eye over the machine, began a low monotonous whistle, as was his wont through life when in deep study, and ere long suggested an additional, very simple, but wholly original cam motion . . . which upon being applied, relieved the difficulty at once.” (A cam is an accessory that can modify a circular motion into an elliptical or linear path.⁵²)

Blanchard was also well known to Lee. With his ear always to the ground for new technology, Lee quickly learned of Blanchard’s flats and ovals machine and invited him to build one at Springfield along with a draw-grinding machine—a new machine type for fine-sandstone barrel grinding just before buffing.⁵³ Both machines were also later installed at Harpers Ferry. We also know that in 1820, Blanchard had a machine-building contract with the Boston Company, the nine-hundred-pound gorilla of the region’s technology buyers.⁵⁴ Any Waltham-Lowell machine-building contract would have gone through Paul Moody, arguably the top machinist in the area, who normally manufactured his own machines and also had a profitable sideline selling them to other textile firms. A Moody outsourcing to Blanchard was a gold-star endorsement.

Blanchard’s lasting fame is based on the Blanchard gun-stocking lathe, a truly original manufacturing breakthrough with broad implications for all machining of irregular shapes. Once again, the junior Waters tells the story, for the stocking machine arose from his father’s new flats-and-ovals lathe. Delighted with Blanchard’s solution, the elder Waters exclaimed, “Well, Thomas, I don’t know what you won’t do next. I should not be surprised if you turned a gun-stock!”⁵⁵ When Thomas stammered out that he would like to try, the shop workmen broke into guffaws. Making gun stocks had long been a serious bottleneck at government armories: the variety of curves and the multiplicity of recesses and connection points made it impossible to machine.

As Waters tells the story, Blanchard mulled the problem until one day, on a trip home, “the whole principle of turning irregular forms from a pattern burst upon his mind.” A neighbor reported that Blanchard stood in the road shouting, “I’ve got it! I’ve got it! I’ve got it!,” while a passing farmer muttered, “I guess that man is crazy.”⁵⁶

Blanchard’s breakthrough was as simple as it was brilliant. He constructed a lathe with two distinct parts, each separately powered. The first was a V-shaped frame holding on its two branches the target block of wood and a copy, usually in metal, of a finished gun stock, the “pattern.” They both rotated slowly and identically while moving back and forth on the horizontal axis. The second part of the machine comprised the cutting tool, geared to revolve at a high speed, connected on a similar frame to a tracer, just a freely moving wheel. The tracer wheel rested against the pattern, while the cutting wheel rested against the wood block. As the pattern rotated and moved longitudinally, the tracer wheel undulated with the pattern shape, imparting the same action to the cutting wheel—and voilà, with just a few passes, the target block assumed the shape of the pattern.^am

Blanchard perfectly understood that he had solved a general problem—how to machine any irregular shape at all. As usual, Lee heard of it and, even before it was finished, reached out to Blanchard in January 1819. Blanchard responded:

Yours of the 21 int has come safe to hand you wished me to wright you respecting macenary I conclude you mean a machine I hav recently invented for turning gun stocks and cuting in the locks and mounting. Doubtless you have herd concerning it. But I would inform you that I have got a moddle built for turning stocks and cuting in the locks and mounting. I can cut a lock in by water in one minute and a half, as smooth as can be done by hand. The turning stock is verry simple in its operation and will completely imitate a stock made in proper shape, I shal bring the moddle to Springfield in the cours of three weeks I shall want your opinion of its utility.⁵⁷

The Blanchard profiling cutter was a highly original pairing of a moving high-speed cutting tool and a moving workpiece. In the top drawing, both the workpiece blank and the metal pattern rotate and move back and forth on their horizontal axes, while the cutting tool replicates the motion of the feeling piece against the pattern. Once established, the same concept can be applied in a nearly infinite number of arrangements, as illustrated by the second drawing. The invention was later widely replicated in generations of profiling milling machines.

Blanchard demonstrated the machine at Springfield in March. Lee was much impressed and, along with Wadsworth and Stubblefield, arranged for a demonstration at Harpers Ferry in June. Blanchard took two machines with him: one was installed at Harpers Ferry, and the other hauled to Washington to demonstrate for Wadsworth.

A contract with Springfield, however, had to await the resolution of a patent dispute filed by Asa Kenney, a brass founder across the river from Blanchard. Kenney had reason to be unhappy with the dispute process. The patent commissioner set up a three-man panel that included himself and Lee, who was arguably Blanchard’s sponsor. Taking no chances, Blanchard kept Lee closely informed of Kenney’s intentions during the pendency of the hearing and hired one of the most powerful lawyers in the state, Levi Lincoln Jr., later a governor and a Congressman, to represent him. The hearing was quickly resolved in his favor. ⁵⁸

Blanchard was clearly no naïf, and he always bargained hard with Lee. When he was in Washington with his stocking machine, he even had the effrontery to wangle a letter from Wadsworth ordering Lee to buy it.⁵⁹ But then there is a charming series of letters with Lee after Blanchard had begun installing the machine at Springfield. Blanchard pleaded a “grait want of fifty dollars of money,” since he’d been buying machine castings. Lee replied that he would be delighted to pay him but needed a bill. A shocked letter from Blanchard shows him to be quite ignorant of billing processes. Lee finally prepared the bill for him and sent it to him for signature. ⁶⁰ (Clients like Waters probably contracted with a handshake and paid in cash.)

The spring and summer of 1820 was something of a crisis for Blanchard, for the Springfield stockers presented a solid wall of opposition to the stocking machinery. Adonijah Foot, the master armorer at Springfield, reported to Lee in early 1820, “The turning of Stocks progresses very well but I think the Machines for cutting in the lock and the one for jointing the face of the Stock will not be of very great advantage.”⁶¹ When Lee passed that on, and expressed his own worries, Blanchard sent an alarmed letter hoping that he had not complained to “head quarters.” (Lee in the meantime had been proselytizing hard for the machine to the navy and private contractors.⁶²) Blanchard had reason to worry. A turndown at Springfield could kill the machine’s prospects throughout the industry, and he promised Lee that he would have early solutions for all the difficulties.⁶³

Blanchard did have a serious problem with the locks. His lock-cutting machine probably worked fine, but like all surviving models, it cut a standard seat for a standard lock plate. Springfield lock plates, however, were handmade and varied considerably in size. Oddly, rather than make the lock makers conform to a standard plate—as he had seen at North’s plant—Lee left it to Blanchard to machine a standard stock to fit whatever lock plates the artisans turned out, which seems benighted. Conceivably, since Lee’s tenure at Springfield had been dogged with labor disputes, he may have been reluctant to precipitate another fight.

Blanchard being Blanchard, he solved the problem anyway. It must have occupied him for the rest of the year, for the first working solution was installed at Lemuel Pomeroy’s shop in Pittsfield, where he had gone at the end of 1820 to create a stocking system. The following February he wrote triumphantly to Lee (italics added):

I have got the Machine in good working order at Pitsfaeld, I have made greate inprovements in cuting in the work, I can cut in the whole lock with great dispatch and exactness let the variation of the plate be as it may. I can make a good joint to every lock, I can cut in side plaete and heel plates, I have discovered a method by which I can vary the jig and set it to evrey lock part side or heel and make a good joint it is done by a verry simple method I am about to commence building a machine for the above mentioned purpose and will practice on the same in my shop until I can do as good work as can posabily be done by hand. ⁶⁴

Charles Fitch, a census analyst who produced a seminal 1883 report on the development of small-arms machinery, described the “curious invention” thusly: “Blanchard devised a combination of dies sprung inward toward a center, so they would conform inside to any shape of lock-plate set in the interior, while the outer ends formed a surface which was used as a former, and thus every cut in wood was made by machinery to conform with the irregularities of the metal work^” (emphasis added).⁶⁵ Unfortunately, no example of the device has survived, but it would have been used at Springfield only until lock-plates were standardized sometime in the 1830s.

The final set of machines Blanchard installed at Springfield, like the one at Pittsfield, “half-stocked” the musket. As Lee described it to Wadsworth: “What we call half Stocking is to face and turn the Stock, fit on the heel plate, let in the barrel, put on the bands, fit on the Lock & trigger plate and bore the holes for the side and tang pins;—the other half is to let in the side plate & guard, hang the trigger, make the groove & bore the hole for & fit the ramrod, let in the band springs, smooth & oil the Stock.”⁶⁶

Lee insisted on an extended bake-off at Springfield, pitting Blanchard and his crew and machine against the current stocking crew and standard methods, with the winner to be the one with lowest all-in costs. Blanchard was brought into the armory as an outside contractor, a common arrangement in nineteenth-century factories. He was paid a piece rate and paid for his own workers and machinery, but used armory space, raw materials, and waterpower. Blanchard reconfigured his machinery into fourteen different machines, each dedicated to a single operation, almost all of them completely self-acting, so he could use unskilled hands. He laid them out in a natural production flow. That was still unusual in America, but reminiscent of the Portsmouth block-making production line, which Blanchard had expressly referenced in a patent application.

Blanchard and his machines were the clear winner. Over the life of the contract, he received a gross of $18,500, a large sum for the day. The contract does not appear to have been a full-time occupation, since he continued filing other patents during its operation. His engagement was supposed to end in 1825, but the machinery was destroyed in an armory fire, and Lee strong-armed him to stay on until a replacement line was up and running. It was 1827 before he made his exit, although some of the delay was due to Blanchard’s hard bargaining before agreeing to a 9-cent royalty on each musket stock subsequently produced with his machinery.

The Blanchard production line had been completely made over by the early 1850s and extended to incorporate many of the “full-stocking” production tasks. Most of the re-work was done by Cyrus Buckland, a legendary Springfield master machinist. An original Blanchard stock-turning lathe is the only one of Blanchard’s own designs to have survived. Over several decades, mostly because of the new machinery, the time for producing a stock was reduced from a day and a half to only about an hour and a half on a timed test.

Freed from his armory obligations, Blanchard became the hardest-nosed, and most innovative, of patent managers, coming up with a great variety of licensing arrangements to accommodate particular situations. His pursuit of infringers throughout the country—in the shoe-last, hat-block, and carriage-parts industries among others—must have kept a small phalanx of lawyers at work. He personally lobbied Congress for patent extensions and won two, bringing him protection through 1862, despite protests from his licensees. He also became something of a showman. To help win his third extension, he used his machine to produce marble busts of congressmen from plaster likenesses. (A wag said Blanchard “turned the heads” of Congress.⁶⁷) He repeated the feat at the Paris Exposition of 1857, executing a bust of the empress Eugénie.

Even while he was at Springfield, Blanchard had become interested in steam transportation and had built a steam carriage before concentrating on steamboats. He built and operated a line of steamboats for the Springfield-Hartford traffic with the shallow drafts for traversing rapids that later proliferated on western rivers. He patented a number of machines for nautical woodworking and created a large pulley-block production line in Burlington, Vermont. A well-known Blanchard wood-bending machine solved a long-standing problem of breakage in objects like plow handles. Blanchard came to the solution by careful study of the internal dynamics of bent wood. Before bending the wood, his machine first compressed it lengthwise to give it greater structural integrity—not unlike forging in metalwork—so it could be readily reformed without cracking.

Blanchard was only fifty-two in 1840, when he was immortalized in Henry Howe’s collection of essays, “Great American Inventors.” He lived for another twenty-five years, was thrice married and twice widowed, enjoyed a large family, and died a sophisticated and well-traveled gentleman of considerable wealth—even as he long maintained a small workshop to make and sell decorative busts and statues. A eulogy said, “One can hardly go into a tool shop, a machine shop, or workshop of any kind, wood or iron, where motive power is used, in which he will not find more or less of Blanchard’s mechanical notions.”⁶⁸

The Quest for the Holy Grail

For all the brilliance of Blanchard’s machinery, wood was a much more forgiving medium than metal. The challenge of aligning the lock’s tumbler and sear was of a different magnitude than that of seating a lock plate in a stock. Achieving consistent interchangeability of metal parts in volume production turned out to be a tougher challenge than the early enthusiasts for uniformity had ever imagined. The practical methodologies evolved over many years, and the most important armory contribution came from John Hall, a gunsmith from Portland, Maine, and inventor of the Hall rifle.

Mastering the interchangeability challenge was not part of Hall’s original business strategy. Rather, in the manner of Eli Whitney, when he was anxious to retain a much-needed government contract, he promised he would produce machine-made rifles with interchangeable parts—but he really did it.⁶⁹

John Hall was born into an upper-middle-class family during the waning days of the Revolution. After his father’s death, he opened a woodworking and boat-building business, married into a politically connected family, and had a very close marriage with seven children. A stint in his state militia sparked a fascination with firearms, and he switched his business to gun making. In 1811, at age thirty, he applied for a patent on a new type of breech-loading rifle, which eliminated the clumsy process of pushing ammunition down the muzzle at each reload. As Hall described his invention in an 1816 pamphlet: “The Patent Rifles may be loaded and fired . . . more than twice as quick as muskets . . . ; in addition to this, they may be loaded with great ease, in almost every situation.... [Since] the American Militia ... will always excel as light troop . . . quickly assembling and moving with rapidity . . . these guns are most excellently adapted for them.”⁷⁰ The critical advantage of the breechloader, of course, was that the soldier didn’t have to stand and expose himself to reload.

But nothing came easily for Hall. In contrast to Blanchard, who moved readily from one product or technology to another, Hall was grimly focused, with perhaps a touch of the fanatic, and he could be impatient and confrontational with critics. Nevertheless, after many years of financial struggle, he obtained an armory contract that paid him an average of nearly $2,000 a year in salary and royalties for more than twenty years—a decidedly upper-class income. Total production of his rifles in all versions was about 40,000, and they were widely distributed among both state and federal troops, although their performance was controversial.⁷¹ Scholarly recognition of his achievements was similarly delayed. Until relatively recently, he was merely a footnote in a fable dominated by Whitney and others.

The first harbinger of the stony path ahead came when Hall applied for his patent. The commissioner of patents, William Thornton, notified Hall that there was a prior claim. From whom? inquired an incredulous Hall. From me! came the reply, although Thornton hastened to reassure him that he was prepared to share the rights.⁷² Thornton was a member of Jefferson’s circle, the scion of a wealthy American family, educated in Europe, a medical doctor, with artistic and cultural pretensions, and a bit of a scientific dabbler. Standard biographies treat Thornton as an accomplished inventor, for he “held patents for improvements on steamboats, distilling equipment, and firearms.” One can imagine how he got them. The story of Hall’s patent has the ring of modern machine-politics graft.⁷³

Upon receiving Thornton’s letter, Hall arranged to see him in Washington, whereupon Thornton showed him an older British breechloader that had never gone into production, and averred, according to Hall that, “he had thought of a plan which would have resembled mine & had given orders for its construction but nothing (except the drawings) had been done toward it (& they were not to be found).”⁷⁴ When Thornton made it clear that a patent would not issue unless it was in both their names, an outraged Hall appealed to James Monroe, the secretary of state, requesting a conflict-of-claims hearing under the patent law. Monroe blandly advised him not to rock the boat, because Thornton’s influence “in that case . . . would be exerted against me.”⁷⁵

To his lifelong regret, Hall caved. In the final arrangement Hall retained manufacturing rights but agreed to share licensing income. In retaliation against Thornton, however, Hall vetoed all licensing deals, which crippled the marketing of the weapon. Thornton thereafter became his nemesis, as Monroe had warned.

The central problem for breechloaders was the gas seal, and the military had great interest in any weapon that could solve it. The rear of the barrel of a muzzle-loader is tightly closed with a threaded breech-plug, so all the explosive power is focused on the projectile. Breechloaders necessarily have a working opening in the rear of the barrel. Unless it is closed quite tightly, muzzle velocity is compromised, and the shooter can be severely injured. The problem was finally solved only when the primer, powder, and bullet were prepacked in self-contained metal cartridges.

Hall’s solution placed a chamber in the breech end of the barrel held in place by lugs and a spring lock. When the lock was released, the chamber tilted up above the stock to receive the standard powder-and-ball charge. The rifleman pushed the charge into place with his thumb and then snapped the chamber back into the locked position. During firing, the positioning of the closed chamber had to be exact and tight. Hall managed that well enough to make a barely adequate gas seal. Charles Fitch considered Hall rifles from 1824 not “fine” by 1880s standards, although a great achievement for the day. “The joint of the breech-block [the chamber] was so fitted that a sheet of paper would slide loosely in the joint, but two sheets would stick.” An 1837 test at West Point showed that the rifles’ muzzle velocity was about three-quarters of that of a comparable muzzle-loaded rifle, and their penetrating power only about as third as high.⁷⁶

Bomford liked the weapons and arranged a small trial in 1816 that gave them high marks, but with the war over, interest waned. Hall then upped the ante with his promise of achieving interchangeability with precision machining, which was sure to get Ordnance’s attention. In the meantime, his wife’s family had convinced John Calhoun, the new secretary of war, to arrange more comprehensive tests. Two separate trials and a rigorous military board review in 1818–1819 ringingly confirmed Hall’s claims. The rifles proved as accurate and powerful as the standard rifle, and even more durable.^an Both rifle types scored much higher than any musket, but in ease of loading, Hall’s rifle had a two-to-one advantage over the standard rifle and three-to-two over the musket. The board rated the ease of loading “of infinite consequence in the rifle, the difficulty of loading this arm being the great objection to its more general introduction.”⁷⁷

The result was an R&D contract, somewhat like Blanchard’s. Finalized in 1819, it would have answered Hall’s fondest prayers, but for a near fatal catch-22 that plagued the rest of his days. He was awarded a salaried armory position as director of a “Rifle Works” with an appropriation for equipment and a workforce and, to boot, a $1 royalty for each delivered rifle. But the contract had to be performed at Harpers Ferry rather than at Springfield. Harpers Ferry was the Southern armory, heavily politicized in part because of its proximity to Washington, financially corrupt, and never as technically aggressive as Springfield. The Harpers Ferry superintendents predictably undermined him at every turn—skimming his appropriations, shortchanging him on equipment and space, filing endless complaints about the wastefulness and ineffectiveness of his methods—while Hall slowly and steadily made genuinely important advances in the mass production of precision-metal parts. He later conceded that his own naïve underestimate of the challenge lent credibility to his critics: “I was not aware of the great length of time that would be consumed . . . to effect the construction of the arms with the perfect similarity of all their component parts.... I had been told it had been pronounced impossible by the French Commissioners . . . and I know that all attempts to effect it in Great Britain and this Country had failed; but from an unswerving reliance on my own abilities I expected to accomplish it in a short period.”⁷⁸

The precise extent of Hall’s achievements will never be known, for all of his patent drawings and all of his machinery have been lost. His contributions to the art of specification and inspection, however, are indisputable. Accurate mass production required much more than better machines. It was essential first to define the target product with great precision. Once an ideal model had been constructed, all subsequent specifications should be taken only with reference to that ideal. Hall insisted on special purpose machines for each part and also special purpose machines to make the production machines. Placing and fixing a part in a machine required the same attention as the precision of the machine itself; it was essential, for instance, that every operation on a part be controlled from a fixed bearing point.^ao ⁷⁹

In his drive for exactness, Hall also made substantial contributions to the technology of gauging. Nineteenth-century gauging consisted of molds, or receivers (a good part should fit snugly into the gauge); groove and hole gauging (the gauge fits snugly into the part); thread gauging; and limit, or go/no-go gauging, like barrel diameter plugs (plug 1 must fit, and plug 2 must not fit). Under Lee, Springfield was an early leader in gauging systems, but it was still not until 1821 that the armory began to insist on exactitude in barrel bore diameter, outside muzzle diameter, inside diameter of the bayonet socket, and the form, dimensions, and screw-hole placement of the lock plate. And even then, it took another two years to manufacture sufficiently accurate gauging to enforce the rules. By about 1823, Lee had a set of eleven gauges that should have allowed reasonable control—although he was apparently not yet attempting to enforce uniformity on lock plates.⁸⁰

Hall carried gauging much further. Precision gauges were constructed for every measurement: there were reportedly some sixty-three separate gauges for the rifle, leaving nothing to a workman’s judgment. Hall’s gauges were always made in three sets, one for workmen, one for inspectors, and a master set in the plant manager’s office to monitor wear on the other two.

Hall’s influence can be seen in the tenfold increase in the number of Springfield gauges between 1815 and 1845. French influences were involved as well, for Springfield had multiple French contacts, but Lee and Bomford were well aware of Hall’s work, and senior artisans rotated through Springfield, Harpers Ferry, and private contractors like North. Hall’s final fillip was inspections of inspections, to ferret out any nonconforming part. Hall may also have been unique among the military arms designers of the day in making machinability a consideration in his weapon’s design. Better design could have greatly simplified the machining of the Charleville, for instance.

It is generally accepted that Hall made substantial contributions to forging, milling, and cutting machinery, although the details are mostly missing. He greatly expanded the application of drop-hammer die forging—shaping hot or cold metal in a steel mold with blows from a mechanized hammer—which wasted much less metal than rough-forging and grinding, and he worked on techniques to retain the dimensions of the forged shape during cooling. Most notably, Hall created a wide range of cutting machines, including a profiler milling machine^ap in which the work table was moved in conformity with a profile, similarly to the way Blanchard’s gun stocking machine operated.

A. Hall invented one of the very first breechloading rifles that had acceptable performance, although its reception was limited by its indifferent gas seal, and maintenance problems associated with rust and powder fouling in the loading breech block. B. Far more important were Hall’s mechanical contributions in die forging, milling, and in principles of machine construction and management. Hall also pushed gauging practice to new levels of precision and rigor. Note that the gauge in the bottom drawing includes more than a dozen gauging fixtures, each addressing a particular part.

Almost all of Hall’s machines were self-acting: after the workpiece was loaded and the power train engaged, the machine moved the tool or workpiece until the operation was complete. Semiskilled young men could often run several such machines at a time under the direction of a “Principal.” Hall also lavished attention on dampening vibration and chatter in his machines: they were typically much heavier than standard machines, with bases of iron rather than wood, redesigned drives and spindles, wider belts to improve trueness, and gauging to track drift from accuracy. James H. Burton, a former master armorer at Harpers Ferry, said that the Hall works housed “not an occasional machine, but a plant of milling machinery by which the system and economy of manufactures was materially altered.”⁸¹

It was not until 1824, about five years from his starting at Harpers Ferry, that Hall could finally invite Calhoun and Bomford to examine a production run of rifles manufactured on his principles. They could see for themselves “the manner in which the several parts, promiscuously taken, came together, fitted and adapted to each other.” But Congressional complaints driven by the carping at Harpers Ferry forced Bomford to suspend all production activities pending a full field trial of the rifles and an external review of Hall’s manufacturing methods.⁸²

Two more years were consumed convening the review boards and completing the investigations, but the final reports were stunning vindications of Hall. After a five-month field trial, the military board expressed “its perfect conviction of the superiority of this Arm over every other kind of Small Arm now in use,” and supplied a statistical analysis of its great advantages in speed of firing, accuracy, and durability.

The manufacturing review was even more glowing. Hall’s system was adjudged to be “entirely novel” with “the most benefitial results to the country.” The inspectors, who were all experienced, had never before seen arms “made so exactly similar to each other . . . [that] parts, on being changed, would suit equally well when applied to every other arm.” They conducted an experiment of freely intermixing parts from two hundred rifles drawn from different annual production runs and found that “we were unable to discover any inaccuracy in any of their parts.” The reviewers also made particular note of Hall’s poor working conditions and hoped that he might “receive that patronage from the Government that his talents, science, and mechanical ingenuity deserve.”⁸³

The board’s hopes were in vain, and the sniping from Congress and Harpers Ferry continued, but the report saved Hall’s contract. The first substantial order for Hall rifles, in 1828, came from state militias, which the armories by law could not supply, so Bomford directed it to Simeon North, although he deputed Hall as his inspector. To Hall’s great frustration, Bomford sent North one of the Hall rifles as a pattern—so even Bomford had missed the point that a “pattern” is a rigorously created and maintained ideal, not just a sample unit pulled from the production line.

The relationship with North got off to a rocky start when Hall arrived at Middletown with his full panoply of gauges and pronounced North’s output unacceptable. But North, of all gunsmiths, could understand what Hall was about, and as he came to appreciate Hall’s achievement, reproduced the system in his own factory—interestingly, with different machines and different gauges. It took another several years, but by 1834, Hall and North had fully demonstrated to the War Department that parts from both Middletown and Harper’s Ferry could be “promiscuously” intermixed and readily reassembled into perfectly functional rifles.

Modern comparisons of Hall rifles confirm that the breech parts, which were the most demanding construct, were in fact interchangeable both between the North and Hall plants and between different rifle vintages. Tolerances on the breech opening among rifles in new condition were between 0.002 and 0.004 inches. In that same period, Joseph Clement was reportedly striving for 0.002 tolerances in constructing the Babbage calculating engines.⁸⁴ So Hall and North were working at the very edge of the era’s machining technology.

By then Hall was in his mid-fifties and increasingly ill, possibly from tuberculosis. His rifle was slowly becoming obsolete and was soon eclipsed by more modern weapons from gunsmiths like Christian Sharps—the Sharps rifle may have been the favorite of Union troops—and B. Tyler Henry, whose Henry rifle was a prototype for the long-running Winchester. Hall quietly continued on salary at Harpers Ferry, tinkering with his system until his death in 1841. His place in the story gradually faded into a mere footnote. As one popular history written in the 1950s put it, “by 1820, Hall, using Whitney’s techniques of interchangeable manufacture, was turning out his rifles at Harpers Ferry.”⁸⁵

Production of the Hall rifle ceased in 1848. Despite some passionate advocates among the officer corps for its accuracy and speed of loading, the weapons seem not to have been much favored by troops. In the Second Seminole War of 1836, which was fought in the Florida Everglades, troops complained that the Hall loading chamber quickly rusted shut and that spilled powder collected in the bottom of the loading chamber, corroding the iron and forcing the chamber out of alignment.⁸⁶ In theory, such problems could be easily solved. The military issued small wire brushes to clear chambers of excess powder, and regular application of lubricant would have prevented rusting. But these are precisely the kind of fussy maintenance tasks that soldiers in the field can be depended on to neglect. A large number of the rifles were still in inventory when production ceased, but they were mostly distributed during the Civil War.

The Significance of Armory Practice

The achievement of Hall and North in manufacturing rifles with fully interchangeable precision parts was a signal milestone toward the final realization of the Wadsworth-Bomford-Lee program of “Uniformity” laid down twenty years before. By about mid-century, practical interchangeability became a fact for virtually all military muskets produced by the armories and private contractors. But legend to the contrary, machines by themselves were still far from being able to consistently achieve such tolerances. A series of microscopic analyses by Robert Gordon showed that precision fitting of firing action components required improvements in hand-finishing at least as great as those in machining. And until very recent times, it was rarely cost effective to attempt to replace all hand-finishing with machinery. The achievement of armory practice and high-precision interchangeability, therefore, was the creation of an integrated process of specification, measurement, work flow, and the highest-quality standards in both machining and hand work.⁸⁷

The value of mass production processes was proven in the crucible of the Civil War. Firearm production at Springfield in 1860 was a bit lower than 10,000 weapons, but it was ratcheted up quickly during the war. Springfield produced 14,000 weapons in 1861, 102,000 in 1862, 218,000 in 1863, and 276,000 in 1864. Colt was the most important private-sector military arms producer. From 1861 through 1863, its peak year, Colt’s output of firearms increased from 27,000 to 137,000—or by a factor of 5—while at the same time Springfield’s rose by a factor of 21.8.⁸⁸

But the way in which the military achieved its interchangeability objective suggests its limitations. In the first place, it was a development that took many decades, which is usually possible only in the type of hothouse environment afforded by military settings. The original 1763 French Charleville musket, with only minor modifications became the Springfield Model 1795 and, with another set of small changes, the Model 1816. That was virtually unchanged until the flintlocks were replaced by percussion caps in the 1840s—they worked much like children’s cap pistols, although the caps were made of copper. (The unnecessary lock parts were machined away, and a percussion cap receptacle was placed on the breech top with a new touch hole. The hammer remained on the side but was curved to strike the breech-top percussion lug.) With the 1848 advent of the Minié ball, which facilitated muzzle loading without fouling,^aq the musket barrels were rifled. Civil War troops, therefore, mostly went to war with “rifle-muskets,” which were quite respectable weapons, with an effective range of six hundred yards, ten times that of the old smooth bores.⁸⁹

The basic infantry weapon carried by Union troops in the Civil War, therefore, was still a modified version of the Springfield 1795 musket. Such longevity is not especially unusual in military procurement. The B-52 bomber, launched in 1952 as an anti-Soviet nuclear bomber, has long since been converted to deliver cruise missiles and other standoff precision weapons. There are a number of instances in which sons of first generation B-52 pilots also became B-52 pilots, and some reports of grandsons. Military platforms, in general, like bombers and ships—and the Springfield 1795—tend to have very long lives, so long as their lethality can be steadily upgraded.

While no private company could follow such a strategy, it is well-suited for military technologies. The sheer logistics of maintaining a global military force imposes an extreme conservatism in getting the most out of existing technologies. But by the same logic, the long planning time frame justifies spending resources on promising technologies with a very long payoff. The American military started working on the basic technologies of the Internet some four decades before it finally burst into commercial prominence. In the same way, armory practice in machining laid down a substrate of technologies—including gauging, pattern making, profiling, and milling—that were seized on later and taken in many different directions by private companies.

The apotheosis of armory practice—machine production lines with special purpose machinery turning out fully interchangeable parts with little or no manual intervention—came only with the first Ford Model T assembly line in 1913. That production model dominated much of American manufacturing in the twentieth century. For most of the nineteenth century, however, highly organized production lines using precision special-purpose machines accounted for a very modest share of national output.