The Dawn of Innovation

Canadians long mistrusted American designs on their country, not without reason. As late as 1935, the American military had a “War Plan Red” for an invasion of Canada, including poison gas attacks on Halifax. There was at least one formal exercise.

Frigates, with just one gun deck, did not rank as ships of the line. Used mostly for detached duty, they were glamor commands because officers and crewmen could make large amounts of money by taking enemy prizes. American frigates were typically much heavier and better armed than their British counterparts. More importantly, overconfident British captains plunged into the early encounters as if the Americans didn’t know the rudiments of fighting. In the last of the frigate-to-frigate battles, however, the two were evenly matched, and it was the American captain who was reckless. It took only minutes for the British Shannon to destroy the Chesapeake.

The throw-weight calculation assumes a single shot from all the guns, each using a ball of the rated weight. In battle, gun crews often stuffed their cannon with multiple cannon balls, bags of scrap metal, grape shot (a half dozen small iron balls on a rod or a chain), and almost whatever else came to hand. Overweighting the payload reduced a gun’s punching power, but at close range it could increase its shattering impact on a wooden hull, and if you were attacking a ship’s crew, a deck-top blast of flying shards of metal was viciously effective.

The scavenging of army forts explains Barclay’s high proportion of long guns. Perry was likely stuck with carronades because Chauncey was picking off new long guns for himself. Perry also bought cannon in Pittsburgh, where there were two new cannon foundries, but newer foundries naturally focused on carronades, since they were easier to make.

He subsequently got the required clearances and more. The Admiralty even came up with the idea of prebuilding hulls in England and shipping them to Kingston. Yeo and Prevost agreed it was a terrible idea. Building hulls wasn’t a bottleneck, and transporting them would require completely new St. Lawrence barges and severely disrupt already-strained logistics. Wary of rejecting an Admiralty gift, they convened an advisory committee that, after an appropriate deliberation, recommended against it. By that time, the Admiralty had already started on one, which was delivered in late fall and launched at Kingston in December 1814 as the Psyche, a big sixty-gun frigate, although it never sailed on the lake.

Sackets Harbor was a cul-de-sac, unlike Kingston, which was on a free-flowing channel. The substantial permanent population created “excrementious” conditions in the harbor, plaguing the force with gastroenteral disease.

Different Iroquois tribal groups fought on both sides, and tribal elders were appalled at how fearfully they were being ground up. Indians and whites had very different ideas on the value of a soldier’s life. Whites were horrified at the Indian penchant for killing prisoners and roasting and eating their hearts and livers. But the Indians were shocked at the whites’ habit of lining up at close range in full view and firing away until one side or the other crumpled. Indians, one chief said, fought “to kill the enemy and save our own men.” A white could never “lead a war party with us.”

Most historians would accept that list, or one much like it. Scholars differ on whether one particular factor, like the early spread of a scientific outlook, the rise in British wages, or an underlying culture, was the ultimate trigger. While there is much to learn from examining each strand separately, in the real world they were tightly intertwined.

Coke is the almost-pure carbon residue from slow burning of coal. Iron ore is smelted in direct contact with the fuel to facilitate carbon binding with the mostly oxide impurities in the ore. Subsequent processing steps, like creating wrought iron or steel, also involve manipulating the carbon content of the product and are most easily accomplished by direct-contact heating. Great Britain made the shift to coke roughly a century before the rest of Europe and the United States. In part because of its greater heat potential, coke is the superior fuel for large-scale processing, so all serious competitor countries were eventually forced to follow the British lead.

Dava Sobel’s best-selling Longitude may be too hard on the astronomers. They certainly rallied against Harrison’s solutions, and several may have been motivated by personal animus toward Harrison (who was easy to dislike). But one could fairly argue that his clocks were not a true solution, despite their clear qualification under the rules. They were resounding proof that clock-based solutions were possible. But each took years to build, and at least through their first three versions, the designs were secret and not obviously reproducible. In short, they were not of much use to the navy, which was the whole point of the exercise.

The machine is of a class of instruments called “comparators,” meaning that they measure only the differences between objects rather than the dimensions of the objects. Maudslay’s micrometer was of the same type. The British were quite late to shift to dimensional measuring, leaving that development primarily to the Americans. The Providence firm of Brown & Sharpe was the world leader in dimensional measuring in the second half of the century.

The name came from Babbage’s adoption of a subtraction algorithm for computing tables. For example the sequence of squares 1, 4, 9, 16, 25 shows first-order differences of 3, 5, 7, 9, which in turn have a constant second-order difference of 2. To compute the square of 6, the engine would take the previous first-order difference (25–16 = 9) and add the constant: 25 + 9 + 2 = 36; 7² then is 36 + 11 + 2 = 49. Higher order exponents required finding higher-order differences before arriving at the constant.

The original award notice from the Exchequer said that the grant was “to enable Mr. Babbage to bring his invention to perfection, in the manner recommended.” By “the manner recommended,” the Exchequer presumably meant Babbage’s own estimate of time and cost, while Babbage insisted the phrase was an open-ended commitment to fund development until the engine reached his recommended level of perfection.

Twenty years later, Swade is heading another team to attempt to build the Analytical Engine (AE), a much more complex challenge. Since the working drawings are not complete, they will have to reconstruct them from Babbage’s “Mechanical Notation,” which team members can read fluently. Tentatively, the project will be completed in distinct stages—completing the drawings, creating a virtual AE entirely in software, and finally constructing the engine itself. Indeed, if they actually succeed in building just the virtual AE, it would answer all of the remaining questions of whether Babbage actually created a workable computer design.

The biography is an admiring memoir published in 1862, almost sixty years after the plant opened, so it is unlikely that the author had accurate information on previous contractor manning levels. At the plant’s volumes, the sheer physical work of loading logs and moving parts from station to station probably required more than ten men, not even including machine tenders.

Iron is one of the most common elements on earth, and ground water is often rich in dissolved iron. In swamps and bogs, the iron is precipitated by oxidation and bacterial action, forming low-grade ore deposits that sink into the mud. Ore raisers recovered the iron by dredging. It was then roasted and thoroughly dried (steam could explode a blast furnace) and pulverized in a stamping mill to reduce the impurities before smelting. Most ancient iron was from bog iron, but it is no longer used for industrial purposes. New Jersey’s iron bogs now produce cranberries.

John Harrison used wooden gears in the first longitude chronometers because the oils in the wood made them self-lubricating. Brass was harder and more durable but needed external lubrication, which trapped dust and other contaminants. Harrison feared that their accuracy would degrade on an extended voyage. As he shrank his chronometers, small-dimension machining mandated the switch to brass components. His prize-winning chronometer was watch-sized and had all brass works.

Because of nuances of the earth’s orbit, the length of a day differs from the canonical 24 hours by + 16 to–14 minutes throughout the year. Mechanical clocks are based on solar mean time, which treats all hours and days as if they were the same length. When most time was taken from sundials or solar reckonings, mechanical clocks were glaringly out of sync with “real time”—hence, the equation clock. The conversion formula is both complex and tedious, although it is likely that Terry worked it out for himself. As mechanical clocks proliferated, the industrial world quietly chose to forget about solar time.

Americans were painfully retaught that lesson by Japanese manufacturers in the 1970s and 1980s. Japanese products were designed for manufacturability from the start. In America and Europe, by contrast, product design engineers and manufacturing engineers rarely talked to each other. Part counts in Western products were invariably far higher than in Japanese versions, and welds and fittings often in unnecessarily awkward places.

Terry’s patents refer to the escapement mechanism as the “verge,” an older term that actually referred to a different escapement type but reflected the usage of wooden clockmakers. I’ve used the modern term to avoid confusion.

Nineteenth-century upper-class work schedules were generally quite leisurely, but commercial business was inherently stressful because of the large fraction of one’s wealth constantly at risk. Ships were lost, commodity prices see-sawed, trade bills defaulted. Cycles of credit expansion and contraction regularly wreaked havoc in all merchant communities.

It does not dim Lowell’s accomplishment to doubt the conventional tale that he returned from England with a complete design of a power loom in his head. It is more plausible that he returned with a few sketches and a firm conviction that power weaving was readily achievable, and that Paul Moody then designed and built the looms. Moody, after all, was both a skilled weaver and a great machinist, who probably needed only Lowell’s backing to develop a working loom.

The primary advantages of the breast-fed wheel were, first, that it was always moving in the same direction as the water, whereas in the overshot wheel, race water could spill down its back and slow the returning wheel buckets. Secondly, the diameter of an overshot wheel was limited by the height of the mill race, while the breast-fed wheel could be of almost any size, and the water entry point could be adjusted to different levels of water flow.

Burden was most famous for his horseshoe making machine. A continuous stream of hot bar iron was fed into a round contraption two and a half feet in diameter that executed a dozen high-speed swaging and punching actions that produced a finished horseshoe every second. During the Civil War, Burden supplied all the horseshoes for the Union army at a price only pennies above the cost of the iron.

The naming convention for Francis turbines was established by others, mostly after his death, in recognition of the consistent impetus he provided over many decades toward smaller and more compact turbine designs with ever higher energy-conversion ratios.

The phrase “lock, stock, and barrel” was a favorite of my grandmother, as in “She left and cleaned him out lock, stock, and barrel.” I knew what she meant, but the phrase conjured up an odd pastiche of door handles and wooden kegs.

If there was insufficient powder in the pan, or the touch hole was fouled, the powder in the barrel would not ignite—and the soldier would get only “a flash in the pan.” If the sear or the tumbler was worn or improperly aligned, a primed, half-cocked gun might discharge by accident—“going off half-cocked.”

A “Stand” of arms was a musket and necessary auxiliary equipment, which in federal contracts usually included the musket, bayonet and holder, and the ramrod. Other equipment like ammunition and accessories was purchased separately. Musket contracts also often outsourced the bayonet knives to specialists like Nathan Starr.

Pistols were usually contracted in pairs (by the “brace”). North’s first contract specified 250 brace of pistols at $13 a brace. References in the sources are not always consistent, however, so I use a single weapon as the basic unit.

By this time Whitney had long since closed out his initial contract and was steadily turning out muskets under both federal and state contracts. Irvine was pushing a new musket design, in which he may have had a financial interest. Whitney was properly scornful: this was wartime, and supply problems were rife; pressing for a revised musket design was idiotic. The muskets Governor Tompkins bought from Whitney for the 1814 Niagara campaign had been sitting in a warehouse because Irvine and his favorite inspector, who was the designer of the new musket, refused to inspect them.

Simeon North’s 1813 pistol contract, for instance, came with a pattern and a list of exceptions, e.g., “Hammer pin [is] too large ... the threads on the small lock pins to be increased about one size . . . the cock pin and jaw too loose.” Even that was more precise than any of Whitney’s contemporary contracts.

Prompting the reasonable complaint from a competitor, “I should not think of gitting a patent for . . . applying a trip to welding a gun barel any moure than plating a scythe or a hoe it seems to me that a strange fanatism has operated on sum peope for gitting patents for some simple things.” The complaint anticipates today’s controversies over software “business process” patents. The trip-hammer had been around a long time, so Waters appears to have patented the application to barrel welding, rather than the tool.

The frame arrangement described here was used for gun stocks. The original patent application shows a shoe last, a more compact shape, and is arranged differently. The application identified a variety of possible arrangements for different product types.

As Robert Gordon has pointed out, interpreting such results is uncertain without knowing all the context of the tests, the target audiences, and hidden agendas. Note the wide variance between this report and the one at West Point.

Hall hypothesized a man cutting a plate exactly square. Normally he would “Square the 2d side by the 1st, and the 3d. by the 2d., and the 4th by the 3rd., but on comparing the 4th side with the 1st, it would still be found that they are not square; the cause is that in squaring each side by the preceding side, there is a slight but imperceptible variation and the comparison of the 4th with the 1st gives the sum of the variations of each side from a true square. And so in manufacturing the limb of a gun,” and he explains the process of measuring every operation on a part from the same fixed point.

Milling machines evolved from older rotary filers. A flat workpiece was clamped to a movable table under a fixed vertical spindle-cutter with a rasp end. In operation, the table moved the workpiece rapidly back and forth on one axis and more slowly on the other to achieve an even reduction. Miller cutting tools rapidly evolved into a great variety of shapes including both end cutters and side cutters to cut drill flutings or screw holes. Workpieces could be rotated as they were cut or fixed at various angles against the tool, using either vertical or horizontal spindles. By the 1870s, “universal” milling machines could cut metal into almost any shape at all. Almost all the development took place in the United States.

The Minié ball was shaped like a bullet, with a hollow base that expanded on ignition to grip the rifling grooves. Loose-fitting Miniés could be easily pushed down the barrel but still have a tight spin for good range and accuracy.

In this chapter, I use “West” in this sense, for it was the usage of the time. In the next chapter, which is more centered in mid-century, I will call the same region the Midwest, in accord with the newly current usage.

Gouge insisted that Biddle’s success in managing monetary crises was a compelling argument for withdrawing the BUS charter, since it necessarily implied an excess of power and discretion vested in a single individual.

Besides his engineering accomplishments, Shreve spent fifteen years as superintendent of western rivers. In that capacity, he invented several large engines for clearing debris from river channels. His greatest accomplishment was to clear the Great Raft, a 150-mile-long tangle of submerged logs and fallen trees that closed off the Red River to navigation and, as it silted, threatened to turn the river basin into a giant swamp. Shreveport, located where the far end of the raft had been, was founded and named in his honor.

Ironically, the Clermont/Steam Boat was of a western-style flat-bottom design. Fulton, indeed, seems to have targeted the western rivers from the start. But as the Hudson trade expanded, Fulton/Livingston adjusted designs to fit the local market and then transferred them wholesale to the West.

Although Oliver Evans was the co-inventor of the high-pressure engine and opened a Pittsburgh factory in 1812, he and his son, who ran the western business, won little steamboat business. The Evanses dominated western mill and factory engines, but Shreve and French, who were steamboat men first and engine makers second, were far ahead in adapting engines to the special demands of western steamboats.

Trollope was “forcibly” reminded of Milton’s lines:

Blind mouths! That scarce themselves know how to hold
A sheep-hook, or have learned aught else, the least
That to the faithful herdsman’s art belongs!
—But when they list their lean and flashy songs,
Grate on their scrannel pipes of wretched straw:—
The hungry sheep look up, and are not fed!
But swoln with wind, and the rank mist they draw,
Rot inwardly—and foul contagion spread.

In a famous analysis, Robert Fogel made a case for the potential economic equivalence of railroads and canals. The capital costs of canals were considerably less than railroads, but they externalized their equipment and services costs. The Erie Canal was almost as long as the Erie Railroad, but was completed in less than half the time—which speaks more to the turmoil at the railroad than the inherent difficulty of the projects.

Data assembled by Robert Fogel show that in 1700 an average American male consumed 2,300 calories over and above that required for basal metabolism and vital maintenance functions, like eating. A French or British worker’s diet would have furnished only 400–750 “extra” calories, respectively, if they had been the same size as Americans. How much their smaller statures compensated for the gap is not known. Anecdotally, British sailors noted their own superior staying power against French sailors during the Napoleonic wars, presumably because of the poor diets of the Frenchmen.

The study I am relying on here is a careful one, by two respected economic historians, an Englishman and an American. Inevitably, as data are collected and filtered and run through spreadsheets, they produce output at second-decimal precision. But given the nature of the data sources, any such conclusions are very approximate. Their study concludes that American worker productivity across all industries was about 95 percent that of British workers, which is tantamount to equality within reasonable margins of error.

Nugent and Bird remained in touch after she left. Within a year, however, Nugent was shot and killed by Griff Evans, another friend of Bird’s, whom she had characterized as “weak.” Evans was hoping to open a tourist hotel; Bird was his first and possibly only guest. The lemon-gloved Englishman who had arrived on the stage represented a British investor, the fourth Earl of Dunraven, who hoped to create a private British hunting preserve. Evans became his agent and was strongly opposed by Nugent and the majority of the locals, who justly regarded it as a land steal. Evans was reportedly drunk when he killed Nugent, and the shooting has been labeled “cowardly,” although the details are murky.

Business historians have shown that Chicago’s giant postbellum meatpacking plants were far larger than scale efficiencies warranted. Competition should have led to a redistribution of slaughterhouses around interior rail centers, closer to the herds. The packing oligopoly led by Armour and Swift colluded to forestall such a redistribution.

The normal practice was to use reciprocating (up-and-down) blades, which could not cut nearly as rapidly. The large surface area of the circular-saw blade, its high speeds, and temperature differences between the center and the periphery created sufficient complexities to require a highly skilled operator.

A compound engine, often called a “double engine” in this period, combined both a high-pressure (expansive) and low-pressure (vacuum) engine. The excess steam from the high-pressure component was piped into a Boulton & Watt–style apparatus, improving the power-to-fuel consumption ratio.

High-quality iron was much more expensive in the United States than in Great Britain, making iron ships uneconomic except for military uses. Rough iron and steel price / quality parity was not achieved until sometime in the 1880s.

The “Master Mechanic” problem was the industry’s name for the proliferation of design shops at the larger railroads. In fact, few of the customization demands were frivolous but rather reflected the steadily improving engineering capabilities at the bigger lines. Different lines operated over quite different terrains, often with different average-weight loads. Procurement engineers were learning to specify optimum power outputs and other features to suit their operating environments.

I have so far used “Baldwin” as if the man and the company were the same. After his death in 1866, however, the company had many different names, as the successive partners changed. “Baldwin Locomotive Works, Inc.” was adopted only in 1909, but for convenience I use “Baldwin” throughout the text.

Pilot boat captains provided pilots to guide large ships into the New York Harbor. Each captain had a stable of pilots and competed by being the first to the best customers. The trade was well-paid, so large rewards went to the daredevil captain with the fastest boat. Brown’s Mary Taylor was a sixty-six-foot ocean schooner.

They got used to it. Stevens’s syndicate donated the cup to the New York Yacht Club in 1857, on the condition that the club would accept international challenges from time to time, with the cup to be held by the winning nation under the same stipulation. The first challenge, from a British syndicate, was in 1870. Competitions continued intermittently, usually against British challengers. The Americans turned away all comers, until an Australian team captured the cup in 1983, ending a 132-year winning streak.

Root and Colt had met when Root was still an apprentice. Colt was giving a demonstration of an underwater torpedo, which then went awry and sprayed water and mud on the crowd. Root prevented the crowd from inflicting bodily harm on Colt. Colt also paid a visit to Collinsville before he set up his first gun venture and took notes on the machinery, so he would have formed a good impression of Root’s capabilities.

The investors complained that Colt had fraudulently concealed the problem that multiple firing chambers often ignited simultaneously. The board was dominated by Dudley Selden, a friend of Colt’s father, who had acted as Sam’s lawyer in forming the company. He engineered a cash call among the investors, and cut off Colt’s salary. Colt had no money, and his failure to meet the call would have triggered the loss of all his stock and control over his patents. (Venture capitalists often resort to such devices to squeeze out inconvenient founders.) Colt’s alleged fraud was to demonstrate his weapons with only a single chamber loaded. But these were all professional investors, and Colt was a fast-talking twenty-two-year-old. If they were gulled that easily, it is hard to feel sorry for them. The dispute petered out after the bankruptcy, and Colt kept control of his patents.

In the nineteenth century, with the notable exception of the 1870s, there was usually little difference between nominal and real, or inflation-adjusted, growth rates. Long-term prices were quite stable although year-to-year fluctuations could be quite sharp.

The explanation for the apparent distress as the real economy boomed probably lies in the temporary distribution of the adjustment pain. The better-off, who were likely to be creditors, would have greatly benefited from the rise in monetary values, and it is likely that wage cuts applied first to the most vulnerable workers. The question has recently drawn much interest, but definitive answers will require a great deal of scholarly digging to reconstruct data on wage levels and household balance sheets to match the generally good census data on physical output.

The shift is apparent in a mid-century leap in American worker productivity. Even as its manufacturing sector grew rapidly, American labor maintained its roughly 2:1 edge over British workers for the rest of the century. At the same time, American transportation productivity jumped sharply, doubling that of the British by century’s end, while wholesale and retail trade made gains nearly as great. Aggregate data somewhat muffle the shift, in part because as an older population, Great Britain had a considerably higher percentage of its people in the work force.

Henry Bessemer, a prolific British inventor, had discovered that injecting a blast of cold air into a container of molten iron touched off a violent reaction that decarburized the iron within minutes. Adding a small amount of carbon back into the mixture produced excellent steel, as the chemical violence within the container ensured total mixing. A technical problem, which took some time to overcome, was that the process worked only with low-phosphorus ores. Fortunately, the vast Mesabi ore range around the Great Lakes was ideal for the Bessemer process.

All such statements reserve the significant exception of African American slavery. Even after the Civil War, freed slaves and their descendants—particularly, but not just, in the South—were kept in a state not unlike that of Russian serfdom for nearly another century.

A GE official pooh-poohed the idea that they would allow such critical technology to be exploited by China: “Why would we give away our future?” he asked. But senior bankers made that same argument in the years just before the financial crash. In most cases, they readily sold their firms’ futures for the sake of temporarily higher revenues, higher stock prices, and higher bonuses.

Hu Jintao is the “paramount leader,” holding the posts of general secretary of the Communist Party, president of the People’s Republic of China, and chairman of the Central Military Commission. Wen Jiabao is the premier of the State Council, which oversees the day-to-day operations of the government. The members of the Party Politburo and the State Council all come from a relatively small body of senior leaders, who make all important policy and personnel decisions. All important positions are filled by party members. At this writing, the Hu-Wen government will complete its second five-year term in office in 2013 and will cede its positions to a new team that has not yet been officially designated.

Chovanec worries that the government can manipulate inflation data by managing prices on goods tracked in the official consumer market basket. Impressionistically, the prices he sees every day in Beijing are rising by double-digits, but taxi fares, which are in the basket, have been held flat, causing a visible diminution of taxi service. Other data, based on physical movement of goods and the like, suggest that important sectors of the economy are contracting, even though reported growth continues to be strong.

The patent system then in effect required a petition with a short description, separate detailed descriptions and drawings, and a model if the secretary required it, which Jefferson did in this case. The secretary did not “approve” patents: he officially accepted completed filings to give notice to other inventors. A filing conferred fourteen years’ exclusive production and use subject to challenges in courts.

After the fire, the Patent Office solicited families and other sources for old records to reconstruct lost filings. The family provided Whitney’s written specifications but not drawings. The drawing shown in Figure A.1 is a copy of drawings, certified as accurate in 1804, which were found in the records of one of the cotton-gin patent cases. The Patent Office later made a model of the Whitney gin and apparently made drawings from that, which are now bundled with the original specifications. A 1960 article by a senior patent official confirms that the patent office commissioned a model of the Whitney gin in 1845, and that the drawings now on file “are merely drawings of the cotton gin made in 1845” and include “the addition of some figures . . . showing alternative forms of teeth.” But it adds, “Whitney is known to have made a model showing three different forms of teeth.” The original description, on the other hand, nowhere mentions other forms of teeth. Instead of clearing up the record, therefore, the Patent Office may have merely made it conform to the legend.

Miller wrote Whitney in 1797 that “We”—presumably referring to the gin business—“have been literally bankrupts for one year past—and I have been reduced to the cruel and mortifying necessity of appropriating the Property of the Estate to prevent this Bankruptcy from becoming public—In consequence of which a Plantation belonging to the Estate of Gen. Greene of Carolina will this spring be sold for half its value.” Whitney in the meanwhile was complaining to Stebbins that Miller was skimming too much money, and there is a later very bitter letter from Miller to Whitney.