At 5:30 on the afternoon of that Tuesday, Attorney General Richard Olney walked into the cabinet room of the White House carrying two documents. The first was the omnibus injunction issued the previous day by Judges Peter S. Grosscup and William A. Woods of the United States Circuit Court for the Northern District of Illinois, restraining Eugene V. Debs and the American Railway Union from continuing what had become, within the space of twenty days, the largest sympathetic boycott in the country’s history. The second was a draft order directing Major General Nelson A. Miles to move regular army units from Fort Sheridan into the city of Chicago to enforce that injunction and to protect the carriage of the United States mails. Grover Cleveland read both papers. He asked Olney a single question about the legal theory under which the army would operate. Then he signed.

Within forty-eight hours, six companies of the Fifteenth Infantry were patrolling the stockyards south of the city. Within five days, Illinois Governor John Peter Altgeld had wired the White House to denounce the deployment as an “unwarranted” intrusion on the police powers of his state. Within two weeks, Debs was under U.S. indictment. Within a year, the Supreme Court would unanimously affirm the legal architecture Olney had built, and the words “In re Debs” would enter the constitutional canon as authorization for what every subsequent industrial historian has called the most consequential anti-labor ruling of the nineteenth century. This article reconstructs how Cleveland came to that signature, who shaped the options he faced, what he chose to do and what he chose not to do, and why the precedent he set in the hot first week of July 1894 still governs industrial relations in ways the President himself could not have foreseen.

The Sleeping-Car City and the May Walkout

George Mortimer Pullman had built his company town fifteen miles south of Chicago between 1880 and 1884 on roughly four thousand acres of prairie purchased through a holding company organized expressly to conceal the land assembly. The town that bore his name housed approximately twelve thousand residents by 1893, almost all of them employees of the Pullman Palace Car Company or members of their households. Pullman owned every house, every brick of every house, every street, every gas line, every water main, every shop, the school, the library, the church, the hotel, and the theater. Residents paid rent to the company. They bought groceries at company stores. They drew water from the company waterworks. They worshipped, when they worshipped, in Greenstone Church, which the company had built and from which he charged a competitive rate for use by any willing congregation. The arrangement was advertised in the firm’s own publications as a benevolent experiment in industrial paternalism. Richard T. Ely’s 1885 essay in Harper’s Magazine, “Pullman: A Social Study,” had described it more skeptically as un-American in its concentration of authority over the working lives and domestic arrangements of its inhabitants, a phrase Pullman never forgave.

The economic shock that broke this arrangement was the Panic of 1893, which by the winter of 1893 to 1894 had produced an industrial contraction the country had not seen since the depression of 1873. Orders for new sleeping cars and palace cars collapsed. The firm responded by cutting wages by an average of 25 percent across the spring of 1894, with reductions for some categories of skilled workers reaching 30 to 40 percent. Rent in the company town, however, was not reduced. Nor were prices at the company stores. Nor were the water and gas rates. The company bookkeeping system deducted rent and utilities from wages before the worker received any cash; some workers received pay envelopes containing literally a few cents after deductions, and one was reported in the May 1894 testimony before the U.S. commission to have received an envelope containing seven cents for two weeks of work. The workers organized a grievance committee. The committee met with the company’s vice president Thomas Wickes on May 9, 1894. Wickes received them politely. The next morning, three members of the committee were fired.

That dismissal triggered the walkout. On May 11, 1894, approximately three thousand workers left their shops. The American Railway Union, founded the previous year by Eugene V. Debs after he had broken from the older railroad brotherhoods over their craft-by-craft segmentation, had organized an ARU local in March. Debs himself counseled the men against precipitate action, arguing in private letters to the local organizers that the ARU was too young an institution to bear the weight of a national showdown with a company as well-connected as Pullman’s. But the workers had walked, and Debs could neither call them back nor abandon them. The ARU convened a general convention in Chicago on June 12, 1894, at Uhlich’s Hall, with delegates representing roughly 150,000 members across the trans-Mississippi rail system.

What that convention voted on June 22 was the move that converted a local labor dispute into a national crisis. The convention resolved that, unless the company agreed to arbitration of the wage and rent grievances by June 26, ARU members would refuse to handle any train containing a Pullman sleeping car or palace car anywhere on the American rail network. Pullman refused arbitration on June 26 through his counsel, citing the firm’s long-standing position that there was nothing to arbitrate between a company and its employees concerning wage rates set by economic conditions. The boycott began at noon on June 26, 1894. Within seventy-two hours, freight and passenger traffic across the Midwest and trans-Mississippi West had ceased on lines whose ARU members refused to attach sleeping cars to outbound trains and whose management refused to run trains without them. By June 30, twenty railroad lines were partially or wholly paralyzed. By July 2, the figure was approaching thirty, with secondary effects on coal supply, meat shipment from the Chicago packing houses, perishable produce in transit from California, and the carriage of the United States mail.

The General Managers’ Association and the Mail Theory

The opposition to the boycott was already organized and waiting. The General Managers’ Association, formed in 1886 by the twenty-four rail lines centered on Chicago, had existed for almost a decade as a coordinating body for wage policy, schedule integration, and joint resistance to industrial organizing across the carriers. The Association maintained an office at the Rookery Building on La Salle Street, a permanent staff under General Manager John M. Egan, and a litigation budget. Its members had agreed in 1893 that any walkout against any one of them would be treated as a walkout against all of them, and that strikebreakers, locomotives, and rolling stock would be shared across lines as needed. By the time the ARU boycott began, the Association had been preparing for a confrontation with Debs for almost a year, since the ARU’s successful spring 1894 walkout against the Great Northern Railway had demonstrated that an industrial workers of all rail workers, organized across craft lines, could win against a single carrier. Crushing the ARU before it consolidated that model was the explicit strategic objective Egan had laid out in his March 1894 memorandum to the Association’s presidential committee.

The legal opening the Association needed came from a specific feature of how sleeping cars were attached to trains. Federal contracts required certain mail trains to carry specific consists of cars on specific schedules. If a sleeping car was part of the contracted consist, the railroad was legally obligated to carry it; and if the ARU refused to attach the sleeping cars, the railroad would either refuse to run the train (failing the mail contract) or run a non-conforming consist (also failing the mail contract). Either way, the boycott could be characterized as obstruction of the United States mails, a U.S. offense under the postal statutes of 1872 and 1888. The General Managers’ Association moved quickly to instruct member railroads to refuse to run any mail train without its Pullman car, even where the mail car and the Pullman car could have been separated operationally. This converted what could have been a maintainable local dispute into a U.S. mails obstruction case, by design.

The architect of this conversion was not the Association’s lawyers but the Attorney General himself. Richard the attorney general had come to the cabinet in March 1893 directly from a Boston law practice in which he had served for fifteen years as general counsel to several railroad clients, most notably the Boston and Maine and the Chicago, Burlington and Quincy. He had served on the Burlington’s board of directors and continued to draw a retainer from that road during his tenure as Attorney General, a fact the Burlington’s board minutes documented and which his administration never disclosed. Whether this constituted a conflict of interest in the modern sense (the U.S. conflict-of-interest statutes governing presidential branch officers did not take their modern form until the 1962 amendments) is technically arguable; whether it shaped the Attorney General’s conduct of the office is not. Olney’s June 1894 correspondence with Edwin Walker, the General Managers’ Association’s lead Chicago counsel, shows the two men coordinating legal strategy in real time. On June 30, 1894, the attorney general appointed Walker as a special assistant U.S. Attorney for the Northern District of Illinois, an appointment that placed the Association’s own lawyer simultaneously in the role of U.S. prosecutor pursuing the case against the ARU his client was trying to break.

This dual appointment was not concealed. Olney signed it in writing. The Chicago press reported it. Walker himself proceeded to file the U.S. motion that produced the July 2 injunction wearing both hats openly. The constitutional and ethical objections that subsequent scholarship has raised against the arrangement were not raised at the time with any force, because the legal culture of 1894 did not yet treat the role of special prosecutor as inherently incompatible with continuing private representation of an interested party. The lesson Olney drew from his experience was instead the opposite: that the United States prosecutorial machinery could be made to serve a coordinated industrial purpose without producing an institutional scandal so long as the relevant officers acted under color of clear U.S. statutes.

The Injunction Expands

The petition Walker filed in U.S. circuit court on that Monday sought an injunction restraining Debs and the named officers of the ARU from interfering with the operation of any train carrying mail or moving in interstate commerce, and from inducing any railroad employee to leave his job or to refuse to handle any car. The petition relied on two U.S. statutory hooks. The first was the Postal statute, which made it a U.S. offense to obstruct the mails. The second was the Sherman Antitrust statute of 1890, which prohibited “every contract, combination in the form of trust or otherwise, or conspiracy, in restraint of trade or commerce among the several States.” Olney and Walker had jointly developed the theory that the ARU boycott was a “combination in restraint of trade” within the meaning of the Sherman statute. This was the same Sherman statute that had been written, debated, and passed largely as a response to industrial trusts of the Standard Oil variety. Its application to a trade union was a legal innovation whose audacity was understood at the time only by a small number of observers, principally Henry Demarest Lloyd and the ARU lawyer Clarence Darrow, who would later represent Debs in the contempt proceedings.

Judges Grosscup and Woods granted the injunction the same day it was sought, that Monday. The order, in its final form, did far more than the original petition’s narrow framing suggested. The injunction prohibited Debs, the named ARU officers, and “all other persons whomsoever” from “compelling or inducing, or attempting to compel or induce, by threats, intimidation, persuasion, force or violence, any of the employees” of the named railroads to refuse to perform their duties; from “inducing in any manner” any employee to leave his job; from interfering with any train carrying interstate commerce or mail; and from in any way communicating, signaling, or coordinating any further action that might be construed as continuing the boycott. The phrase “or persuasion” appearing alongside “threats, intimidation, force or violence” was the operative innovation. Persuasion, by its plain meaning, encompassed speech: a workers officer telling another worker that he ought to honor the boycott was now subject to U.S. contempt proceedings for the move of speaking. The blank-prohibition language “all other persons whomsoever” extended the order’s reach to roughly 150,000 ARU members nationally, none of whom had been served with process or had any opportunity to be heard.

The legal scholar David Ray Papke, in his 1999 study “The Pullman Case: The Trial of Labor and Capital in Industrial America,” characterizes the omnibus injunction as the founding instance of what subsequent union lawyers would call the “government by injunction” doctrine, in which U.S. circuit courts substituted equitable orders binding on undefined classes of persons for the older criminal-law machinery of riot, conspiracy, and individual prosecution. Papke’s central claim is that the injunction’s expansion from a narrow mail-obstruction theory to a sweeping prohibition on speech, communication, and association was not a legal accident produced by hurried drafting but a deliberate strategy to convert the Pullman walkout into a U.S. contempt proceeding in which the ARU’s First Amendment, due process, and trial-by-jury defenses would be circumvented. The U.S. contempt power, after all, did not require a jury. The judge who issued the injunction was also the judge who tried the contempt. Conviction required proof of disobedience to the order, not proof of any underlying crime. And the punishment ran to imprisonment of the convicted person without any of the procedural protections that would have attended a criminal prosecution for, say, mail obstruction. Papke argues, and the evidence in the General Managers’ Association’s own minutes supports him, that the injunction’s procedural advantages over criminal prosecution were precisely the point.

Almont Lindsey, in his 1942 study “The Pullman Strike: The Story of a Unique Experiment and of a Great Labor Upheaval,” takes the analysis a step further. Lindsey’s reading is that the Olney-Walker coordination represented a deliberate institutional fusion of U.S. prosecutorial power with private corporate interest, in which the General Managers’ Association obtained, at no charge to itself, the use of the United States Army, the United States Marshal Service, and the U.S. circuit courts as instruments of walkout-breaking. Lindsey’s account of the case files he examined at the Newberry Library and the Chicago Historical Society documents specific instances in which the General Managers’ Association supplied evidence to the U.S. prosecutor, coordinated witness selection, drafted portions of the legal briefs that Walker signed in his U.S. capacity, and footed bills for legal services that were then reimbursed in part by U.S. appropriations through expense vouchers Olney signed. The picture Lindsey draws is of a U.S. prosecutorial apparatus operating, for the duration of the boycott, as a subsidiary of the railroad industry.

Allan Nevins, whose 1932 biography “Grover Cleveland: A Study in Courage” won the Pulitzer Prize and remained for half a century the standard scholarly treatment, sees the matter differently. For Nevins, Cleveland’s decision was a principled defense of the rule of law against a union combination that had unilaterally suspended the operation of the national mails system and major sectors of interstate commerce. Nevins concedes the closeness of Olney’s railroad ties but argues that the legal theory under which the injunction was granted was sound: the Sherman statute did prohibit conspiracies in restraint of trade; the boycott did, as an empirical matter, restrain trade; the obstruction of mail was a U.S. offense; the U.S. circuit courts had jurisdiction; and the President’s constitutional duty under Article II to “take care that the laws be faithfully executed” applied to mail and commerce statutes as much as to any other. Nevins reads Cleveland’s signature as the move of a President doing what the Constitution required, even at significant political cost to himself and to his party. He notes that Cleveland, alone among prominent Democrats of his generation, would lose the 1896 nomination in part because of his Pullman conduct, and that his willingness to accept that cost is evidence of principle rather than corporate capture.

The disagreement between Nevins on one side and Papke and Lindsey on the other is not primarily about facts. The factual record is largely shared. The disagreement is about the relative weight to assign to the rule-of-law framing (Nevins) versus the corporate-state-alignment framing (Papke and Lindsey). Where this article comes down, on the evidence, is closer to Papke. The injunction’s procedural design (covering “all other persons whomsoever,” including “persuasion” alongside “intimidation,” structured to evade jury trial through contempt), the staffing arrangement (the General Managers’ counsel sworn in as special U.S. prosecutor), and the legal theory’s novelty (Sherman statute applied to a trade union one year after the Knight case had cabined its application to industrial trusts) together exceed what a neutral rule-of-law explanation can account for. The rule of law was the vocabulary; the substance was a deliberate use of U.S. machinery to break a union that the railroad industry had targeted for destruction since the Great Northern walkout of the previous spring.

The Cabinet Meeting and the Troop Order

The decision the White House faced on that Tuesday was not whether to enforce the injunction. The injunction would be enforced by U.S. Marshals as a matter of course, and Marshal John W. Arnold had already begun deputizing additional men to serve the orders and arrest defiant strikers. The decision was whether to deploy regular army troops in addition to the marshals. The distinction mattered. Federal marshals were civilian officers; their use to enforce a court order was within the routine machinery of U.S. law enforcement and did not raise the constitutional questions associated with using the standing army for domestic policing. The regular army, by contrast, was governed by the Posse Comitatus statute of 1878, which prohibited the use of the army to enforce civil law except where expressly authorized by the Constitution or by statute. The constitutional authorization the White House relied on was the duty under Article II to ensure execution of U.S. law, supplemented by statutes permitting use of such force where a state failed to suppress an obstruction to U.S. functions. The statutory question of whether the existing statutes actually authorized what the administration proposed to do was, in 1894, genuinely contested, and would not be resolved until the Supreme Court’s 1895 ruling in In re Debs.

The cabinet that met on July 3 was the administration’s second-administration cabinet, smaller and more cohesive than his first. Olney attended; Secretary of War Daniel Lamont attended; Secretary of State Walter Q. Gresham attended; Postmaster General Wilson Bissell attended. Gresham was the only voice in the room with deep prior experience in the railroad industry from a different angle: he had served as a U.S. judge in the Seventh Circuit during the ARU unrest of the late 1870s and 1880s, and his judicial experience had given him a more cautious view of federal court intervention in labor disputes than Olney’s. Gresham’s caution did not extend to opposing the deployment, however; the cabinet minutes summarized in Robert McElroy’s 1923 biography of the administration and in Henry F. Graff’s 2002 American Presidents Series volume on the administration show no recorded dissent from the troop decision. The discussion was about timing, the size of the force, and the chain of command, not whether to send troops at all.

The chain-of-command question matters because of how it constrained General Miles. The order Cleveland signed directed Miles to use his force to enforce the omnibus injunction and to protect the U.S. mails. It did not place the troops under the direction of Illinois state authorities, nor did it require Miles to coordinate with Mayor John Patrick Hopkins of Chicago or with Governor Altgeld. Miles reported directly to the War Department and through it to the President. This was the constitutional and operational feature that Altgeld would, two days later, denounce as the most fundamental violation of federalism in the entire affair: regular army units were being deployed inside a state, in a city with functioning police and a functioning state militia, without the request, consent, or coordination of either the mayor or the governor. Cleveland did not deny this. He affirmed it as the appropriate posture given that the federal interest (mail and interstate commerce) was independent of and superior to any state authority over the same territory. This was a federalism claim of considerable scope, and we will return to it.

The order moved into execution on the evening of July 3. The first units to arrive in Chicago were six companies of the Fifteenth Infantry from Fort Sheridan, approximately twenty-five miles north of the city, which boarded special trains during the night and reached the Chicago stockyards by the morning of July 4. Additional units from Fort Riley, Fort Leavenworth, and Fort Brady were ordered to move toward Chicago over the following days. By the peak of the deployment in mid-month, approximately 2,000 regular army soldiers were operating in and around Chicago, supplemented by 4,000 Illinois National Guard troops (whom Altgeld had federalized himself, despite his political opposition, once violence in the rail yards made state action necessary) and 5,000 deputized marshals. The total U.S.-and-state armed force deployed against the boycott approached 12,000 men.

Altgeld’s Telegram

On the evening of that Thursday, Governor John Peter Altgeld sat at his desk in the executive mansion in Springfield and composed a telegram to the President of the United States that ran to roughly 1,800 words, longer than most state papers the administration had received from any governor since the start of the term. Altgeld was a Democrat. He had been elected governor of Illinois in 1892 on the same ticket that elected Cleveland to his second term. He was, by all standard partisan reckoning, a Cleveland ally. His telegram was nevertheless one of the sharpest documents in nineteenth-century American federalism. It survives in the Cleveland Papers at the Library of Congress, in the Altgeld Papers at the Illinois State Historical Library, and in extensive newspaper reproductions of the period.

Altgeld’s argument operated at four levels. At the constitutional level, he asserted that Article I, Section 8, of the Constitution committed to Congress the power to “provide for calling forth the Militia to execute the Laws of the Union, suppress Insurrections and repel Invasions,” and that the calling-forth statutes Congress had enacted under that authority (most recently the Insurrection statute of 1807) conditioned U.S. use of the army inside a state on either the request of the state legislature, or if the legislature could not be convened, of the governor. Altgeld pointed out that he had made no such request and that the Illinois state legislature, though it could be convened, had not been asked to make one either. The President had therefore deployed regular army units inside Illinois without satisfying the statutory predicate Congress had set for such deployment. This was, Altgeld argued, not merely an inadvisable exercise of U.S. power; it was an exercise of U.S. power that lacked statutory authorization and was therefore unlawful.

At the factual level, Altgeld asserted that local and Illinois authorities were perfectly capable of maintaining order in Chicago and surrounding rail centers, that the Chicago police force under Superintendent Michael Brennan and the Illinois National Guard were on duty and had been on duty since the boycott began, that the violence that had occurred up to July 5 had been on a scale routinely handled by Illinois authorities in similar circumstances, and that the regular army units were not necessary to restore order but had been deployed for a different purpose, which was to break the boycott on behalf of the railroad combination. Altgeld did not, in the telegram itself, explicitly accuse the executive of doing this knowingly. He set out the factual position and let the implication stand.

At the federalism level, Altgeld argued that the precedent Cleveland was setting was extraordinary and dangerous: that if the federal executive could deploy the regular army inside a state to suppress an industrial dispute on the theory that it interfered with mail or interstate commerce, without the consent or coordination of the state’s elected officers, then no labor controversy, indeed no civil disturbance of any kind that touched on a federally regulated commodity, was beyond the reach of unilateral federal military action. Altgeld’s prediction proved accurate. The precedent established in 1894 was invoked by Theodore Roosevelt’s Justice Department in considering intervention in the 1902 coal walkout (though Roosevelt, as Article 32 of this series details, chose the opposite path of mediation); by Woodrow Wilson during the World War I disputes; by Truman in the 1946 rail seizure and the 1952 steel seizure (the latter struck down in Youngstown); and by Reagan in the 1981 PATCO firings examined in detail at “Reagan fires air controllers 1981”. The lineage of executive intervention against organized workers runs through this 1894 moment.

At the political level, Altgeld argued that the administration had been misinformed by Olney about the actual situation on the ground in Illinois, and that the misinformation reflected the partisan and corporate interests of the Attorney General more than any honest assessment by U.S. officers on the ground. The telegram named no specific allegation of bad faith against Olney but documented eight specific factual claims in Olney’s reports to the president that Altgeld asserted were either false or substantially exaggerated.

Cleveland’s reply, telegraphed back to Springfield the morning of July 6, was 350 words. It conceded none of Altgeld’s substantive points. The president’s position was that national authority to enforce U.S. law operated independently of gubernatorial consent; that the requirements of the Insurrection statute applied only to the use of regular army units to suppress general insurrection within a state, not to the use of such units to remove obstructions to specific U.S. functions like the mail and interstate commerce; that the necessity of the deployment was a determination for the national executive, not the governor, to make under the Article II Take Care Clause; and that the deployment would continue. The reply ended with a single sentence acknowledging that the governor’s concerns had been received and considered, which Altgeld understood as a polite version of refusal.

Altgeld sent a second telegram on July 6 reiterating his constitutional arguments and asking specifically for the legal authority under which regular army units were operating without gubernatorial consent. Cleveland’s second reply, sent July 7, declined to engage the question further, citing the U.S. circuit court’s injunction and the duty to enforce it. The exchange went no further. Altgeld then took the unprecedented step of writing an open letter to the president published in the Chicago press, in which he restated the constitutional case at greater length and accused the administration of having sided with the railroad combination. The political consequence for Altgeld, in Democratic politics, was severe; he would lose his reelection campaign in 1896 in part because of the perception (cultivated by the Cleveland wing of the party) that he was a radical sympathizer with the strikers. The political consequence for Cleveland, in the longer term, was equally severe and likely more historically consequential: the Democratic Party in 1896 would repudiate his wing, nominate William Jennings Bryan on a platform that explicitly denounced “government by injunction,” and begin a long realignment that would eventually move organized workers into a Democratic coalition the second Cleveland administration had effectively severed.

The Troops on the Ground

The arrival of regular army units on July 4 did not immediately end the boycott. It intensified the confrontation. Crowds of strikers and sympathizers, joined by significant numbers of local residents who had no employment relationship with the railroads but who treated the deployment as an intolerable provocation, gathered around the stockyards, the Pan-Handle yards on the South Side, the Illinois Central terminals, and the Rock Island junction at Blue Island. The violence that followed (and that has been retroactively used to justify the troop deployment) was largely a product of the deployment itself, not a condition that preceded it. The U.S. commission appointed by the president on July 26 and whose November 1894 report remains the most thorough contemporaneous U.S. investigation of the events documented this point with care. Up to July 4, when troops first appeared in Chicago, property damage in the rail yards had been limited and had largely consisted of obstruction of switches and uncoupling of cars; deaths attributable to the boycott were two. Between July 5 and July 10, with troops on the ground and confrontations escalating, property damage rose dramatically and the death toll climbed to at least thirteen, with hundreds more injured.

The largest single incident occurred on July 6 at the Panhandle yards, where rioting destroyed approximately seven hundred rail cars by fire, the largest such loss on any single night of the conflict. The fires were lit, according to the commission’s reconstruction based on witness testimony, by elements of the crowd that had drawn in unemployed workers, criminal opportunists, and local residents angered by the troop deployment, not by ARU members operating under union direction. The ARU had explicitly disavowed property destruction as a tactic from the start of the boycott, and Debs’s correspondence and public statements throughout June and early July had pressed ARU members to maintain a strict discipline of non-violent obstruction. The discipline had held under the marshals, where confrontation was sporadic and resolvable through arrests of specific individuals. The discipline broke under the troops, where confrontation became continuous and undifferentiated, and where the symbolism of the troop deployment radicalized actors who had not previously been participants.

This was the second-order causal pattern the commission documented and that subsequent scholarship, particularly Lindsey and Papke, has reinforced. The troop deployment did not respond to a violent strike; it produced one. The argument that the deployment was necessary to restore order is therefore difficult to sustain on the evidence the national government’s own commission compiled. Nevins addresses this point indirectly by arguing that even if the deployment caused short-term escalation, it was necessary in the longer arc to restore the operation of U.S. functions. This is a defensible argument as a matter of choice (one might accept short-term escalation for long-term restoration); it is not a defense of the framing that the troops were a response to violence, since the major violence followed the troops rather than preceding them.

By July 10, the boycott was breaking. Trains were moving on most lines, escorted by troops and operated by replacement workers (many of them recruited from the African-American population that the ARU had explicitly excluded from membership, a strategic error whose contribution to the ARU’s defeat the historian Eric Arnesen has documented in his work on Black railroad workers). On July 13, Debs and the ARU’s executive committee, recognizing that the boycott had collapsed as an operational matter, met with leaders of the Chicago Federation of Labor and proposed a settlement in which the ARU would call off the boycott in exchange for amnesty for striking workers and reinstatement of fired ARU members at the railroads. The General Managers’ Association rejected the proposal. There would be no settlement and no amnesty. The Association’s position was that the ARU would be destroyed, not negotiated with.

The Arrest of Debs

On July 17, 1894, Eugene Debs and three other ARU officers (Vice President George W. Howard, Secretary Sylvester Keliher, and Director L. W. Rogers) were arrested at the ARU headquarters on contempt charges based on their alleged violation of the omnibus injunction. They were also indicted, separately, under the Sherman Antitrust statute for conspiracy in restraint of trade. The contempt proceeding was tried before Judge Woods, one of the judges who had issued the injunction, sitting alone without a jury. The Sherman Act prosecution would have required a jury but, in a tactical choice by Olney that subsequent observers have called shrewd or scandalous depending on their inclinations, the Sherman Act indictment was permitted to languish while the contempt proceeding went forward.

The contempt trial, conducted in December 1894 and January 1895 with Clarence Darrow leading the defense and Edwin Walker leading the prosecution, was substantively a trial of the injunction itself: Darrow’s principal argument was that the order Debs was accused of violating was so broad as to exceed the U.S. circuit court’s equity jurisdiction, that it violated the First Amendment by prohibiting persuasion, that it violated the Fifth Amendment by binding persons not parties to the original proceeding, and that the Sherman Act could not be read to cover a trade union. Judge Woods rejected each argument. He found Debs and the other ARU officers in contempt and sentenced them to six months in the McHenry County Jail at Woodstock, Illinois. Debs began serving his sentence in January 1895. He used the months in jail to read Marx, Engels, Bellamy, and Kautsky; he would emerge a socialist, founding the Social Democratic Party of America in 1898 and running for president as the Socialist Party of America’s candidate five times between 1900 and 1920, receiving 901,000 votes from prison in 1920. The radicalization of Debs by his imprisonment is one of the under-recognized consequences of his 1894 decision.

The defense appealed the contempt conviction to the Supreme Court, where the case became In re Debs, decided May 27, 1895. Justice David J. Brewer wrote a unanimous opinion affirming the conviction and, in doing so, established the constitutional architecture that would govern U.S. union injunctions for the next forty years until the Norris-LaGuardia Act of 1932 restricted their use. Brewer’s opinion did not rely on the Sherman Act. It rested instead on a broader theory of the federal government’s sovereign interest in the free flow of interstate commerce and in the operation of the mails, and on the equity power of U.S. circuit courts to enjoin obstructions to those interests. The opinion explicitly affirmed the use of such injunctions to restrain union activity that interfered with commerce, the use of contempt rather than jury trial to enforce such injunctions, and the deployment of regular army units to give effect to such orders. Each of these holdings would shape rail-industry labor relations for the next four decades.

Brewer’s opinion is one of the more revealing documents in nineteenth-century constitutional law. It treats the question of whether the Sherman Act applied to union as essentially irrelevant: the United States’ authority to move, the opinion holds, does not depend on the existence of any specific statute, but on the inherent sovereign power to protect the free flow of commerce and the mails. This is a remarkable claim. It would not be available to a modern court reading the Constitution as a strict charter of enumerated powers. It was available in 1895 to a court that understood the national government’s authority as residing in part in inherent sovereign powers analogous to those of any sovereign state, an understanding that Reconstruction-era jurisprudence had pushed toward in cases like the Slaughterhouse Cases (1873) and the Civil Rights Cases (1883) and that the ARU cases of the 1890s consolidated.

The Olney-Walker Correspondence and the Theory’s Construction

The Olney-Walker correspondence preserved in the Olney Papers at the Library of Congress is the indispensable contemporaneous document for understanding how the legal theory affirmed in In re Debs was assembled. Between June 26 and July 2, 1894, Olney and Walker exchanged at least eleven letters and a comparable number of telegrams, several of which were also copied to Assistant Attorney General John Maynard Harlan in Washington. The exchange shows a deliberate division of intellectual work. Walker, on the ground in Chicago, drafted the factual predicates and the operational language of the petition. Olney, in Washington, drafted the broader doctrinal framing and the strategic justifications for invoking the Sherman Act against a union for the first time. Each letter responded to the previous one with refinements, and the omnibus order entered by Judges Grosscup and Woods on that Monday morning reflects the seventh or eighth iteration of the theory.

What the correspondence reveals, more than any other surviving source, is that the doctrinal claim Brewer would later affirm as resting on inherent sovereign power was originally pleaded by Walker on a much narrower mail-obstruction theory. The expansion to a generalized commerce-protection doctrine was Olney’s addition, made in a letter dated June 28, 1894 in which he urged Walker to plead the broader theory in the alternative so that the equity court would have the option to enter an order that did not depend on the mail-obstruction predicate alone. The strategic reasoning Olney articulated was that the mail theory was vulnerable to the General Managers’ Association’s own conduct (the Association had instructed member lines to refuse to run mail trains unless the union withdrew its boycott of sleeping cars, which made the obstruction question turn on whose refusal preceded whose), while the broader commerce theory was invulnerable to any such argument because the boycott unquestionably affected interstate commerce regardless of who was technically obstructing what.

This is the key doctrinal move of the case, and it was made by Olney in a private letter five days before Brewer would announce it as the holding of the Supreme Court. The architecture of In re Debs was not, in this sense, the Court’s own construction; it was the prosecutor’s construction, ratified by the Court a year later. Brewer’s opinion adopts the Olney framing nearly verbatim in places, a textual fact that the legal historian William Forbath documents in his 1991 study and that establishes the prosecutorial origin of the constitutional doctrine more decisively than any prior account had.

What the correspondence does not show is any sustained effort by Olney to test the doctrinal theory against possible objections. The letters proceed by accumulation rather than by adversarial stress: each refinement extends the theory’s reach without anyone in the chain of drafters asking whether the extension was defensible. Walker, as a working corporate attorney, was paid to maximize his client’s litigation position; Olney, as Attorney General, had a duty to ask whether the maximalist position served the public interest separately conceived. He did not ask. The correspondence reveals a prosecutorial mind operating in advocacy mode, not in counselor mode. This is, by itself, a structural critique of the office Olney occupied.

The Strike Commission Report

In a move that has long puzzled historians of his administration, the president on July 26, 1894 (with the boycott effectively broken but troops still on the ground in Chicago) appointed a three-member U.S. commission consisting of Carroll D. Wright, the U.S. Commissioner of Labor; John D. Kernan of New York; and Nicholas E. Worthington of Illinois. The Commission’s charge was to investigate the causes of the boycott, the conduct of the parties, and the role of national authority. Hearings began August 15, 1894, in Chicago, and continued through September. Testimony was taken from the company’s founder (who appeared reluctantly and was sharply questioned), from Debs (who appeared from his Cook County Jail cell, where he was being held pending the contempt trial), from members of the General Managers’ Association, from railroad workers, from the company town residents, from U.S. officials including Olney’s deputies, and from state and local officials including representatives of Altgeld. The Commission’s November 1894 report ran to 681 pages with appendices.

The report’s findings were unsparing. It found that the Palace Car Company’s wage reductions had been disproportionate to its dividend policy (Pullman had paid full dividends to shareholders throughout the period it was cutting workers’ wages, drawing the cash from the firm’s substantial reserves) and that the company town arrangement constituted an “un-American” concentration of authority over employees’ lives that no employer should have. It found that the General Managers’ Association was, in form and function, a combination in restraint of trade that, if applied to the ARU side of the rail industry as the Sherman Act had been applied to the union side, would have been similarly subject to U.S. prosecution; the asymmetry of enforcement, the report noted, was striking. It found that the omnibus injunction had been over-broad in its terms and that the application of the Sherman Act to a trade union was a stretching of the statute beyond its evident purpose. It found that the deployment of regular army units, while within his authority as he had interpreted it, had been timed and structured in ways that escalated rather than de-escalated the conflict. And it found that the absence of any national mechanism for compulsory arbitration of railroad labor disputes had left the country without a peaceable means of resolving the kind of conflict that this case represented, and that Congress should remedy this absence by statute.

Congress did not remedy it for another four years. The Erdman statute of 1898, which finally established national mediation machinery for railroad union disputes, was a direct response to the commission’s recommendations. The Newlands Act of 1913 expanded the Erdman framework. The Railway Labor Act of 1926 institutionalized the structure that has since governed railroad union relations and that, with modifications, was extended to airline union in 1936. The legislative arc of national mediation, in other words, originates in the commission’s diagnosis of what his intervention had failed to accomplish. The Commission was, in this sense, his own posthumous critic of his own decision, appointed by him and reporting honestly back to a country that the report itself documented had been substantially failed by the national executive in July 1894.

Cleveland did not publicly accept the Commission’s findings, nor did he repudiate them. He let the report stand. In his 1904 retrospective volume “Presidential Problems,” published after his presidency, he defended his 1894 conduct in language that conceded almost nothing to his critics. The “absolute necessity” of national action, he wrote, “rested upon a duty so apparent that no consideration of consequences could be allowed to interfere with its discharge.” This was the Nevins reading exactly, in his own words. The disagreement with the Commission was thus a disagreement the president left intact rather than resolved, and it is the unresolved disagreement that has structured the historiography ever since.

The Decision Tree Reconstructed

his choice on that Tuesday can be modeled as a decision among five available options, only two of which were seriously considered and one of which was selected. Reconstructing the options the way a contemporary observer with access to the cabinet’s actual deliberations would have seen them clarifies what was chosen and what was foregone.

The first option was non-intervention. The national government could have treated the boycott as a private union dispute and a state law-enforcement matter, declined to enforce the mail-obstruction theory beyond routine U.S. protection of individual mail cars, and allowed Illinois state authorities to maintain such order as they could maintain. The cost of this option was the loss of U.S. mails service on affected lines for the duration of the boycott, which by July 3 had reached approximately ten days. The benefit was preservation of the constitutional norm against U.S. military intervention in domestic union disputes without gubernatorial consent. Olney rejected this option in his June 30 memorandum to the president and the cabinet did not seriously revisit it on July 3.

The second option was marshal-only enforcement. The U.S. government could have used U.S. Marshals to serve the injunction, arrest specific defendants who violated it, and provide armed escort to specific mail cars or trains while declining to deploy regular army troops. This was a narrower form of the response actually chosen. It would have provided a U.S. enforcement mechanism while preserving the Posse Comitatus principle against using the army for domestic law enforcement. Marshal Arnold’s actual reports to Olney during the first week of midsummer had described the marshal force as overstretched but capable of focused operations; whether marshal-only enforcement could have moved the trains was contested but not impossible. This option was discussed in the cabinet on July 3 and rejected on the ground that the scale of the boycott exceeded what the marshal service could handle.

The third option was state-coordinated national action. The U.S. government could have requested that Governor Altgeld federalize the Illinois National Guard for service alongside U.S. marshals, with the regular army held in reserve at Fort Sheridan in case the combined state-U.S. civilian force proved insufficient. This was the option Altgeld himself would have preferred, as his July 5 telegram makes clear, and it was the option that would have satisfied the Insurrection Act’s statutory predicate for state consent. The cabinet rejected it on the ground that Altgeld could not be relied upon to fully commit Illinois forces against the strikers given his publicly expressed sympathy for union and his administration’s prior decisions in the Haymarket pardon case of 1893 (the pardon of the surviving Haymarket defendants had cost Altgeld considerable political support among business interests and signaled, to Olney and others in the Cleveland administration, that he was unreliable on questions of union enforcement). Whether Altgeld could in fact have been so relied upon is one of the open counterfactuals of the case; on the evidence of his actual conduct (he did federalize the Guard, on his own initiative, when the violence escalated after the troops arrived), the cabinet’s distrust may have been misplaced.

The fourth option was the path actually taken: unilateral U.S. deployment of the regular army to enforce the injunction and protect the mails, without state coordination and over state protest. This was Olney’s recommended option and his chosen option. Its features were already described.

The fifth option, never seriously considered but worth recovering because it would have set a different precedent, was presidential mediation. Cleveland could have proposed, as Roosevelt would do in 1902 for the coal strike (the parallel case examined in Article 32, “TR and the Coal Strike: The 1902 Mediation Precedent”), to convene the parties at the White House under U.S. auspices and to broker a settlement that would have moved the trains and provided some redress to the strikers. This would have required either Pullman’s willingness to arbitrate (which he had already rejected on June 26) or significant U.S. leverage applied to Pullman to compel his participation. Cleveland had such leverage available: the Palace Car Company operated under various federal-state regulatory regimes, and the mail contracts it benefited from could have been used as a coercive tool. Why this option was not considered is the deepest counterfactual question the case raises. The shortest answer is that Cleveland did not conceive of the office as an industrial broker. He conceived of it as a law enforcer. The conceptual move that Roosevelt would make eight years later, treating the executive as a third party with legitimate interests in industrial peace independent of the parties’ legal positions, was not available in 1894 because the political theory of the office had not yet articulated it.

A Findable Artifact: The Injunction’s Expansion, Hour by Hour

The injunction’s transformation from a narrow mail-protection theory to the sweeping omnibus order issued July 2 happened over roughly seventy-two hours, between the morning of June 29 and the late afternoon of July 2. Reconstructing this expansion is the most concrete way to see the deliberate strategy at work.

On the morning of June 29, 1894, Edwin Walker filed an initial petition in U.S. circuit court seeking an injunction against the ARU’s “obstruction of the United States mails.” The petition was narrow. It named Debs and three other ARU officers, identified specific mail trains being obstructed on specific lines, and requested an order requiring the named defendants to refrain from interfering with the carriage of those specific trains. Walker informally circulated the petition to Judges Grosscup and Woods that afternoon for guidance on whether the equity court would grant injunctive relief on a U.S. mails-obstruction theory. Both judges indicated, in chambers, that they would consider such relief but suggested that the petition’s narrow framing might prove operationally inadequate: the boycott was spreading faster than any city-by-city or line-by-line enumerated relief could constrain it, and the judges suggested that Walker consider a broader theory.

By the evening of June 30, Walker, with input from Olney by telegraph from Washington, had redrafted the petition to add a Sherman Act count and to broaden the requested relief to cover “all railroads engaged in interstate commerce” and “all employees and others” inducing strike action. This was a substantial expansion. The Sherman Act count converted the case from a postal regulation matter into an antitrust matter, opening the U.S. circuit court’s full equity jurisdiction to address restraints on trade. The “all employees and others” language created the class-prohibition feature that would later allow the court to find non-parties in contempt.

On July 1, Walker filed the amended petition. Judges Grosscup and Woods conferred and indicated they would issue the injunction the following day, with one further modification: the inclusion of “persuasion” in the list of prohibited means by which strike action could be induced. The addition of “persuasion” alongside “force, intimidation, threats, and violence” was the most legally aggressive feature of the final order; it converted speech itself into a prohibited move, on the theory that persuading a worker to honor the boycott was equivalent to coercing him. The legal precedent for treating persuasion as equivalent to coercion in equity proceedings was thin (some 1890s state cases had moved in this direction, but no national authority had embraced it explicitly), and the inclusion of the term in the final order would be the principal Free Speech objection Darrow would raise in the Debs contempt proceeding.

On the morning of July 2, the final injunction was entered. Marshal Arnold began serving it within hours. By that evening, copies of the order had been posted at major rail yards in Chicago and were being telegraphed to other U.S. districts for similar enforcement. On the morning of July 3, Olney walked into the cabinet with the order in hand and the troop request in draft. The legal architecture had been built in seventy-two hours, the operational decision was made in one cabinet meeting, and the historical precedent was set by the time the Fifteenth Infantry reached the stockyards on July 4. The speed of construction is part of what the architecture concealed: it was novel, but it moved fast enough that the novelty was a fait accompli before serious opposition could organize.

Where the Historians Disagree

The the case has been a touchstone of American union historiography for more than a century, and the principal interpretive disagreements among its scholars cluster around three questions. Each merits a paragraph in its own right.

The first question is whether he’s action was a defense of national authority against insurrection or an executive instrument of class warfare. Nevins’s 1932 biography established the first reading and gave it scholarly prestige; Henry F. Graff’s 2002 American Presidents Series volume on the president substantially endorses Nevins on this point while conceding more ground than Nevins did to the procedural critiques. Lindsey’s 1942 study and Papke’s 1999 monograph anchor the second reading. H. W. Brands, in “American Colossus” (2010), offers a middle position: the president believed sincerely in the rule-of-law framing but was substantively wrong in his assessment of what the rule of law required, because the legal architecture Olney had built was itself a corporate-state alignment masquerading as neutral law enforcement. Brands’s reading is the position this article endorses. The president was not a hypocrite; he was not a tool of capital; he was a man whose conception of the national executive’s duty took for granted a set of substantive premises about commerce, contract, and union that had been engineered by interested parties faster than he could critically examine them.

The second question is whether the precedent the case set was inevitable given the broader political-economic forces of the 1890s, or whether contingent choices by the president and Olney produced an outcome that other available choices would have averted. The inevitabilist reading, associated with Marxian and structuralist historians like David Montgomery and Melvyn Dubofsky, treats Pullman as the predictable working out of a class struggle whose terms were set long before the individual decisions of that month mattered. The contingentist reading, associated with Lindsey and with the more recent work of Joseph A. McCartin on twentieth-century union cases, treats Pullman as a moment of genuine choice where alternative paths were available and where the particular features of Olney’s legal architecture (the dual prosecutor-counsel role, the Sherman Act stretch, the persuasion-equals-coercion innovation) were contingent moves rather than necessary ones. This article leans contingentist. The same political-economic forces in 1902 produced Roosevelt’s mediation, and the same forces in 1894 could have produced something other than what he chose if he or someone at his table had pushed back on Olney’s framing.

The third question is whether the case is best understood as a union-history episode or as a constitutional-law episode. The union-history reading, which dominates Lindsey, Dubofsky, and Brands, treats the boycott’s defeat as the central event and the legal architecture as the instrument by which the defeat was accomplished. The constitutional-law reading, which dominates Papke and the legal scholar William E. Forbath in “Law and the Shaping of the American Labor Movement” (1991), treats the legal architecture as the central event and the boycott’s defeat as the immediate occasion. Both readings are sustainable; they emphasize different aspects of the same underlying conduct. For purposes of this series, which is centrally concerned with how the modern presidency was built decision by decision, the Forbath-Papke framing is more useful, because it identifies what he’s signature on the troop order added to the executive’s permanent toolkit beyond what was available to his predecessors.

The Complication: Commerce, Capital, or Both

The strongest objection to the corporate-state-alignment reading is that there was, in fact, a real national interest in the operation of interstate commerce and the U.S. mails that the boycott obstructed, and that no national executive committed to enforcing U.S. law could have ignored that interest indefinitely. The boycott, by July 3, had stopped passenger and freight movement on roughly thirty rail lines, had interrupted mail delivery to large portions of the trans-Mississippi West, had cut off perishable produce shipments from California that were rotting on rail sidings, and had threatened the coal supply that powered urban heating and industry across the Midwest. These were genuine U.S. concerns. The Article II Take Care Clause was not invented in 1894 to give he cover; it had been there since 1789 and had been invoked by presidents from Washington (the Whiskey Rebellion of 1794) through Jackson (the Nullification Crisis of 1832) to authorize national action against domestic obstructions to U.S. functions. The mail and interstate commerce were U.S. functions in 1894 in a way that they had not been at the U.S. level for much of the prior century, because the post-Civil War regulatory state had brought them under U.S. supervision through the Interstate Commerce Act of 1887 and the Post Office Department’s expansion. The national interest, in other words, was both genuine and growing. To dismiss it entirely in favor of the corporate-capital reading would be to read 1894 with the cynicism of 1934 retrospectively imposed.

The honest answer is that both readings are simultaneously true. The national interest in commerce and mail was genuine. So was the corporate-state alignment. The skill of Olney’s architecture lay precisely in fusing them: presenting an enforcement of the genuine national interest in a procedural and institutional form that maximally served the corporate interest in destroying the union. Papke’s contribution to the literature is to show this fusion in operational detail. Nevins’s contribution is to insist that the genuine national interest was not nothing, and that a president who treated it as nothing would have been failing in his own constitutional duty. The judgment a careful reader should make is that he satisfied his constitutional duty in a manner that, in retrospect, can be seen to have been corporate-captured at the level of operational design even where it was honorable at the level of presidential intention. Both can be true. The historiography has often forced a choice between them; the evidence does not require the choice.

Verdict

Grover’s that Tuesday signature on the troop order was both a defensible and an indefensible move, defensible at the level of constitutional duty as he understood it, indefensible at the level of operational design as Olney built it. The deployment was unnecessary on the evidence; the marshal force, properly augmented and coordinated with the Illinois National Guard, could have served the injunction and protected the mails without the regular army’s intervention. The injunction itself was over-broad and procedurally novel in ways that betrayed its purpose; the inclusion of “persuasion” alongside “intimidation” and the binding of “all other persons whomsoever” were drafting moves designed to circumvent jury trial and First Amendment defenses, and they should not have survived appellate review even under the constitutional law of 1895. The Sherman Act’s application to the ARU was a stretching of statutory language beyond evident purpose and should have been corrected by the Supreme Court rather than ratified. The Strike Commission’s own findings, appointed by Grover and reporting back to him in November 1894, documented each of these objections in language Grover chose not to engage. He was wrong on the operational design. He was right on the constitutional duty. The first error has had consequences for national-union relations that the second consideration cannot offset.

The verdict here is closer to Papke and Lindsey than to Nevins, but it does not endorse the simpler corporate-capture reading. Grover was not Olney’s tool; he was Olney’s principal, and the responsibility for the operational design Olney built is therefore his. A different attorney general would have built a different architecture, and a more critical president would have asked Olney harder questions on July 3. Grover did not ask them. He asked one question and signed.

Legacy: The Precedent Chain

The lineage of U.S. union intervention that begins with the president’s July 1894 deployment runs through every twentieth-century episode in which U.S. force or such injunctions have been brought to bear on union disputes. Wilson during World War I extended the framework: the 1917 Adamson Act, the wartime nationalization of the railroads, and the use of such injunctions in the postwar union strife (the 1919 coal strike particularly) all drew on the Pullman precedent. Hoover used the framework in considering responses to the 1932 Bonus Army episode, though his use of regular army troops there was a different kind of deployment and is treated in Article 39 of this series, “Hoover and the Bonus Army.” Truman attempted to extend the framework in the 1946 rail seizure and again in the 1952 steel seizure, and the Supreme Court’s 1952 ruling in Youngstown Sheet and Tube v. Sawyer is the case most directly in tension with In re Debs; the two together define the limits of executive labor authority in the modern period. Reagan in 1981 used a different but related framework in the PATCO firings examined in Article 59, “Reagan fires air controllers 1981”; the PATCO response did not rely on regular army deployment but it relied on the same broad theory of executive authority over labor in essential industries that Pullman had established. The institutional architecture Cleveland and Olney built in 1894 has been bent, restricted, and partly reversed by the Norris-LaGuardia Act of 1932 and the National Labor Relations Act of 1935, but its core principles (U.S. power to enjoin union activity affecting interstate commerce, U.S. use of force to give effect to such injunctions, executive authority to determine the necessity of such force) remain available to the modern presidency.

This article belongs to the series house thesis that the modern presidency was forged in four crises and that every emergency power created in those crises outlived the emergency. The the case is not one of those four crises, but it is a precursor to all of them. It established, at low stakes by the standards of the world wars and the Cold War, that the national executive could act unilaterally inside a state, against the protest of that state’s elected officers, in defense of a national interest the executive itself defined. This authority was available to Wilson in 1917, to FDR in 1941, to Truman in 1950, and to every modern president since. The the case is in this sense a small precedent for what later became the large pattern. Cleveland could not have foreseen the use to which his 1894 framework would be put. But he built it, and what came after, came after.

The retrospective irony of the case, and the deepest reason it merits sustained attention even at a distance of more than a century, is that the president’s intervention was supposed to defend the rule of law and ended by demonstrating how the rule of law could be redesigned, through U.S. circuit court procedures and executive deployment patterns, to serve interests the law had not previously been understood to serve. The arc that runs from 1894 through 1981 is not the arc of a national executive defending neutral law; it is the arc of a U.S. executive whose conception of law evolves alongside the political-economic interests it serves. The first link of that chain was signed at 5:30 in the afternoon on July 3, 1894, by a president who thought he was discharging a constitutional duty and who was, in part, doing exactly that, while also building an apparatus he did not understand.

The President’s Self-Assessment and the Verdict of His Party

In “Presidential Problems” (1904), Cleveland devoted a long chapter to the case under the heading “The Government in the Chicago Strike of 1894.” The chapter is one of the more revealing self-assessments any nineteenth-century president has left. Cleveland conceded nothing. He treated the deployment as obviously correct, the constitutional questions as trivially resolved by the Supreme Court’s unanimous affirmance, and Altgeld’s objections as the product of a sentimental sympathy with labor that no responsible federal executive could entertain. He did not address the Strike Commission’s findings directly. He did not address the dual-role appointment of Edwin Walker. He did not address the persuasion-equals-coercion innovation in the injunction. He addressed, simply, the largest stakes (mail and commerce) and the most defensible features of his own conduct (a willingness to act against his own party’s labor sympathies for the sake of constitutional duty). The retrospect is the position of a man who has not been moved by the criticism his own commission produced. This is, by itself, evidence of something about how the case operated in his mind. He did not, by 1904, see what he had built. The opacity to its architect is part of the legacy.

The verdict of the president’s own party was less merciful. The Democratic National Convention of 1896, meeting in Chicago in early July, repudiated the Cleveland wing on every major issue: silver against gold, easy money against hard, government by ordinary process against government by injunction. The platform’s express denunciation of “government by injunction” referred specifically and pointedly to the Pullman case. Bryan’s nomination was the formal repudiation. Altgeld’s prominent role in the convention (he chaired the Illinois delegation and was widely credited with delivering the platform language on injunctions) was the second repudiation, by a Democratic governor whose career the president’s conduct had largely destroyed but whose vindication the convention extended. Cleveland did not attend. He did not endorse Bryan, voting in November for the third-party Gold Democrat candidate John M. Palmer. The break between Cleveland and the future of his party was complete, and the Pullman case was its proximate cause. A president whose handling of a major decision splits his own party in the next election cycle has, by any standard of political consequence, made a decision of historical magnitude. Whether the magnitude was for the better or for the worse is a question this article has answered: for the worse, on operational design; for the more complex, on constitutional doctrine; and for the longest reach, in establishing executive authority over labor that the twentieth century would test, refine, and eventually contest.

Frequently Asked Questions

Q: Why did Cleveland send regular army units to Chicago during the Chicago boycott?

President Cleveland sent the regular army into Chicago on July 3, 1894 to enforce a U.S. circuit court injunction restraining the American Railway Union from continuing its boycott of trains carrying Pullman sleeping cars. The official legal justification was that the boycott was obstructing the United States mails and interstate commerce, both national interest under Article I and the Take Care Clause of Article II. The deeper political reality, documented in the U.S. Strike Commission’s November 1894 report, is that the troop deployment served the interests of the General Managers’ Association, a combination of twenty-four rail lines that had been organizing to crush the ARU since the union’s successful Great Northern strike of spring 1894. The two motivations operated together. Cleveland believed he was discharging a constitutional duty; the structure Attorney General Richard Olney had built around that duty was simultaneously a corporate strike-breaking apparatus, and the historical judgment is that both can be true.

Q: Did Governor Altgeld of Illinois oppose the federal troop deployment?

Yes, vigorously. Governor John Peter Altgeld sent Cleveland a telegram on that Thursday running to approximately 1,800 words denouncing the deployment as unauthorized national intervention into state police authority. Altgeld argued that the Insurrection Act of 1807 required either a state legislative request or, where the legislature could not be convened, a request from the state executive before regular army units could be deployed inside a state, and that Cleveland had received neither. Altgeld also contended that local and state authorities were capable of maintaining order without federal military involvement and that the federal deployment had been engineered to break the boycott on behalf of the railroad combination rather than to address genuine disorder. Cleveland’s July 6 reply rejected each of Altgeld’s arguments. The exchange remains one of the sharpest documents in nineteenth-century American federalism, and Altgeld’s constitutional position has aged better than the president’s reply.

Q: What was the legal basis for the that Monday injunction against Eugene Debs and the ARU?

The injunction issued by Judges Peter S. Grosscup and William A. Woods on that Monday rested on two federal statutory hooks and one broader equity theory. The first hook was the Postal statute, under which obstruction of the mails was a federal offense. The second was the Sherman Antitrust Act of 1890, which prohibited combinations in restraint of trade. The novel application of the Sherman Act to a labor union (rather than to an industrial trust, its evident original target) was the legally aggressive move that Attorney General Olney and the General Managers’ Association’s counsel Edwin Walker had jointly developed. The Supreme Court’s 1895 ruling in In re Debs eventually ratified the injunction not on the Sherman Act theory but on a broader theory of federal equity power to protect interstate commerce and the mails, an authority the Court treated as inherent in national sovereignty rather than requiring specific statutory grant.

Q: Why was Eugene Debs imprisoned after the Pullman Strike?

Debs was convicted of contempt of the federal court that had issued the July 2 injunction, on the ground that he had violated the order’s prohibition on inducing rail employees to refuse to handle Pullman cars. The contempt trial was conducted before Judge Woods, one of the two judges who had issued the injunction, sitting without a jury. The Supreme Court in In re Debs (1895) affirmed the conviction. Debs served six months at the McHenry County Jail in Woodstock, Illinois beginning in January 1895. During his imprisonment he read extensively in socialist theory, including Marx, Engels, Bellamy, and Kautsky, and emerged from jail a committed socialist. He founded what became the Socialist Party of America in 1898 and ran for president five times. His radicalization by federal imprisonment is one of the most consequential second-order effects of the president’s 1894 decision.

Q: What did the U.S. Strike Commission find about the president’s handling of the Pullman Strike?

The three-member Strike Commission appointed by Cleveland on July 26, 1894 (chaired by U.S. Commissioner of Labor Carroll D. Wright) reported in November 1894 with findings sharply critical of multiple aspects of the federal response. The Commission found that the Palace Car Company’s wage reductions had been disproportionate to its dividend policy, that the company town arrangement was un-American in its scope, that the General Managers’ Association was itself a combination in restraint of trade whose asymmetric exemption from antitrust enforcement was striking, that the omnibus injunction had been over-broad, that the application of the Sherman Act to labor was a stretching of the statute, and that the troop deployment had been timed and structured in ways that escalated rather than de-escalated conflict. The Commission recommended national mediation machinery, which Congress eventually provided through the Erdman statute of 1898. Cleveland did not publicly accept the findings but did not repudiate them either.

Q: Was Attorney General Olney biased because of his railroad ties?

Olney came to the cabinet in 1893 directly from a Boston law practice in which he had been general counsel to the Boston and Maine Railroad and the Chicago, Burlington and Quincy Railroad. He had served on the Burlington’s board of directors. He continued to draw a retainer from the Burlington during his tenure as Attorney General, a fact documented in the railroad’s board minutes that Cleveland’s administration never disclosed. Whether this constituted a conflict of interest under the federal conflict-of-interest law of the 1890s is technically arguable; the U.S. statutes governing executive branch conflicts did not take their modern form until the 1962 amendments. Whether it shaped his conduct is not arguable. His June 1894 correspondence with Edwin Walker (the General Managers’ Association’s counsel whom Olney appointed as special U.S. prosecutor on June 30) shows the two men coordinating legal strategy in real time. The dual role of Walker, simultaneously the Association’s lawyer and the federal prosecutor pursuing its industrial opponent, was the operational expression of the bias.

Q: Did the Sherman Antitrust Act apply to labor unions?

The Sherman Act of 1890 prohibited “every contract, combination in the form of trust or otherwise, or conspiracy, in restraint of trade or commerce among the several States.” Its evident legislative target was the industrial trust, of which Standard Oil was the contemporary archetype. The application of the statute to labor unions was a legal innovation pioneered by Olney and Walker in the Pullman case. The Supreme Court’s 1895 ruling in United States v. E. C. Knight Co. had, just months earlier, narrowed the Sherman Act’s reach against industrial combinations on the ground that manufacturing was not commerce; the simultaneous broadening of the Act to reach labor unions while narrowing it against manufacturing trusts produced an asymmetry that subsequent observers (and the Strike Commission) treated as troubling. The Clayton Antitrust Act of 1914 attempted to exempt labor unions from antitrust prosecution, though courts continued to apply Sherman Act doctrine to union activity into the 1930s until the Norris-LaGuardia Act and the New Deal labor statutes substantially altered the framework.

Q: How did the Pullman Strike end?

The boycott collapsed operationally by July 13, 1894, roughly ten days after regular army units arrived in Chicago. Trains were moving on most affected lines, escorted by federal soldiers and operated by replacement workers, many of them recruited from the African-American population that the ARU had excluded from membership (a strategic error whose contribution to the union’s defeat the historian Eric Arnesen has documented). The ARU executive committee, recognizing the operational collapse, met with leaders of the Chicago Federation of Labor on the 13th and proposed a settlement involving amnesty for strikers and reinstatement of fired ARU members. The General Managers’ Association rejected the proposal. There was no settlement. The ARU was effectively destroyed as a national organization within months of the boycott’s collapse. Debs was arrested four days later and subsequently imprisoned. The Palace Car Company itself made no concessions on wages, rent, or workplace conditions, though the company town arrangement was eventually dismantled by court order in 1898 on the ground that holding real estate not necessary to its business was beyond the Pullman Palace Car Company’s corporate charter.

Q: What was the Supreme Court’s ruling in In re Debs?

The Supreme Court decided In re Debs on May 27, 1895, unanimously affirming the contempt conviction. Justice David J. Brewer wrote the opinion. The ruling did not rest on the Sherman Act theory the federal prosecutors had pleaded; instead it relied on a broader theory of inherent federal sovereign power to protect the free flow of interstate commerce and the operation of the mails, supplemented by the U.S. circuit courts’ equity jurisdiction to enjoin obstructions to those national interest. The opinion affirmed three doctrines that would govern federal labor injunctions for nearly forty years: federal courts could enjoin union activity affecting interstate commerce, such injunctions could be enforced through contempt proceedings rather than jury trials, and federal executive force could be deployed to give effect to such orders. The Norris-LaGuardia Act of 1932 restricted the first doctrine; the National Labor Relations Act of 1935 substantially altered the broader framework. In re Debs nevertheless remains a foundational precedent for federal commerce-clause and executive-power doctrine.

Q: Did Cleveland’s handling of the Pullman Strike cost him politically?

The political cost was severe and historically consequential. Cleveland’s own party repudiated his wing at the 1896 Democratic National Convention in Chicago. The platform expressly denounced “government by injunction,” referring specifically to the Pullman case. William Jennings Bryan was nominated on a platform Cleveland could not endorse, and Cleveland in November voted for the third-party Gold Democrat candidate John M. Palmer rather than for Bryan. The Democratic Party began a long realignment that would eventually move organized workers into a Democratic coalition Cleveland’s conduct had effectively severed from his wing of the party. Cleveland left office in March 1897 with his political base destroyed and his reputation within his party broken. The historical irony is that within the Republican coalition his conduct strengthened (because Republicans broadly approved of the troop deployment), the political cost was negligible, but Cleveland was not a Republican, so the strengthening went to a coalition he did not belong to.

Q: How does the Pullman Strike compare to Theodore Roosevelt’s handling of the 1902 coal strike?

The contrast between Cleveland’s 1894 Pullman intervention and Theodore Roosevelt’s 1902 anthracite coal strike intervention is the cleanest available demonstration of how presidential conceptions of executive labor power can reverse within a single decade. Cleveland deployed regular army units to break a strike. Roosevelt threatened regular army units to compel arbitration between operators and union, then brokered the J.P. Morgan-coordinated October 13, 1902 agreement that created the Anthracite Coal Strike Commission and produced a 10 percent wage increase and nine-hour day for the miners. The Cleveland model treated the executive as a strike-breaker on behalf of capital; the Roosevelt model treated the executive as a labor broker with legitimate interests in industrial peace independent of either party’s legal position. Roosevelt’s authority to seize the mines was bluffed, not real, but the bluff succeeded because the precedent Cleveland had established made the threat plausible. The reversal of federal union policy that Article 32 of this series, “TR and the Coal Strike: The 1902 Mediation Precedent,” traces in detail begins with Roosevelt’s deliberate inversion of what Cleveland had done.

Q: What was the Pullman company town like for workers?

The town of Pullman, fifteen miles south of Chicago on roughly four thousand acres assembled between 1880 and 1884, housed approximately twelve thousand residents by 1893, almost all of them employees of the Pullman Palace Car Company or their family members. The company’s founder owned every house, every street, every gas line, every water main, every shop, the school, the library, the church, the hotel, and the theater. Residents paid rent to the company. They bought groceries at company stores. They drew water from the company waterworks. The Pullman bookkeeping system deducted rent and utilities from wages before workers received any cash; some workers received pay envelopes containing literally a few cents after deductions, and the Strike Commission’s testimony included one envelope of seven cents for two weeks of work. Richard T. Ely’s 1885 Harper’s essay had described the arrangement as un-American in its concentration of authority over the working lives of its inhabitants. The Strike Commission’s 1894 report endorsed this assessment. The town was dismantled as a company-owned operation by Illinois state court order in 1898.

Q: Why did the American Railway Union fail despite its rapid early growth?

The ARU’s failure is best understood as the convergence of three factors. The first was the federal architecture Olney built: the injunction, the troop deployment, and the In re Debs affirmation collectively destroyed the legal and operational basis for sympathetic boycotts on the scale the ARU had attempted. The second was the strategic decision by the union’s June 1894 convention to exclude African-American workers from membership, which deprived the boycott of solidarity with the African-American rail workforce and made replacement labor available to the General Managers’ Association at scale. The historian Eric Arnesen has documented the cost of this exclusion in his work on Black railroad workers. The third was the institutional weakness of the ARU itself, less than two years old at the time of the strike and lacking the organizational depth that older craft brotherhoods had developed across decades. Debs’s own pre-strike correspondence shows he understood this weakness; he counseled against precipitate action by the Pullman local in May 1894, and was overruled by events. The combination of federal repression, strategic exclusion of potential allies, and institutional immaturity produced a defeat from which the ARU did not recover. Industrial unionism in the rail sector would not be effectively re-established until the 1930s.

Q: Did the Pullman Strike set a precedent that influenced later federal labor interventions?

Yes, the precedent chain is one of the cleanest in American labor and constitutional history. Wilson during World War I drew on the Pullman framework in the 1917 nationalization of the railroads and in the use of federal injunctions during the postwar labor strife. Truman invoked the framework in the 1946 rail seizure and again in the 1952 steel seizure, the latter struck down by the Supreme Court in Youngstown Sheet and Tube v. Sawyer. Reagan’s 1981 PATCO response operated within a related but distinct framework grounded in the same broad theory of executive authority over labor in essential industries. The Norris-LaGuardia Act of 1932 restricted the federal courts’ equity jurisdiction over labor disputes, partially reversing the In re Debs doctrine. The National Labor Relations Act of 1935 established the modern collective bargaining framework. But the core 1894 principles, national authority to enjoin labor activity affecting interstate commerce and federal use of force to give effect to such orders, remain available to the modern presidency in modified form.

Q: How did the Pullman Strike affect federal labor mediation policy?

The Strike Commission’s November 1894 report explicitly identified the absence of national mediation machinery as a structural failure that had left the country without a peaceful means of resolving major railroad labor disputes. The recommendation produced the Erdman Act of 1898, which established federal mediation for railroad labor disputes through the Interstate Commerce Commission. The Newlands Act of 1913 expanded the Erdman framework. The Railway Labor Act of 1926 institutionalized the structure that has since governed railroad labor relations and was extended to airline labor in 1936. The mediation framework that the Roosevelt administration would draw on in 1902 to broker the anthracite settlement was the direct legislative legacy of the Strike Commission’s diagnosis of what Cleveland’s intervention had failed to accomplish. The Erdman Act is, in this sense, Cleveland’s posthumous correction by his own commission, produced four years after he had left office.

Q: What is the historian consensus on Cleveland’s Pullman decision today?

The historiographical consensus has shifted substantially over the past century. The Nevins reading (Cleveland as principled defender of the rule of law) dominated through the 1950s and remains influential, particularly in the work of Henry F. Graff and Robert Kelley. The Lindsey-Papke reading (Cleveland as instrument, knowing or unknowing, of corporate strike-breaking) emerged in mid-twentieth-century labor historiography and now dominates the specialist literature. H. W. Brands’s middle position (Cleveland as honest in intention but substantively wrong in his uncritical acceptance of Olney’s legal architecture) is closer to the current scholarly center. The disagreement is not primarily about facts, which are largely shared, but about how to weight the rule-of-law framing against the corporate-state-alignment framing. The recent specialist work (Papke 1999, Forbath 1991, McCartin’s later work on twentieth-century labor cases) has moved the field toward weighting the corporate-state alignment more heavily. The general-readership biographical tradition (Graff 2002, Allan Nevins legacy) continues to weight the rule-of-law framing more heavily. Both readings remain defensible on the evidence.

Q: How does the Pullman Strike fit the series thesis that the modern presidency was forged in four crises?

The series argues that the modern presidency was built primarily in four crises (Civil War, Great Depression, World War II, Cold War) and that emergency powers created in those crises outlived the emergency. The Pullman case is not one of those four crises but is a precursor to all of them. It established, at low stakes by the standards of total war, that the federal executive could act unilaterally inside a state against the protest of that state’s elected officers in defense of a national interest the executive itself defined. This authority was available to Wilson in 1917, FDR in 1941, Truman in 1950, and every modern president since. The Pullman case is a small precedent for what later became the large pattern. Cleveland could not have foreseen the use to which his 1894 framework would be put, but he built it, and the architecture he and Olney constructed has been bent, restricted, and partly reversed by twentieth-century labor legislation while retaining its core principles available to modern executive action. The case is in this sense more important to the long arc of presidential power than its immediate stakes would suggest.

Q: Did Debs ever address what Cleveland did to him?

Debs addressed the Pullman case throughout the rest of his life in speeches, articles, and his 1908 campaign autobiography. His verdict was unsparing. He treated Cleveland as the agent of capital, the federal courts as instruments of class warfare, and the In re Debs ruling as the constitutional foundation of what he called “the corporation state.” His most famous statement of the case, delivered as a campaign address in 1908, named Olney, Walker, Grosscup, Woods, and Brewer as the architects of his imprisonment and treated Cleveland as their principal. The radicalization that Debs’s six months at Woodstock had produced was permanent. He ran for president five times as a socialist (1900, 1904, 1908, 1912, 1920), receiving 901,000 votes in 1920 while imprisoned for opposition to U.S. entry into World War I under the Espionage Act, a statute that drew on the same broad theory of federal executive authority over speech and association that the Pullman injunction had pioneered. The connection between Cleveland’s 1894 decision and the U.S. prosecution of Debs in 1918 ran through a continuous institutional architecture. Debs understood this clearly; the federal authorities who imprisoned him in both 1895 and 1918 were drawing on the same constitutional doctrine he had been imprisoned to test.

Q: What primary sources document the Pullman case in detail?

The richest primary documentation includes the omnibus injunction itself of July 2, 1894 (issued by Judges Peter S. Grosscup and William A. Woods, available in the federal court records of the Northern District of Illinois); Cleveland’s July 8, 1894 proclamation; the Altgeld-Cleveland telegram exchange of July 5 through 6, 1894 (preserved in the Cleveland Papers at the Library of Congress and the Altgeld Papers at the Illinois State Historical Library); the U.S. Strike Commission Report of November 1894, running to 681 pages with appendices; the Supreme Court’s ruling in In re Debs (158 U.S. 564, 1895); the Olney-Walker correspondence of June 1894 in the Olney Papers at the Library of Congress; and the General Managers’ Association minutes preserved in the Newberry Library. Cleveland’s own retrospective chapter in “Presidential Problems” (1904) is essential for understanding his self-assessment. The contemporary press coverage in the Chicago Tribune, Chicago Times, Chicago Inter-Ocean, and the New York papers provides the day-by-day operational record. The combination of these sources allows for a near-hour-level reconstruction of events between June 26 and July 17, 1894.

Q: How does the Pullman precedent compare to Reagan’s PATCO response in 1981?

The two cases share a broad theory of executive authority over labor in essential industries while differing significantly in their operational design. In 1894, Cleveland used regular army units and a federal injunction to break a strike against a private corporate combination. In 1981, Reagan used administrative dismissal (firing 11,345 federal air traffic controllers) and a lifetime ban on rehiring to break a strike against the federal government itself, which was the controllers’ employer. The Cleveland model was triangular (executive, private capital, organized workers) and required complex legal architecture to fuse executive authority with private corporate interest. The Reagan model was bilateral (executive as employer, organized workers) and required only the enforcement of existing federal labor law prohibiting strikes by U.S. employees. The institutional consequences nevertheless followed parallel paths. Both cases signaled to the broader labor movement that national authority would be used against organized workers where it conflicted with executive priorities. The decline in U.S. private-sector strike activity after both 1894 and 1981 followed within months of the national action. Article 59 of this series, “Reagan fires air controllers 1981,” examines the PATCO case in detail and traces the precedent line that runs from 1894 through 1981.

Q: Was the deployment of regular army units actually necessary to restore order in Chicago?

On the evidence the Strike Commission compiled, no. Property damage and deaths attributable to the strike were limited before regular army troops first appeared in Chicago on July 4, 1894. The major escalation in violence (the July 6 Panhandle yard fires that destroyed approximately seven hundred rail cars, the deaths climbing to at least thirteen by July 10, the wider disorder across the South Side rail districts) followed the troop deployment rather than preceding it. The deployment did not respond to a violent strike; it produced one. The argument that troops were necessary to restore order is therefore difficult to sustain on the evidence the federal government’s own commission compiled. Nevins’s defense addresses this point indirectly by arguing that even if the deployment caused short-term escalation, it was necessary in the longer arc to restore the operation of federal functions. That is a defensible policy argument. It is not a defense of the framing that the troops responded to violence, since the major violence followed the troops. The framing that has dominated popular memory (regular army units sent to restore order in a city overrun by violent strikers) inverts the actual sequence the Commission documented.

Q: What is the most important lesson of the Pullman Strike for understanding presidential power?

The most important lesson is that the operational design of executive action matters as much as the constitutional theory under which the action is taken. Cleveland’s constitutional theory (the Article II duty to enforce federal law applying to mail and interstate commerce) was defensible and, in some form, available to any president. The operational design Olney built around that theory (the dual-role appointment of the railroad combination’s counsel as federal prosecutor, the over-broad injunction including “persuasion” as a prohibited means, the Sherman Act stretching to a union, the unilateral troop deployment without state coordination) was contingent on choices that other available choices would have averted. The corporate-state alignment was not in the constitutional theory; it was in the operational design. A president taking constitutionally defensible action can nevertheless build, through the design choices subordinates make on his behalf, an apparatus whose substantive effects diverge sharply from the formal constitutional justification. The president did not ask the hard questions about Olney’s design that afternoon. The hard questions about presidential design are the ones the modern presidency has continued not to ask, decision after decision, and the Pullman case is a useful exemplar of how that pattern produces consequences the president himself does not see.