Revere most likely designed this setup with raw material flow in mind. He initially processed all raw materials in the foundry building, where he might cast his own copper pigs or pour bronze into bell or cannon molds. Workers then brought the copper pigs to the hammering building for reshaping, annealing, and possibly some refining before moving to the rolling mill for drawing into spikes or rolling into bolts or sheets. Similarly, workers carried solid cannon to the rolling mill building for boring and finishing.
These processes had evolved considerably since Revere’s earlier days. Forty years earlier, John Singleton Copley’s portrait placed a younger Revere amid his tools and product, illustrating quite literally how the maker’s hand is reflected in his work. As he neared the conclusion of a prosperous career, Revere’s finest legacy remained his crafted items, and we best understand the magnitude of his accomplishments by studying his many product lines, the changes he made to his production routines, and the ways each deliverable moved closer to the ideals of standardized production.
Figure 8.2. Paul Revere’s sketch of his Canton Property, from pen and ink sketch of Revere Property on Neponset River in Canton. From Revere Family Papers, 1746–1964, microfilm edition, 15 reels (Boston: Massachusetts Historical Society, 1979), reel 15. This undated and untitled sketch was almost certainly drawn by Paul Revere himself, depicting his house in the center of the image and two of his shop buildings alongside a curving river. The sketch integrates Revere’s home and work life, and juxtaposes natural elements such as the river and trees with fences and manufacturing buildings. Courtesy of the Massachusetts Historical Society.
Revere maintained his bronze bell and cannon production in the midst of sizeable copper bolt and sheet contracts. His output increased throughout his career, and his son continued both product lines for many years after his death. Revere’s bell and cannon casting initially took place in his Boston foundry under the supervision of Joseph Warren, while the larger standardized operations related to bolt, spike, and sheet production took place in Canton under Paul’s watchful eye. After the 1804 gale destroyed the Boston shop, Revere and Son moved all their operations to Canton and achieved greater efficiency for all their operations and access to waterpower for large cannon-turning equipment. Consolidation also helped Joseph Warren take a larger role in all aspects of the company.
Revere’s pricing indicates that he consistently reserved his cheapest rates for bell manufacture and his highest rates for cannon casting. For example, from 1809 to 1810 he charged 42 cents per pound for bells, 50 cents for bolts and spikes, and 55 cents for cannon. At first glance this pricing might appear inconsistent with the complexity of the bell-making process, but Revere managed to simplify and perhaps even standardize his labor in two ways. First, greater experience allowed his workers to apply consistent procedures, such as an optimized ratio of metals and tried and true pouring processes, thus removing earlier uncertainty while avoiding time consuming “do overs.” Second, he used a limited number of molds to produce a smaller range of bell sizes. After 1799 he restricted his output to bells weighing five hundred pounds or more (with two exceptions), usually used in churches as opposed to schools or ships. This restriction allowed him to focus on a more uniform product line. Since these large bells required an enormous weight of metal the amount of labor per pound was far less than for individually prepared spikes and sheets, which required annealing, hammering, and shaping. Even though bells remained his least standardized product his comfort with the process allowed him to accelerate his production by following an established process known to produce the desired acoustics for a bell of a specific weight and size. Revere made between two and twelve bells a year throughout the 1800s until his retirement.
Regarding the quality of his bells, Revere’s correspondence reveals a single dissatisfied client whose complaints greatly affronted his sense of honor. After receiving their bell on May 31, 1802, Messrs Heywood, Flagg, and Stowell reported that “the general voice of the People seems to be that they are very much disappointed in their expectations that your Bell cannot be heard at a proper distance.” Revere responded in July, briefly explaining that large bells have a more “Majestick” sound than smaller ones, but not necessarily a louder one: had the clients asked him earlier he would have told them as much. His somewhat pompous response failed to win them over, and the same clients responded with a far angrier letter, accusing Revere of a “want of honor and integrity.” Predictably, Revere responded with utter outrage.
You are the first persons who ever charged me with deceiving them or with the want of Honour, or Integrity, in business, and I despise the man who would charge me with either . . . If you are “mortified with a miserable thing” don’t blame me, after saying in your letter that “the sound is agreeable” and after hearing the Bell sounded in the Furnace forty times & approving of it, before you took it away; Then was the time for you to have disputed the sound, before you took it from the Furnace. Your Bell is the 52nd Church Bell which we have cast. We have never had occasion to write one letter to, or have had one word of dispute with either of the Committees; yours is the first.24
Throughout this exchange Revere professed his unshakeable faith in both the workmanship of the bell as well as his own integrity, and even though he wanted to make his customers happy he could not understand their dissatisfaction. His quality assurance policies do back up his position, since he let the clients listen to forty peals prior to their purchase. Revere eventually received the payment for this bell, and his other records indicate that these harsh complaints were in fact a rare occurrence. The records do not contain any other criticism of his bells, and one customer spoke for the majority when he told Revere, “The Bell proves a very good one—it gives universal satisfaction.”25
Although cannon casting might appear similar to bell casting, it involved two expensive and unpleasant processes that explained the higher price. The cannon-casting process became fairly routine once a metal recipe and mold preparation procedure survived the test of time. After that, however, Revere had to bore the cannon, which he accomplished at Canton as early as 1804. In September 1810 he told factor Joseph Carson that his works could only bore one gun at a time “but should they [more cannon] be wanted at short notice we could extend them.”26 Boring was time consuming and difficult work: any misalignment between the cannon and drill created an asymmetric chamber that decreased the cannon’s accuracy and greatly increased the chance of a deadly explosion. Following the boring, all cannon had to be proved under highly adverse conditions. Proving not only necessitated additional transportation fees but also ran the risk of bursting the cannon and starting the entire process from the beginning at Revere’s cost.
Even in 1802 Revere rightfully considered himself a cannon-casting expert. Amasa Davis of the Massachusetts militia remained the primary customer for ordnance, buying pieces on many occasions. Revere also lobbied the federal government for contracts. In 1802 he informed Attorney General Levi Lincoln that he made “very great Improvements in the Foundry line, especially [words missing] Casting Cannon. I think I can say without vanity that I cast them as well as they are cast in Europ . . . I believe I have cast more brass cannon than [words missing] people in America. I have cast more than one hundred for this & some of the other states.”27 Revere received a small number of federal orders and occasional requests from other states. In 1810 he even received a secret request from the Spanish government to furnish twenty pieces of cannon, a tremendous order at the time that was probably never fulfilled.28
Neither Revere nor anyone else would ever mistake his bells and cannon for standardized products. Even the weights of different pieces made from the same mold varied. These processes still depended too heavily upon human judgment and skill, and the demand for bells and cannon never grew high enough to justify additional research into cost cutting or process improvement. Standardization simply did not make sense for these product lines, as the existing procedures satisfied his clients while af
fording a tidy profit. Revere continued these operations without any major procedural changes, and they provided him with a steady income and an expanding network of positive referrals.
Revere started producing bolts and spikes in his Canton mill shortly after buying the Canton property. The high opinion in which Revere valued his bolts and spikes, as well as his notion of honor, is evident in his response to a rare grievance from Mess. Beck and Harvey in Philadelphia, who complained that Revere’s bolts and spikes were too brittle and out of proportion. His response speaks for itself:
we were never so much astonished at any thing of the kind as at the contents of your letter . . . That “the bolts” should be “extremely brittle” is impossible, for they are drawn hot, under the forge hammer . . . the spikes are pronounced by judges here as good as ever was made, these sent you are exactly the same . . . We have manufactured the Bolts and Spikes for more than twenty Merchant vessels and you are the first persons who have found fault with our bolts or spikes . . . We have but one rule to do our business by, the strictest rules of Honour; we have sent you Bolts & Spikes such as our Carpenters approve of here, had you sent us patterns we should litteraly have followed them. We are Gentlemen.29
As with the bell grievance cited above, Revere saw no separation between his business and his personal reputation: a technical complaint about his product struck him as a personal attack upon his honor and challenged the “gentleman” status he awarded himself. In addition, he refused to accept the validity of this criticism in light of his own substantial experience, finding it “impossible” that these bolts and spikes might be deficient. Revere’s next letter to these merchants was far less outraged, and indeed, almost humble. He discussed terms for reshipping the bolts and spikes back to Boston and clarified his defense of his bolts and spikes: “that they are as smooth as English we do not contend but that they are as good we do. Our Manufactory is yet in its infancy. I think I told you when in Phila. that we finished ours with a hammer which is the reason they are rough, the English is Rolled . . . we are improving & expect to finish our Bolts the same way.”30 Revere concluded by referring these men to Joshua Humphreys for his opinion of bolt and spike quality. Whether or not Beck and Harvey were correct about the quality of their bolts and spikes, Revere did not exaggerate his experience in this field. Throughout all his Canton operations, bolt and spike work proceeded with ever-increasing volume, eclipsing all other products in total sales. Any improvements he could make to this process would have a major effect on his balance sheet.
By 1805, Revere implemented the improvement mentioned at the end of his letter to Beck and Harvey. He either purchased or cast a special pair of rollers for the mill, for use in producing bolts rather than copper sheets. He explained this system to Levi Hollingsworth in 1808 and implied that in this instance he made his own rollers, which contained grooved indentations. If aligned properly, these rollers produced copper cylinders out of flat copper passed through them. The rollers squeezed the heated sheet and forced the copper into the rollers’ indentations, producing what Revere called “strings” of copper—long, bolt-sized rods between four and ten feet long. Revere’s reply to Hollingsworth included a cross-sectional diagram of iron rolls highlighting the diagonal indentations for bolt production.31
His workers still made other fastening products such as spikes at the forge, working them hot and finishing them cold to ensure that the points remained hard but not brittle. Revere alleged in one letter that his bolts and spikes were nearly as hard as iron and claimed his braces and pintles were the best in the state, because “we can give the metal a better heat and we cast the whole for one ship at one melting.”32
Revere produced enormous quantities of bolts and spikes throughout his operations. The applicability of the term standardized output to products of this sort is a matter of debate. “Standardized” goods often imply complicated products such as mechanisms constructed from interchangeable parts: for example, rifles or bicycles. However, simpler items can be standardized in the same way that they can be heterogeneous. Revere’s use of his rolling mill to produce copper bolts represents a major milestone in his manufactory because he shifted a previously handmade item into the realm of machine production. He made a similar though less dramatic improvement to his spike production process when he added machines that headed spikes, a process previously performed with a hammer and special tongs. As he offered products in several fixed sizes and weights, and as he offered to repurchase unused quantities, he increasingly treated them as uniform commodities. For example, a memorandum in his 1806–1812 bankbook lists the weights of 1-foot-long copper bolts of different diameters: 1.5-inch diameter bolts weigh 4 pounds, 1-inch diameter bolts weigh 3 pounds, and so on.33 This calculation clearly assumes constancy among his output. In any early industrial endeavor the potential for irregularities still existed, but Revere’s manufacturing and managerial policies indicated he wished to make them infrequent exceptions to a standardized rule.
In December 1803 Revere wrote to Joshua Humphreys to discuss his sheet-rolling process and how he learned by doing:34
I agree with you that the manufacturing of copper in this country is a new thing & that every allowance ought to be made & every encouragement given to [so] usefull a Branch for tis by Experience that the Manufacturer becomes perfect & experience will not be gained without encouragement is given. We are dayly gaining experience. I have had & now continue to have the whole to feel out, for I have not been able to get any information from any person. Should I live & be able to take care of the Business for seven years to come, I should not get to the Zenith . . . I cannot help acknowledging that I have done better than my expectations. Our sheets are as well finished and as soft & as free from scales & cannot be distinguished from English.35
His final claim was not completely accurate. Revere continued his learning process on the job, and made his fair share of mistakes. As mentioned earlier in this chapter, he informed Robert Smith of some consistent errors in his first shipment of sheet copper in May 1802. Instead of producing sheets measuring 14 inches by 4 feet, a “small part” of the shipment measured 13.25, 13.5, or 13.75 inches wide by 4 feet long. Revere also mentioned that he consulted with some shipbuilders, and was informed that many of the sheets on a ship are cut down to a narrower size anyway. In other words, he believed his error would not have any harmful impact.36
This problem with the roller alignment was only one of many mistakes he and his workers must have made in their early years. His success in this field testifies to his workshop’s ability to adapt and continue perfecting their processes. This in large part resulted from his experience with silver, iron, brass, and copper, which gave him a great instinctive awareness of metallurgical processes. He demonstrated his confident expertise in various letters to the Navy Department or other questioners, on subjects such as annealing, work hardening, and different finishing processes. Although his contract asked him to roll his sheets cold, he asked for permission to roll them hot, which would make them less brittle, allow nails to pass through them easily, and form a tighter seal with the ship hull. He knew this through practical experience and not from a study of the chemical and physical theories behind each process. Captain Edward Preble agreed with Revere’s advice, and Navy Secretary Smith asked for more information. Revere responded: “The nature of copper is such that when it is in its pure State it is nearly as soft as Lead. It cannot be wrought till it is in that state, hammering it, pressing it, or Roleing it makes it hard & stiff; then heating it red hot, or Annealing it, brings it to its natural State again.” He then mentioned that bolts and spikes were often finished while cold, to harden them enough to drive into wood, while British sheet copper was “finished as soft as the Annealing makes it.” Revere then described his own method: “I think we have made an improvement, for after it is annealed and cleaned, we pass it once thro the Roles, which finish each sheet flat, smooths it, and adds a little to their stiffness.”37 Revere spoke with absolute
authority on this matter and illustrated how he combined hot- and cold-working processes to produce both flexibility and strength. Smith adopted all his suggestions.
Just as Revere eventually treated bolts in a standardized way, he also came to view all copper sheets as interchangeable. As early as 1802, he mentioned that his sheets weighed “34 ounces to the superficial foot,” or 34 ounces per square foot. The weight of a fixed area of copper depended on its thickness since the density of the raw copper did not vary much. Revere told Robert Smith that copper sheet inspectors recommended one-fourth of his sheets should weigh 30 ounces per square foot, a thinner sheet that he could easily provide if asked. Not only was he more in touch with the needs of the actual shipbuilders than the navy secretary, but he also began to consider the utility of producing different grades of sheeting for different uses. His 1806–1812 bankbook reflected this division of sheet thicknesses in a memorandum listing the weights of different thicknesses of sheet copper.38 This method allowed Revere to verify the uniform quality of his sheets with a well-understood and easily used measuring device, his scale. He did not mention whether he also used gauges to measure the width of his sheets, but this seems highly likely considering the increasing mention of sheet width in his correspondence with clients such as Robert Fulton. In spite of Revere’s move toward standardization, his output occasionally suffered from imperfectly machined rollers. He was still a bit ahead of his time, and only when other industries perfected machining processes to create his own equipment could he truly establish standardized production.
Midnight Ride, Industrial Dawn Page 40
