What is the state, and how does it exercise power? These seemingly abstract questions become much more concrete inside a 21st-century federal immigration courtroom. First-time observers of these proceedings tend to notice two features. For one, immigration hearings are overwhelmingly bureaucratic affairs. Judges and lawyers speak in an impenetrable legalese of I-589 forms, A-numbers, filing deadlines, and master calendars. As defendants wait for their names to be called, the proceedings look more like a trip to the Department of Motor Vehicles than a television courtroom drama. But unlike renewing a driver’s license, the state’s capacity for violence looms large in federal immigration court. Armed guards watch over shackled defendants dressed in prison jumpsuits. Judges decide whether and how the government will punish them through incarceration, fines, and deportation. This combination of mundane paperwork and muscular force captures two foundational characteristics of the state: bureaucracy and coercion.1
The early 20th-century sociologist Max Weber championed bureaucracy as one of the defining organizational models of modern social structures (and modern states in particular). He described the “ideal type” of bureaucracy as having certain features: a specialized and hierarchical organization that follows a set of internal rules and that employs a full-time, professionalized workforce that relies on written documents. Although bureaucracies exist across the private sector, they are most often associated with government institutions. Weber similarly emphasized coercion in his discussions of the state, which he defined as “a human community that (successfully) claims the monopoly on the legitimate use of violence within a given territory.” Can the state wage war or put down insurrections? Can a government make people follow laws? Will they be punished if they don’t? But this framework of bureaucracy and coercion is misleading. Government institutions aren’t always organized as bureaucracies, and they don’t always exercise power through coercion.2
A Weberian lens of bureaucracy and coercion produces a history of the American state divided into two acts. In act 1, a fear of monarchy led the founders to limit the authority of the national government in favor of state governments. By Weber’s standards, the 19th-century federal government lacked both bureaucratic capacity and coercive power. It was, in a word, weak. In act 2, the 20th-century American state transformed into a powerful modern leviathan. Over the course of the Progressive Era, the New Deal, two world wars, the Cold War, and the Great Society, the American state built up new civilian agencies (bureaucracy) alongside the world’s most powerful military force (coercion). Remove this Weberian lens, however, and the story changes. Rather than a sequential story about a weak state becoming a strong one, historians have uncovered a powerful federal apparatus stretching all the way back to the nation’s founding.3
In the early years of the American republic, the federal government waged war, mapped territory, built infrastructure, collected tariffs, carried mail, provided disaster relief, and propped up the institution of chattel slavery.4 The Civil War further expanded the state’s capacity, as the federal government mobilized a vast military and civilian machinery to put down the Confederate rebellion. In the wake of that war it extended citizenship to millions of formerly enslaved people while launching new campaigns of conquest against Native groups in the West. By the end of the century, the federal government had seized, surveyed, and distributed vast tracts of western land and resources and begun to acquire an overseas empire.5 Much of this was accomplished, paradoxically, by keeping federal authority “out of sight.” Rather than establishing large civil-service bureaucracies or wielding nakedly coercive power, the federal government often delegated tasks to surrogates and intermediaries through a system of fees and bounties, including state and municipal governments, private corporations, entrepreneurs, voluntary associations, and religious organizations.6
Despite a surge of scholarship on the 19th-century American state, the topic is often discussed in abstract ways when it comes to geography. Otherwise sophisticated treatments of the state tend to use a spatial vocabulary—the “scales” of national, state, and local authority, the “sites” of governance, or the “boundaries” separating civil society, the market, and the state—without much spatial specificity. Little is known about the distribution of the state’s infrastructure and workforce or when and how that geography changed over time. In concrete terms, where was the state? It is easy to take this question for granted. After all, states are inherently spatial things, with borders that appear on maps. In short, they have territory. This is what makes a state a state. But just because a state lays claim to a particular area of the globe does not mean that it actually controls that space evenly. Understanding where, when, and how a state extends authority over claimed territory is a crucial part of its history. And nowhere was this more true than in the western United States.7
Over the second half of the 19th century the American state incorporated the western United States into the nation. The speed and success of this project can lend it an air of inevitability, as the state takes on the appearance of a “weaponized filing cabinet” rolling across the region, violently displacing Native inhabitants in order to pave the way for farms and towns, mines and dams, railroad lines and telegraph wires. But there was nothing inevitable about the speed, success, or scope of western expansion.8 At mid-century, the federal government’s territorial claims may have stretched all the way to the Pacific Ocean, but its authority over this land was virtually nonexistent. It had little to no presence in the region outside of a handful of isolated army forts and some small footholds on the Pacific coast, and it lacked even basic knowledge about on-the-ground conditions in the region. In many places Native sovereignty trumped American law and governance. The state had to extend its authority across a vast and sparsely settled environment unlike anything it had faced in the eastern United States. It had to conduct long-distance military campaigns against powerful Indigenous groups while repeatedly facing down local challenges to federal authority across the region. It had to map land, catalog resources, and allocate them to settlers and private companies. It had to carve out federal territories, admit new states, and build an administrative apparatus to oversee the region. The American state’s efforts to incorporate western territory faced daunting challenges of geography and distance. This would be a fundamentally spatial project, and the US Post would provide the underlying circuitry to complete it.9
“I had forgotten where I had left my overcoat,” Clark Thompson wrote to the post trader at the Yankton Indian Agency in January of 1864.10 While the fate of the overcoat has been lost to history, the fate of the post trader has not. The man in question, Andrew Jackson Faulk, was a successful lawyer and businessman in Pennsylvania before he moved to Dakota Territory in 1861 to take up the job of post trader at the newly created Yankton Indian Reservation. Faulk likely secured this position through his political connections within the Republican Party, which had recently taken control of the presidency. Faulk used those same connections in 1866 to catapult himself into a far more prestigious position: governor of Dakota Territory. From start to finish, Faulk’s tenure in office was completely dependent on the mail.11
Andrew Faulk’s appointment placed him quite literally on the periphery of the American state, far removed from the federal government’s seat of power in Washington, DC. Unfortunately for Faulk, 19th-century governance was a never-ending exercise in long-distance communication. Legislators had to field grievances or requests from their constituents. Circuit judges needed to be able to disseminate decisions to courts across the country. Customs officials needed to collect tariffs and deposit them in the public treasury. The vast majority of these interactions unfolded at a distance rather than through face-to-face interactions. And as Thompson and Faulk could attest, in the United States those distances were quite large.
Even at the nation’s founding the Constitution’s framers worried whether a republican model of government could survive in a country as large as the United States. As historian Richard R. John has shown, the US Post held together the United States’ fledgling political system. It carried newspapers and periodicals that kept citizens informed about politics and government. It provided common links to a dispersed population, cultivating a shared national sense of identity. And, crucially, it fostered connections between constituencies and their elected representatives. When officials convened in Washington, most of them were far removed from their constituents. Unable to regularly travel back to their districts, they instead relied on the so-called franking privilege. First instituted under the 1792 Post Office Act, the “franking privilege” allowed certain government officials to send and receive mail without paying any postage. Congressmen used this franking privilege with abandon, mailing not only letters home to their constituents but also pamphlets, copies of their speeches, printed government documents, and newspapers. This service gradually extended across the federal government, to postmasters, officers in the executive branch, soldiers during wartime, and even the widows of former presidents. These were the early long-distance links of the nation’s representative democracy.12
As the United States seized more and more territory during the 19th century, the challenges of distance grew ever more urgent. In the western United States, the US Post was a prerequisite for even rudimentary functions of government. In fact, the first post office in what would later become Dakota Territory preceded its organization as a federal territory by a full decade. By the time the first territorial assembly met in Yankton in 1862, there were already 11 post offices up and running in Dakota. As elected officials started writing civil and criminal laws, drawing up a tax code, and establishing county boundaries, they used these nascent postal connections to collect and disseminate information. By comparison, the first telegraph line wouldn’t reach Yankton until 1870, which meant that Dakota’s early territorial politics and statecraft relied on letters rather than telegrams.13
Andrew Faulk’s correspondence illustrates the extent to which the American state depended on the underlying federal infrastructure of the US Post. Appropriately enough, his tenure kicked off with a letter, an official notification from the Department of the Interior instructing him to fill out the required forms and forward them to Washington, DC.14 Almost immediately, the new governor started using the mail to fulfill his new responsibilities. Government agencies like the Treasury Department’s Bureau of Statistics sent him inquiries about Dakota and its population, with similar requests pouring in from private citizens who were considering moving there.15 Faulk sent official letters appointing territorial officials, from a local probate judge to the paymaster of the Dakota militia.16 He also corresponded with a US Treasury agent named Enos Stutsman who was stationed at Pembina on the far side of Dakota Territory along the northern border with Canada. Stutsman’s letters to Faulk had to travel some seven hundred miles, detouring into neighboring Minnesota before winding their way back to Yankton. But this connection nevertheless allowed Stutsman to submit official election returns from a distant corner of Dakota to its seat of government on the opposite side of the territory.
Letters were part of a larger constellation of long-distance correspondence between government officials. All sorts of paperwork flowed across the US Post, including certificates, registries, forms, contracts, receipts, and commissions.17 Take the case of Andrew Faulk’s own appointment as governor. To make the position official, he had to fill out a bond and oath of office (including a “loyalty oath” to guard against ex-Confederates taking office) and then send those forms to his superiors in Washington, DC.18 These pieces of paper made up the forgotten cogs of governance. Handwritten letters might make for more interesting reading material than a printed government form, but the latter is what kept the larger machinery of the state running. Without the postal network carrying those scraps of paper between distant places and seats of government, they would be just that: scraps of paper.
Faulk’s appointment as governor came with the additional title of superintendent of Indian affairs for Dakota Territory, which made him the federal government’s local proxy for dealing with the thousands of Native people who lived inside Dakota’s borders. This put him at the frontlines of the state’s larger project of Greater Reconstruction. Articulated by historian Elliott West, the era of Greater Reconstruction encompassed the 1840s through the 1870s, when the federal government consolidated its authority over both the southern and western United States, putting down the Confederate rebellion in the South while also waging war against Native groups in the West. Questions of race and citizenship loomed large as the nation debated whether and how to incorporate African American freed people, American Indians, former Mexican citizens, religious minorities, and new immigrants into the national polity.19 Faulk used the mail to conduct Greater Reconstruction’s on-the-ground operations. He sent letters coordinating peace envoys with the US Army and arranging for gifts to be sent from Yankton up the Missouri River to Dakota’s northern tribes.20 He used the mail to request an increase in federal troops to monitor these northern groups, appoint government agents to work on specific reservations, and ask for extra supplies for these reservations.21 Indian agents in turn mailed him on-the-ground updates, including official letters that he forwarded on to the Commissioner of Indian Affairs to include in his annual report.22
Even in-person governance depended on long-distance communication. In early 1867, Andrew Faulk led a diplomatic envoy of 30 Yankton and Santee tribal leaders to Washington, DC, part of an effort to improve living conditions on Dakota’s government reservations. The trip included a face-to-face audience with President Andrew Johnson, in which Johnson referred to himself as their “great father” and pledged “care and protection of [his] wards.”23 The language of the meeting reflected the federal government’s pivot away from negotiating with western tribes as sovereign nations and toward a policy of removing them onto reservations where they could be assimilated into white “civilized” culture. Notably, the entire visit was coordinated through the mail. In the year leading up to the trip, Faulk and other government officials exchanged a series of letters hammering out the trip’s details, down to what the delegates should wear on their visit, deciding between “full suits of [white] Citizen’s clothing” or “Indian costumes” (they opted for both).24 Faulk then added the cost of these clothes to a running tally of trip expenditures that included hotel accommodations, railroad tickets, and meals. He would later mail this handwritten expense report to his higher-ups in the Office of Indian Affairs for reimbursement.25
Andrew Faulk’s dual position as both governor and superintendent of Indian affairs meant that he repeatedly played the role of peacemaker between his constituents and the US Army. 26 Many of Dakota’s white residents wanted the federal government to annex the Black Hills, an area in western Dakota that the Lakota held as a sacred site called He Sapa.27 The US Army, meanwhile, was battling the Lakota and their Northern Cheyenne and Northern Arapaho allies. Facing a string of defeats, a war-weary public, and spiraling costs, army officials increasingly looked toward peaceful negotiations. When word reached them that a group of private citizens were planning to march into the Black Hills, a general wrote to Faulk to warn him that “the military authorities will not protect them in an invasion of land still held unceded by the Indian tribes.”28 Although Faulk himself was an advocate for opening up the Black Hills to white settlement, he duly printed the general’s letter in Yankton’s Union and Dakotaian newspaper.29 After Yankton’s citizens disregarded the notice and continued to ready themselves for an expedition, a fed-up army officer finally warned Faulk in a letter that any such action would be “in violation of law” and instructed him to prevent it “using force if necessary.”30
The correspondence between army officers and Andrew Faulk over federal Indian policy highlights the ways in which the American state’s coercive power flowed through less obvious channels. The US Army supplied the violent force behind the state’s project of Greater Reconstruction. But exercising this power required other kinds of administrative capacity, including the ability to coordinate actions from a distance. Again and again, the US Post provided the necessary infrastructure to do so. It shuttled communications up and down the military’s chain of command, including directives about troop movements, accounts of skirmishes and battles, muster rolls, casualty lists, requests for more supplies or troops, and official reports. Soldiers themselves relied on the mail to bring them letters from friends and family, national newspapers and magazines, and even (depending on their home state) absentee ballots to vote in elections.
The army’s reliance on the mail itself depended on the US Post’s ability to extend its network into isolated places. During the mid-1860s, for example, the army built a string of forts along the Missouri River, deep in Dakota Territory’s interior and hundreds of miles away from the nearest American settlements. This military infrastructure provided the state’s coercive power. But the state’s civilian infrastructure extended alongside it: by 1867, six of Dakota’s forts also had post offices. This included the Fort Randall Post Office, which processed more mail that year than any other post office in Dakota Territory. The sheer quantity of correspondence and paperwork that passed through Dakota’s military outposts underscores how the American state’s violent campaigns were conducted with envelopes as well as rifles.31
In the end, Andrew Faulk’s tenure in office drew to a close much as it began: through the mail. In April of 1869, Faulk opened up an envelope containing yet another sheet of government paper: a letter officially notifying him that he was being replaced as governor. As soon as his successor arrived in Yankton, Faulk was “to turn over all the records, books, correspondence, and general property of the Executive office.” His removal likely stemmed from the changeover in administrations that had come with the inauguration of President Ulysses S. Grant the previous month. Although little comfort to Faulk himself, the letter was a testament to the postal connections that knit together the larger apparatus of the American state. From the headquarters of an executive department in Washington, the letter had traveled some twelve hundred miles to the desk of a western territorial official. Its journey had taken just one week.32
Eight hundred miles away from Yankton, government officials in Colorado Territory were wrestling with similar issues as Andrew Faulk. As in Dakota Territory, Native people still occupied much of Colorado when it became a federal territory in 1861. In particular, groups of Utes including the Mouache, Caputa, Parianuche, Sabuagan, Tabeguache, Weeminuche, and Yampa controlled large areas of western Colorado. Over the following years, officials attempted to carve out more and more land for white settlement. In 1868, a collection of Ute leaders signed an agreement with federal officials that shrank their land claims while establishing the western third of Colorado as Ute territory. As part of the agreement, non-Natives were forbidden “to pass over, settle upon, or reside in” this area. Americans promptly ignored this prohibition, trespassing on Ute land in search of gold and silver in the San Juan Mountains. In 1873, federal officials renegotiated with several Ute leaders and reached the so-called Brunot Agreement, which ceded an additional chunk of mineral-rich Ute land to the federal government. Congress officially ratified the agreement in 1874, opening up an additional 3.7 million acres to white settlers.33
Figure 1.1 Occupation of the San Juan Mountains
The seizure of Ute land (light gray) in southwestern Colorado was immediately followed by the rapid spread of US Post Offices between 1873 and 1878. Terrain imagery comes from Stamen Maps (http://maps.stamen.com/terrain/), downloaded September 27, 2019.
The seizure of Ute land in southwestern Colorado was part of a larger process of settler colonialism, in which a colonizing society attempts to exterminate or remove Indigenous peoples and replace them with its own population. This was a project steeped in violence and explicitly backed by the coercive power of the American state.34 However, while violence and coercion explain the seizure of western land, they do not explain the speed with which Americans then occupied this plundered territory. As historian Virginia Scharff emphasizes, colonization in the West wasn’t just about removing Native peoples and taking their land; it also required Americans to move to places that “were unpromising, to say the least.”35 The mountainous corner of southwestern Colorado was one of those places. The area may have been promising in terms of its mineral wealth, but it was also one of the least accessible places in the contiguous United States. One newspaper editor in the summer of 1874 called it the worst “of all the rough and rugged countries that it has been my fortune to travel over.”36 Earlier that year, a prospector named Alferd Packer had resorted to cannibalism when his winter expedition became lost and snowbound in the mountains.37 The American state had officially laid claim to this tract of Native land, but that did not in and of itself guarantee that Americans would occupy it.
But occupy it they did. Thousands of prospectors and settlers poured into southwestern Colorado in the wake of the Brunot Agreement, bringing the US Post with them. As shown in Figure 1.1, the first post office in the area opened shortly after Congress ratified the agreement in 1874, and within three years there were nearly 20 post offices up and running. By 1877, horseback riders and wagons were carrying the mail to small and otherwise remote mining camps like Mineral City, Tellurium, and Animas Forks.38 These new lines of communication carried news about mining strikes to newspapers in eastern Colorado, helping further stoke the “San Juan Fever” that brought so many people swarming into southwestern Colorado during the 1870s. The US Post did not cause settlers to occupy this area; the lure of gold and silver would have brought them with or without the mail. But when they reached this otherwise distant and inhospitable place, an accessible postal service helped make it a little more livable. The US Post’s ability to rapidly expand its gossamer network facilitated a larger process of colonization.39
The impact of the US Post comes into clearer focus when situated within the wider landscape of the American state. By 1877, the project of Greater Reconstruction had reached a kind of culmination in the West. The government’s large-scale military mobilizations against western tribes were winding down, and the state was building up its civilian infrastructure in order to administer all of this conquered territory. Three institutions in particular—the US Army, the Office of Indian Affairs, and the General Land Office—were tasked with some of the state’s central functions in the region: conquering and dispossessing Native peoples, confining them on reservations, and parceling out their land to white settlers. So where were these three main government actors located? How were their offices and personnel distributed? Several sources list the locations of government offices, forts, and outposts that were operated by the US Army, the Office of Indian Affairs, and the General Land Office in 1877. Together, they start to answer the seemingly basic question: Where was the state?40
Figure 1.2 The American State and Greater Reconstruction
These maps compare the locations of personnel working for the US Army, Office of Indian Affairs, and General Land Office in 1877. Data for the Office of Indian Affairs and General Land Office were collected from The Official Register of the United States, 1877 (Washington, DC: Government Printing Office, 1877). Many thanks to Benjamin Brands for providing the locations of US Army forts.
The maps in Figure 1.2 have their limitations. They only capture the places where people were stationed according to official government records, but people didn’t always stay put: soldiers on patrol might range hundreds of miles from a fort, or a government surveyor might lead a mapping expedition far away from his office. Relatedly, the maps do not show other government agents who “resided” in Washington, DC, but may have nonetheless traveled or worked in the western United States. In short, they imply a degree of stasis that did not exist on the ground. Nevertheless, they are a starting point for excavating the spatial skeleton of Greater Reconstruction in the West. In some ways, the maps confirm suspicions. The three main government institutions of the US Army, the Office of Indian Affairs, and the General Land Office operated across a largely rural geography: think of General George Custer’s cavalry galloping across the plains, a schoolhouse on a windswept Indian reservation, or the rectangular grid of a homestead survey. Despite the scholarly attention that has been paid to these three institutions, however, they weren’t the only state actors in the West. The Official Register of the United States, a biennial directory of federal workers, reveals a much longer cast of characters.41
To take one example, the Treasury Department employed more people in the West in 1877 than the Department of the Interior, a classically “western” institution that housed both the Office of Indian Affairs and the General Land Office. Moreover, Figure 1.3 reveals that the Treasury Department operated a mainly urban and coastal geography rather than the better-known frontier geography of the rural western interior. This leads to a more complete picture of where the state was, who constituted it, and what those people did. “The state” wasn’t just made up of people like James Craven, a government agent working at the Cheyenne River Indian reservation in Dakota Territory. It was also Maggie Cook weighing coins at the US Mint in Carson City, Nevada, and Julia Williams running the Santa Barbara lighthouse off the coast of California.42 Government workers pursued a much wider range of state projects than just military campaigns, land surveys, or Indian reservations. Mapping the material presence of the federal government illustrates a more expansive state and its operations in the West.43
Figure 1.3 “Eastern” versus “Western” Institutions
Despite its association with the eastern United States, the Treasury Department (left) employed more people in the western United States than the Department of the Interior (right), with its workforce concentrated largely in urban and coastal locations. Data collected from The Official Register of the United States, 1877 (Washington, DC: Government Printing Office, 1877).
Finally, surveying the full landscape of the federal government in the West shows that no government institution had a more expansive spatial footprint than the US Post. In 1877, the Post Office Department operated more than five thousand offices in the West, or 10 times as many locations as all other government organizations (civilian and military) combined. Compared to army forts, Indian agencies, customs houses, land offices, courthouses, and lighthouses, the federal government’s thousands of post offices blanketed the region. Small-town post offices popped up practically anywhere a group of people coalesced: in mining and timber camps, in farming and ranching communities, at stage stations, railway stops, health resorts, and mill towns. No other government entity was capable of moving as quickly and operating in as many different places. Its expansive network extended the reach of the state into thousands of “in-between” places in which the federal government had no other institutional presence. The US Post was the lowest common spatial denominator of the state; this was the source of its power and the starting point from which all else flowed.44
Figure 1.4 The Most Expansive Arm of the State
The spatial distribution of post offices (left) dwarfed that of all other federal institutions (right) in the western United States 1877. All other government locations include personnel for the Department of the Interior, the Federal Judiciary, the Treasury Department, the War Department, and the US Army. Data for civilian institutions was collected from The Official Register of the United States, 1877 (Washington, DC: Government Printing Office, 1877). Benjamin Brands provided the locations of US Army forts.
How did the US Post operate such a spatially expansive network? Studying the inner machinery of the 19th century’s largest public institution leads to a deeper understanding of how states are organized and the broader relationship between territory, geography, and state power. Once again, southwestern Colorado provides a useful example. The post offices that sprouted up in the 1870s on recently seized Ute land were not the kind of freestanding post office buildings that operate today. Not a single one of these offices was run by a full-time, salaried government employee. Instead, most Colorado postmasters worked part-time for small commissions and fees, typically housing their post offices in private residences or businesses. In 1877, the post office at Eureka, Colorado, was run by a hotel and restaurant owner named Thomas G. Andrews. Four and a half miles up the road, a mine and mill owner named Edwin M. Brown was in charge of the Animas Forks Post Office. Neither Andrews nor Brown was paid very much ($86.83 and $76.50, respectively), and neither of them held onto the position for very long. Two year later, both of them had lost their appointments.45 The two may seem like unlikely candidates to illuminate the American state and its history. After all, they were not bureaucrats. They didn’t have any particular technical expertise when it came to handling the mail, didn’t work in a government building, didn’t draw a regular salary, and weren’t dedicated to a career in the public sector. They were just two local businessmen who occasionally sorted and distributed mail for their neighbors. These localized, semi-privatized, and temporary positions are difficult to square with the typical impression of a government job. Andrews and Brown were instead part of a less familiar structure of governance: the agency model.
The agency model was one of the principal organizational frameworks of the 19th-century United States. In basic terms, it was a solution to the challenges of distance and geography. An organization would delegate tasks to local representatives who were responsible for a surrounding geographical area. This allowed the organization to provide services to a dispersed populace scattered across thousands of different places. In the private sector, sales agents received commissions for peddling books, newspapers, and magazine subscriptions on behalf of companies like Scribner and Company. Life insurance companies like New York Life enlisted local agents to sell policies and vet the long-term health prospects of applicants. The credit rating company R.G. Dun, & Co. relied on local agents to file on-the-ground reports. So too did the Pinkerton National Detective Agency.46
The agency model was the engine behind the nation’s rural mail service. Roughly 95 percent of all post offices operated according to this model. Small-town postmasters who ran these offices had more in common with sales agents than civil servants, offering services to their “clients” in exchange for small payments. The bulk of these payments came through a commission system that was calculated as a sliding-scale percentage of the outgoing mail from their office. Postmasters also received fixed fees for extra services, such as processing postal money orders or renting out letterboxes. Once a postmaster’s annual combined compensation of commissions and fees reached one thousand dollars, their post office was reclassified from a “fourth-class” office to a “Presidential” office. This new designation meant that the postmaster’s appointment had to be approved by the Senate and that they began receiving a fixed annual salary from the government.47 The vast majority of postmasters, however, never came anywhere close to that level: at the end of the 19th century, roughly half received less than one hundred dollars a year.48
Rural postmasters worked with a second kind of local agent: contractors who carried mail along thousands of so-called star routes, or any form of transportation that was not a railroad or steamship. The Post Office Department awarded temporary contracts to companies or individuals to transport the mail along these star routes, most often via stagecoach or on horseback. For the majority of Americans who didn’t live directly on a railway or major river, this was how they got their mail.49 Like rural postmasters, the public responsibility of transporting the mail was largely a part-time task grafted onto an existing private business. Transportation companies that won a mail contract often simply added a few extra mail bags alongside their passengers, freight, or other cargo. Like postmaster appointments, postal contracts were impermanent: they lasted for a maximum of four years (many were shorter), and postal officials could alter the routes at any time.
The US Post used the agency model to rapidly expand into areas like Colorado’s San Juan Mountains during the 1870s. The federal government didn’t have to construct buildings, buy vehicles, or dispatch employees. It just appointed a local resident to work part-time as a postmaster and contracted with a staging company to carry the mail. The government’s equipment costs were minimal. New postmasters received postage stamps, stamped envelopes, postal cards, and a book of rules and regulations from department headquarters. Small-town postmasters were responsible for supplying everything else, including letterboxes and keys, pens and pencils, rubber stamps, scales, money drawers, and bookkeeping forms—and of course the physical space of the post office itself. Mail contractors, meanwhile, were similarly expected to furnish nearly all of their own equipment, including vehicles, draft animals, barns, and waystations. The only physical equipment provided by the Post Office Department consisted of official sacks, pouches, and canvas bags to carry the mail.50
The agency model has been overshadowed by a bureaucratic model of the state. But in the 19th century, large swathes of the US federal government did not function at all like the idealized bureaucracy championed by Max Weber. The US Post’s rural mail service highlights this neglected machinery of governance in action and provides a lens for recognizing the agency model in other contexts. Postal officials were not the only ones facing the challenge of American geography. Again and again, government institutions turned to the agency model to provide public services in far-off places. Returning to southwestern Colorado in 1877, two other federal institutions in addition to the US Post were operating on recently seized Ute land: the General Land Office and the federal judiciary. Both of them relied on the agency model. The surveyor general of the United States created the San Juan Land District and located its headquarters in the booming mining town of Lake City in early 1877. As with all of its field offices, the Lake City land office was staffed by two people, a receiver and a register, who were responsible for processing land applications. Although they received a modest base salary from the government, most of their public income came from a 1 percent commission on each land sale. Neither of the workers were full-time civil servants; one was a storeowner and assayer, the other a newspaper editor. They are a reminder that despite the trained, technocratic expertise of the General Land Office’s surveyors and cartographers, the task of actually transferring public land into private hands was conducted by a scattered workforce of local agents doing minor clerical work on behalf of the federal government.51
The federal judiciary had also staked out a presence in southwestern Colorado by 1877, and much like the Post Office Department and the General Land Office, it had done so using the agency model. The federal court system relied on US commissioners, or local agents who performed minor judicial functions on its behalf. The position of commissioner was first established in 1793, when Congress passed a piece of minor legislation allowing a court to authorize a local person “learned in the law” to take bail in remote parts of a judicial district.52 The responsibilities of commissioners gradually expanded to include tasks such as taking affidavits, administering oaths, or issuing search warrants. By 1877, more than 1,800 commissioners were scattered across the United States—including Colorado’s San Juan Mountains, where John L. Pennington acted as both the private supervisor of a local mine and as a publicly appointed US commissioner in the town of Silverton. Like postmasters, commissioners did not receive a salary, instead receiving fees for part-time work. They also worked with little administrative oversight; anybody could be appointed as US commissioner (there were no requirements that they had received legal training), and they did not have to report the fees they earned for their work. And, like postmasters, the position of commissioner was also a contingent one. John Pennington, for instance, had left or lost his job as US commissioner in Silverton within two years of taking the position.53
Mapping the federal judiciary underscores the agency model’s role within the larger apparatus of the American state. As shown in Figure 1.5, the federal judiciary had two distinct workforces with two distinct spatial patterns. The first consisted of circuit judges, district attorneys, court clerks, and bankruptcy registers. This workforce tended to function more like a bureaucracy, with salaried, professionalized employees who worked in official government buildings located in major population centers like Denver and Pueblo. The second workforce consisted of US commissioners like John Pennington. These part-time local agents outnumbered all other federal judicial employees combined and worked in many more locations, including smaller towns and cities like Silverton, Colorado. The difference was especially clear in California, where 67 US commissioners were spread over 54 different locations—compared to eleven other judicial employees stationed in just two cities, San Francisco and Los Angeles. Of course, size and geography did not directly translate into influence and power. A federal circuit judge in San Francisco wielded far more power than all of California’s US commissioners combined. But part-time commissioners allowed the judicial branch of the federal government to function over a much wider area than it would have otherwise. Administrative structure and institutional geography went hand in hand, as the agency model’s flexible administrative arrangements produced a much more expansive spatial footprint than more bureaucratic organizational corners of the American state.54
Figure 1.5 The Federal Judiciary’s Two Spatial Footprints
Locations of US Commissioners (left) versus all other workers in the federal judiciary (right) in 1877, sized according to the number of people working in that location. Data collected from The Official Register of the United States, 1877 (Washington, DC: Government Printing Office, 1877).
Some of the 19th century’s most important government institutions used the agency model to function over long distances and large areas. The Pension Bureau, for instance, faced many of the same geographical challenges as other federal institutions. This organization was tasked with distributing payments for injured veterans along with the widows and dependents of soldiers who had died in service, an undertaking that became vastly more complex after the mass casualties of the Civil War. In order to establish who did and did not qualify for a pension, it first had to conduct in-person medical exams of hundreds of thousands of veterans. Rather than hire a full-time workforce of trained government doctors and dispatch them out across the country, the Pension Bureau grafted the medical examination process onto the existing private practices of local physicians. These “neighborhood practitioners” were tasked with evaluating pension claimants in the surrounding area. Instead of a full-time salary, they received a small fee for each exam. The Pension Bureau’s physicians were not unlike the Post Office Department’s postmasters or the federal judiciary’s commissioners: private local agents responsible for providing services on behalf of the federal government within a given area.55
The US Census also used the agency model to fulfill its responsibilities. Each decade the federal government had to collect information about millions of households and individuals, in an attempt to make its populace legible to central planners, officials, and other government actors. This was no easy task in the United States, especially as Americans occupied more and more territory during the 19th century. From 1790 through 1870, US marshals were in charge of collecting information within their particular judicial district. Each marshal appointed “assistants” to act as enumerators—private residents who collected information about households in a given area. In 1879, Congress transferred responsibilities from the US marshals to an official Census Office. But the agency model didn’t disappear with more specialized administrative oversight; if anything, it grew more entrenched. In June of 1880, 31,382 enumerators fanned out across their districts, going door to door to collect information about the nation’s roughly fifty million inhabitants. This group looked a lot like postmasters: local agents responsible for a particular territory who were paid through commissions for part-time labor. Some postmasters, in fact, moonlighted as census enumerators. Moreover, their commissions were calculated based on geography, as sparsely settled parts of the country garnered a higher commission than urban districts because enumerators had to travel longer distances to count the same number of people. Despite advances in tabulating technology, statistical methods, and technocratic specialization within the Census Office, the success of its operations ultimately depended on an army of temporary, semi-privatized local agents acting on its behalf.56
The presence of the agency model did not preclude bureaucracy. “The state,” after all, is not a singular entity. Even within the same institution, elements of the localized agency model often worked in conjunction with more centralized, bureaucratic arrangements. In the case of the US Post, the Railway Mail Service was an early beachhead of bureaucracy within the federal government, using salaried, full-time employees to transport and process mail along railway routes. These employees were subject to tightly managed, top-down oversight and were some of the first postal workers to take civil-service exams.57 But once the mail was unloaded from the nation’s railway cars, this bureaucracy bled into the agency model of the countryside. Railway mail agents would hand off mail bags to a stagecoach driver who would then deliver them to the doorstep of a local storeowner. The agency model could and did work in tandem with bureaucratic models of governance.58
The western operations of the US Post challenges longstanding frameworks of state power. Rather than acting through coercion and force, as emphasized by Max Weber, its influence came from structural power, or the ability to define the conditions under which people take actions and make decisions.59 Although the US Post’s structural power is less visible than the American state’s more muscular capacity for war and policing, it is just as crucial for understanding the federal government’s efforts in the western United States. This structural power was on full display in Andrew Faulk’s office in Yankton and the mining towns of the San Juan Mountains. There, the gossamer network’s distinctive geography helped define the conditions under which government officials worked and settlers migrated. Its rapid extension and expansive coverage made it the underlying circuitry for the state’s larger project of territorial conquest and occupation.60
The diffuse administrative structure of the agency model challenges a second assumption about state power: that the state is an inherently centralized entity. Under this framework, power flows from the top down and the center outward. Heads of state declare war, legislators pass laws, judges issue legal decisions, and administrators implement policies, and all of this gets transmitted through the rest of the government and across civil society. Some of the most influential theoretical models of the state define it in these terms. These models of the state often conflate organizational centralization with spatial centralization, assuming that leaders take actions from seats of governance and then implement them outward over a state’s physical territory, like an earthquake’s aftershocks reverberating from its epicenter.61 Sociologist Michael Mann, for instance, defines the state as “a territorially centralized form of organization” that wields both “despotic” (or coercive) power and “infrastructural” power, or the ability to penetrate civil society and implement decisions throughout a territory.62 Under this kind of framework, the state is powerful if the center can exercise control over the periphery, and a state is weak if the periphery can circumvent this control. Regardless of the outcome, the action largely flows in one direction.63
This book reverses the traditional direction of state power. The western postal system was dominated by local demands, local conditions, local politics, and local actors. By and large, officials in Washington did not decide when or where to expand the nation’s rural postal infrastructure. Instead, local business owners and civic leaders would write petitions, gather signatures, and send their requests for new mail service to Washington, DC. Often these requests went directly to Congress, which had authority over designating new mail routes. Loath to deny their constituents’ requests, Congress would duly rubber-stamp hundreds or thousands of new mail routes each year and send them on to the Post Office Department. Postal officials were then tasked with evaluating whether and how to establish service on the requested route. The department had no official guidelines for approving or denying these requests. There was, for instance, no requirement that a new post office serve a certain number of people. As postal officials readily admitted, requests for more mail facilities were rarely rejected.64 Given the sheer size of the network, officials in Washington relied on local politicians and intermediaries to advise them on where to establish new post offices or who to appoint as postmasters to run them. Once an office was up and running, it was subject to minimal administrative oversight: in the words of one postmaster general when describing the nation’s rural post offices: “The touch of the Department upon them is very slight . . . The machinery is set up and then let alone.”65 The periphery, rather than the center, drove the western postal network’s growth and managed its on-the-ground operations.66
Centralized models of state power risk attributing a false sense of purpose, coherence, and intentionality to the state. After all, “the state” is not a pseudo-person who calculatingly compels people to act in a certain way, and not every state action can be chalked up to the rational decision-making of policymakers and officials.67 Take the case of the agency model. It is tempting to see the US Post’s reliance on this framework as some carefully planned tactic to extend the state’s “infrastructural power” across the periphery. In reality, postal administrators found themselves barely hanging on to the reins of a regional expansion project that careened forward ahead of them.68 The expansion of the western postal network lacked intentionality, top-down planning, and centralized coordination. But that didn’t make it any less influential. In fact, its decentralized structure had a profound impact on the western United States.69
The US Post’s embrace of the agency model directly shaped the network’s geography and, by extension, its impact on the region. For one, a lack of centralized coordination contributed to its status as the world’s most expansive postal network. Generally speaking, postal officials in other countries exercised more top-down oversight than their counterparts in the United States, which in turn led to a more cautious and limited expansion of new infrastructure. The Canadian Post, for instance, dispatched an inspector to personally visit each proposed mail route or post office in order to decide whether to grant the request. In Japan, the Meiji imperial government treated the national post as a vehicle for its modernization campaign, which subsequently made it one of the most closely regulated and centralized postal systems in the world. The Russian government exercised tight control over its post offices, requiring local communities to fund new post offices themselves and shutting down any offices that failed to turn a profit within three years.70 There were no equivalent regulations in the United States, which meant that the US Post expanded far more capaciously: by 1878, the United States had seven times as many post offices as Canada and roughly 10 times as many as Japan and Russia. The agency model was what allowed the US Post to dwarf every other postal system on earth.71
In the western United States, the decentralized administrative structure of the US Post facilitated a breakneck brand of expansion. Again and again, westerners were able to mobilize postal infrastructure to serve their needs. Although settlers depended on a range of federal institutions, from the General Land Office to the US Army, the US Post proved most responsive to their demands. Westerners often saw the federal government as slow, unresponsive, or even downright opposed to their interests, such as when officials tried to keep squatters from illegally trespassing onto Native land or reservations. By comparison, westerners themselves were often in the driver’s seat when it came to the region’s postal system. The decentralized agency model meant that no matter where they moved, they could generally count on the postal network to follow them. Historian Gary Gerstle writes, “How the United States managed expansion without a large centralized state is . . . among the most significant stories of American history.”72 In some ways, Gerstle has it backwards. A better question might be: Would a centralized state have managed this kind of expansion at all?
Had the US Post followed the more centralized models of postal administration in countries like Canada, Japan, or Russia, it would have constrained its geography and slowed its expansion into rugged, sparsely populated areas such as the San Juan Mountains. Prospectors would have trespassed onto Ute land with or without the US Post. Mining companies would have constructed mills and sluices, and settlers would have built cabins and surveyed townsites. But without reliable mail service they would have struggled to maintain social ties, conduct business, join civic organizations, or participate in national politics. Some might have given up, while others might have never decided to relocate to a distant mountain range in the first place. Without the US Post’s expansive network, the pace of settler colonization would have been slower, its reach more limited, and its prosecution more difficult.73
Instead, the US Post helped accelerate the seizure of Native territory and its transformation into a colonized landscape of settlements, mining claims, and post offices. The structural power of the western postal system to shape this process was far from the only way in which the American state exercised power. But it remains one of the least appreciated. Even today, large-scale, indirect forces like the US Post are difficult to pin down. They function in the realm of the everyday, the routine, even the mundane. Studying the sprawling geography of the US Post and placing it within the wider landscape of the federal government paints a new picture of the American state and its western operations. But fully understanding this larger system requires shrinking the scale of analysis. The US Post’s structural power comes into focus when seen through the eyes of individual people.