“I Would Advise That It Be Kept from the Schools at All Cost”: The Influenza of 1918–1920 at Nonreservation Indian Boarding Schools

Introduction

On 25 October 1918, Superintendent Charles T. Coggeshall of the Fort Bidwell Indian School in California wrote an alarmed telegram to Commissioner of Indian Affairs Cato Sells. The state board of health had recently ordered all schools in Modoc County closed due to a deadly wave of influenza that was then sweeping the country. Coggeshall was taking “all precautions possible” to prevent infection among the Indigenous students under his supervision, but he warned the commissioner that he had “no adequate hospital or segregation facilities” to manage the disease if it entered the school. He wondered if he should follow state guidelines and send the children home.¹ Two days later, Sells responded and directed Coggeshall not to close the school but instead to employ emergency medical help as needed. Sells believed that Indigenous children could be “better cared for in school than at their homes,” and thus should not be released from school during the influenza crisis.² Coggeshall dutifully complied with the commissioner’s order. Within ten days, all seventy-eight of his students as well as six employees and two nurses were down with the virus.³

As the exchange between Coggeshall and Sells demonstrates, it was the Indian Office’s policy to keep its boarding schools open during the global influenza pandemic of 1918–20 despite the fact that many public schools across the United States closed at least temporarily to stem the spread of the virus. This was true even when Indigenous parents specifically requested the return of their children.⁴ Faced with a new and deadly public health emergency, officials insisted that they knew what was best for Indigenous children. Yet the reality was more complicated. Although the individual decisions of superintendents led to a range of outcomes at the schools, overall Indigenous children at the nonreservation boarding schools experienced high rates of infection and correspondingly high mortality rates. Despite these results, school superintendents and Indian Office officials downplayed the health crisis at the boarding schools, choosing instead to focus on stories of success.

Scholars of Indigenous America have long recognized Indian boarding schools as a fruitful terrain for analysis. Since the 1980s, numerous researchers have focused their attention on these federally run institutions by describing both the policies and systems that structured them as well as the lived experiences of the Indigenous students who attended them (see McBeth 1983; Trennart 1988; Hyer 1990; Lomawaima 1994; Adams 1995; Ellis 1996; Child 1998; Cobb 2000; Trafzer, Keller, and Sisquoc 2006; Vučković 2008; Gilbert 2010; Bahr 2014; Gram 2015; Woolford 2015; Landrum 2019). Although many scholars of Indian boarding schools discuss the health conditions at these institutions, few have made disease outbreaks and mitigation efforts a focus of their work. This lacuna is somewhat surprising given that matters of health were of central concern to boarding school administrators and federal officials in the early twentieth century (Keller 2002: xvi–xvii). Those works that do focus on student health include Diane T. Putney’s (1980) unpublished dissertation on Indian morbidity and mortality and federal policy, Jean A. Keller’s (2002) monograph on the Sherman Institute in California, David H. DeJong’s (2007) article on federal efforts to improve health outcomes at the boarding schools, and Mikaëla M. Adams’s (2020, 2022) previous two articles on the influenza outbreaks at the Haskell Institute and the Pierre Indian School. Although these last two works explicitly focus on the influenza of 1918–20 in the boarding schools, they examine the outbreaks at just one institution each and thus do not provide comparative analysis of how different schools approached the pandemic.

In this article, we investigate the 1918–20 influenza pandemic in all of the federal government’s twenty-five nonreservation Indian boarding schools, which served approximately 6,200 Indigenous youths at the time.⁵ Our article makes two main contributions. First, it uses underutilized manuscript sources from the Bureau of Indian Affairs’ Central Classified Files and Annual Narrative and Statistic Reports to document in detail how boarding school administrators managed this national health crisis at their various institutions. This is a story not previously told. By bringing these sources to light and comparing the various approaches that school administrators took to this disease outbreak, we add to scholarly understandings of both the Indian boarding school system and the influenza pandemic of 1918–20.⁶ Second, the interdisciplinary expertise of the authors allows us to provide both qualitative and quantitative analysis of how the pandemic played out at these schools. By examining why and how boarding school administrators made the choices they did during influenza outbreaks and then statistically analyzing the outcomes, we provide valuable insight into how the on-the-ground decisions of federal officials affected Indigenous lives. In the following sections, we first describe the Indigenous education system at the time, then document and evaluate patterns of infection, mitigation efforts, and medical aid as well as responses by the superintendents and Indian Office to the overall experience. Drawing from this discussion, we use quantitative analysis to investigate the roles of different factors in morbidity and mortality outcomes.

Indigenous Education in Early Twentieth-Century America

Following US military victories over Native peoples in the late nineteenth-century Indian wars, reformers and federal officials embarked on a new mission to address the nation’s so-called Indian Problem through education. The Indian Office, which had previously been a fairly small and insignificant agency, expanded tremendously in this period to take on this task; hundreds of newly hired employees marched into Indian Country determined to “civilize” its people and bring them into the American fold (Cahill 2013: 2–3). White Americans believed that the education of Indigenous children, in particular, would speed up the assimilation process and ultimately transform them from dependent tribal members into self-sufficient American citizens (Hoxie 2001: 189–91; Cahill 2013: 53–55). The federal government’s first federally operated boarding school, the Carlisle Indian Industrial School in Pennsylvania, opened in 1879. By 1919, another 24 nonreservation boarding schools, 80 reservation boarding schools, and 193 day schools existed across the continental United States.⁷ Altogether, these various government schools served around twenty-nine thousand of the ninety thousand school-age Indigenous children under federal supervision. Another six thousand Indigenous youths received training in mission and private schools; thirty thousand more attended state-run public schools.⁸ As Commissioner Sells explained to Congress in 1919, education had become the “primary reason” for the ongoing existence of the Indian Office as a federal bureau.⁹

Although the Indian Office determined the curriculum, set the regulations, and scheduled intermittent inspections of the boarding schools, in practice school superintendents had considerable power over the day-to-day implementation of federal policies on the ground. It was the superintendents who recruited and enrolled students, oversaw the school staff—including principals, teachers, disciplinarians, matrons, cooks, clerks, and medical personnel—and managed school finances (Cahill 2013: 56–58). Superintendents frequently corresponded with the commissioner of Indian Affairs for direction and advice, but ultimately they had to make do with the resources at hand, manage their own staff and students, and negotiate relationships with various stakeholders, including Indigenous parents and local white citizens. As a result, the boarding schools were not a monolith—much depended on the competence and efficiency of each school’s particular superintendent. Differences between the schools became especially apparent during moments of crisis.

One of the greatest challenges boarding school superintendents faced was lack of resources. Indeed, chronic underfunding from Congress severely hindered federal aspirations for the education of Indigenous children. By the late nineteenth century, lawmakers who embraced eugenic thinking and scientific racism began questioning the value of devoting significant federal funds to the Indian education project (Hoxie 2001: 191–92). Without appropriate resources, Indian boarding schools often deteriorated into run-down and unsanitary facilities, which resulted in numerous student fatalities from infectious diseases. Although the Indian Office made an effort to improve student health in the 1910s following several investigations that revealed the extent of morbidity and mortality rates at the schools, the diversion of federal funds to World War I made this task challenging (DeJong 2007). Just one year after the United States entered the war in 1917, a deadly new strain of influenza struck the nation.

The First Wave of the Pandemic, Spring 1918

US law required physicians to report outbreaks of a number of infectious diseases to public health authorities in the early twentieth century, but officials did not consider influenza serious enough to warrant special attention. Influenza’s status as a nonreportable disease during the early months of 1918 makes tracing the first wave of the pandemic difficult (Crosby 1989: 18).¹⁰ The wave was widespread in the United States but generally mild in impact; furthermore, it overlapped in time with normal seasonal influenza and could be difficult to distinguish from it (Crosby 1989: 17; Sattenspiel et al. 2023). The best definitive evidence for early outbreaks comes from institutions that kept formal health records, such a prisons, army camps, and—pointedly for our purposes—federal Indian boarding schools.

At least three of the twenty-five nonreservation Indian boarding schools that existed at the start of the pandemic reported significant influenza epidemics during the spring of 1918, with another three mentioning more contained outbreaks. By far the most serious of these epidemics occurred at the Haskell Institute in Kansas. Superintendent H. B. Peairs later wrote that the disease “struck us . . . like a thunder bolt [sic] out of a clear sky.”¹¹ The first case appeared at the school on 15 March 1918. Within just two weeks, the virus had “seized half the pupils with more or less severeness.”¹² Students complained of sore throats, headaches, and exhaustion. Some experienced nausea, vomiting, and chills. Temperatures ranged between 101 and 106 degrees Fahrenheit.¹³ Although the school physician recognized the outbreak as “grippe,” it was not like any flu he had seen before. On 25 March, the first student succumbed to the virus.¹⁴ Over the next week, another five students died in the school hospital; at least one other student perished after returning home.¹⁵ Desperate for help, Superintendent Peairs called in members of the US Public Health Service (USPHS) to investigate.¹⁶ Following his visit to the school, USPHS Senior Surgeon Charles E. Banks deemed the outbreak serious enough to report to his agency’s April 1918 Public Health Bulletin.¹⁷ The epidemic also appeared in the surgeon general’s 1917–18 annual report.¹⁸

The other two nonreservation schools that experienced significant outbreaks were the Flandreau Indian School in South Dakota and the Mount Pleasant Indian School in Michigan. Flandreau superintendent Charles F. Pierce reported two major outbreaks among his pupils during the 1917–18 academic year: a mild measles epidemic and “another of grip which was more serious than the measles.” “Grip”—influenza—attacked “about half of the employe [sic] force and student body.” Although the epidemic did not last long, it “so lowered the vitality of several pupils that it was deemed advisable that they be sent home.” Pierce did not record what became of these students.¹⁹ Mount Pleasant experienced similar outbreaks of both measles and influenza during the 1917–18 academic year. Measles came first, attacking sixty-seven students and leading to tubercular flare-ups among several of the afflicted, whom the superintendent promptly sent home. Influenza followed not long after and was “rather severe in many instances.” Both pupils and employees fell ill, which overtaxed the school hospital to the point that student dormitories had to be repurposed as convalescent wards. Ultimately, there were 105 cases of influenza at the Mount Pleasant Indian School that spring, but all recovered.²⁰

Smaller, relatively mild outbreaks of influenza also occurred at the Springfield Indian School in South Dakota, the Santa Fe School in New Mexico, and the Carlisle Indian Industrial School in Pennsylvania. At Springfield, students came down with both “grip and ‘Liberty’ measles [rubella]” during the spring of 1918, “but none were seriously sick.”²¹ The Santa Fe School reported “a half dozen cases of pneumonia” in April 1918, one of which “proved fatal,” although it is unclear whether influenza was the underlying culprit for the infections.²² Carlisle also experienced a “mild form of grippe” in mid-April, which sent “twelve or fifteen students” to the school hospital. The outbreak abated by early May, and all of the students recovered.²³

The Second Wave of the Pandemic, Fall 1918

By far the most serious wave of the 1918–20 influenza pandemic occurred in the fall of 1918. After seemingly going dormant in the United States over the summer of 1918, the virus returned in full force in late August, likely transported from Europe via military convoys (Barry 2004a: 183). Influenza swiftly made its way across the country, taking advantage of new transportation routes and wartime population movements to spread from coast to coast.

The first outbreak at a nonreservation boarding school that fall occurred at the Rapid City Indian School in South Dakota. The virus reached the school at the beginning of the fall semester, apparently traveling in the lungs of the first returning students. Before school employees understood the danger, some 150 students had arrived at the school. The first symptomatic case developed on 22 September 1918. From 2 to 4 October, another one hundred pupils fell sick. Within just a few more days, 131 pupils and 16 employees were down with influenza.²⁴ Among the sick was Superintendent Jesse F. House. As soon as he was well enough to resume his duties, House wrote an impassioned letter to the Indian Office describing the outbreak. “It is the most terrible experience I have ever had,” he lamented, “so many sick and so very sick, and so difficult to give the care necessary.” He advised the Indian Office to do its utmost to keep the disease from other schools.²⁵

After receiving various reports about influenza, the Indian Office sent out a telegram on 11 October 1918 to warn superintendents about the danger posed by the new disease.²⁶ Unfortunately, by the time this telegram went out, influenza had already infiltrated many of the nonreservation boarding schools.²⁷ At least nine of the twenty-four nonreservation Indian boarding schools that existed that fall experienced outbreaks during the first two weeks of October. By mid to late October and early November, influenza had attacked another ten of these schools. Just five nonreservation boarding schools escaped influenza during the fall 1918 wave of the pandemic.²⁸

Influenza entered the schools in a variety of ways. In some cases, school employees introduced the virus. At the Flandreau Indian School, the superintendent and the assistant clerk were the first to fall ill on 12 and 14 October, respectively. By the following week, “pupils commenced to be effected [sic] very rapidly and within a short time [the] school hospital was full to overflowing.”²⁹ School visitors may also have inadvertently carried the disease to these institutions. In particular, former students who had enlisted in the army frequently visited friends at the boarding schools while on furlough. Superintendents encouraged such visits as a way to instill a “greater spirit of patriotism” among their Indigenous charges, but given the prevalence of influenza in army training camps, these visits may have brought unintended consequences.³⁰ Another pathway of infection came from the students themselves as they traveled back to school for the beginning of the fall term. The source of the outbreak at the Sherman Institute in California was likely “a party of students who evidently came in contact with the disease on the train.”³¹ Similarly, the infection of students at the Pierre Indian School in South Dakota “occurred on the train while enroute [sic] to Pierre.”³² Influenza arrived at the Phoenix Indian School in Arizona, meanwhile, by way of two student “deserters, who spent one day in a cotton camp about twenty miles from Phoenix, where Papago Indians were ill with this disease.” The boys returned to school and “mingled with the student body before they were known to be on the premises.”³³

Experience with the spring 1918 wave of the pandemic may have offered certain schools some immunological protection against the second wave of the pandemic. The Flandreau Indian School experienced a renewed outbreak of influenza in October 1918, during which only 130 pupils out of 231 fell sick.³⁴ This morbidity rate of 56.27 percent was less than that of many of the nonreservation boarding schools that had not experienced a spring outbreak, several of which experienced a fall morbidity rate of 100 percent. Superintendent Charles F. Pierce noted the protective effect of prior influenza exposure at Flandreau: he reported that of the 125 students who had contracted influenza by 19 November 1918, 111 were newly enrolled that fall, leaving just “14 old pupils who took the disease.” Of these fourteen, “only 2 were sick last spring” and there was “some doubt as to whether they had the grippe or influenza at that time.”³⁵

These protective effects were mitigated, however, by the enrollment of new, immunologically naive pupils, including the transfer of significant numbers of students from the Carlisle Indian Industrial School in Pennsylvania to other boarding schools. Carlisle closed its doors to Indigenous students during the summer of 1918 after the army commandeered the institution to serve as a hospital for wounded soldiers. Many of these students transferred to the Haskell Institute in Kansas, which was comparable to Carlisle both in its size and in the grade levels it served.³⁶ Due to the addition of these new students, attendance at Haskell rose from 750 to 830.³⁷ Although Carlisle had experienced a mild outbreak of influenza during the spring of 1918, reportedly only twelve to fifteen students had fallen ill. Other Carlisle students may have been exposed to the virus without developing clinical cases, but many likely remained immunologically defenseless against the new strain of influenza.³⁸ With the addition of these new students as well as other new enrollees, Haskell experienced another serious outbreak of influenza, which began in early October and lasted through November 1918. Reportedly, 343 students fell sick, and nine died at the school, with at least one more pupil dying at home after he fled the school during the height of the epidemic.³⁹

In addition to offering some—if limited—immunological protection, experience with the spring 1918 wave of the pandemic may also have influenced the behavior of at least one superintendent when it came to making quarantine decisions during the fall 1918 wave of the pandemic. Superintendent R. A. Cochran of the Mount Pleasant Indian School implemented an extraordinarily strict quarantine at his institution as soon as he heard about the return of influenza that fall. According to his correspondence with the Indian Office, he put the school under quarantine as soon as “the disease appeared in Michigan.” He prohibited visitors to the school, including the parents of the students.⁴⁰ He also banned pupils from going to the nearby town of Mount Pleasant. Only he and one or two other employees made occasional excursions off school grounds to send and receive mail.⁴¹ Although Cochran worried that parents might protest his decision to ban visits during the Christmas holidays, he preempted such complaints by asking for—and receiving—support for his efforts from the Indian Office.⁴² Cochran did not lift the quarantine until 1 February 1919, at which point the nearby town of Mount Pleasant finally seemed “clear of the disease.” Because of these stringent measures, the Mount Pleasant Indian School avoided the fall 1918 and spring 1919 waves of the pandemic even though “the surrounding country had quite an epidemic.”⁴³

Most of the nonreservation boarding schools, however, did not benefit from the protective effects of prior experience. Once influenza entered the schools during the fall of 1918, the virus spread rapidly and infected large portions of the student body as well as school employees. The Sherman Institute in California had 622 cases, 440 of which were confined to their beds at the same time.⁴⁴ More than 75 percent of the students at the Salem Indian School in Oregon contracted influenza between 1 October and 3 November 1918. Sixteen minors died at the school, along with three adults.⁴⁵ The Hayward Indian School in Wisconsin, meanwhile, reported 200 cases—the entirety of the student body; nine of these were fatal.⁴⁶

Many influenza cases developed into pneumonia, which proved especially deadly. As Sherman superintendent F. M. Conser recounted, “the influenza in itself does not seem to be so serious, but the complications that follow have proven quite serious.” At his school, fifty-two of the afflicted children developed pneumonia and eight died between 10 October and 22 October 1918.⁴⁷ The superintendent of the Hayward Indian School explained “In nearly all cases of death, Influenza was followed by Pneumonia and there appeared to be no possibility to save the patients.”⁴⁸ Ten students at his institution developed pneumonia following influenza, and seven of those died by 22 November 1918.⁴⁹ Another two perished in the subsequent weeks. At the Santa Fe School in New Mexico, pneumonia killed two students. The first boy developed influenza symptoms two days after his arrival at the school. This, “together with the fact of probable exposure on the drive of some sixty miles to get to the school, brought on an attack of pneumonia,” which proved fatal on 20 October 1918.⁵⁰ The second, a nine-year-old boy, contracted influenza at the school in mid-November 1918. Although he survived his initial infection, he developed pneumonia and died on 15 January 1919.⁵¹ One hundred thirty-one cases of influenza developed at the Phoenix Indian School in November 1918, and thirteen of these progressed to “broncho pneumonia.” Two of these patients died.⁵²

Influenza also left patients weak after recovery, which necessitated ongoing care even after the worst of the outbreak passed. As Greenville superintendent Edgar K. Miller relayed to the Indian Office in late October 1918, “Our hospital will be full for sometime [sic] yet, for the disease often leaves a child with something else that takes sometime to over-come or get rid of. This is one of the bad features of this disease. We have to be extremely careful with the children on this account.”⁵³ Similarly, Hayward superintendent Henry J. McQuigg reported that the epidemic at his institution was largely over by mid-December 1918, yet “many of the convalescents” remained “very weak.”⁵⁴ In some cases, students weakened by influenza then succumbed to other diseases. At the Cushman Trades School in Washington, six students died directly from influenza, while another two died shortly afterward “from scarlet fever in complication with the Spanish Influenza.”⁵⁵

Renewed Outbreaks and the Third Wave of the Pandemic, Winter–Spring 1919

The majority of the nonreservation boarding schools experienced outbreaks of influenza during the fall and early winter of 1918. For this reason, they generally avoided new infections during the early weeks of 1919 as well as the third wave of the pandemic, which started in March 1919, since much of their student body had already developed immunity to the virus. Quarantine efforts during the winter and spring may also have spared the schools additional outbreaks. Haskell superintendent H. B. Peairs, for example, instituted “a partial quarantine” on his school beginning in December 1918 and lasting through the spring of 1919. He suspended town leave and declared that no one would be “permitted to return home at any time during the holiday season, and also that no parents or friends from the students’ home [would be] permitted to visit.”⁵⁶ Although this ban was difficult for students and their families, Peairs believed it was the only way to prevent a renewed outbreak.⁵⁷

There were, however, exceptions. For the most part, new outbreaks that appeared during the early weeks of 1919 can be properly classified as second peaks of the second wave of the pandemic rather than true third wave outbreaks. Although both the Albuquerque Training School and the Cushman Indian School experienced influenza outbreaks in October 1918, the disease struck again in early 1919. At Albuquerque, the first outbreak was relatively mild, with just eight students and five employees falling sick in October. All recovered except for the dining room matron, who died in early December following a long fight with pneumonia. Influenza returned to the school on 20 January1919. This time, it spread rapidly among the students. There were 138 cases, but only three were severe. No student died at the school, although one boy perished after being sent to the sanatorium at Laguna.⁵⁸ In contrast, the Cushman Indian School experienced a severe outbreak in October 1918 that turned the school into “one big hospital” and killed five students. Although the worst of the epidemic passed by the end of the month, a few students remained “dangerously ill.”⁵⁹ Two more boys died in early November, and a third lingered until December 1918 before succumbing to “complications following influenza.”⁶⁰ A renewed outbreak in mid-February 1919 carried off an additional pupil who died of “nephritis and uremic poisoning, following an attack of grippe.”⁶¹ The Sherman Institute, meanwhile, experienced a long, drawn-out epidemic that started on 2 October 1918 and carried on through 8 January 1919.⁶²

The only nonreservation boarding school to experience a true “third wave” outbreak was the Carson Indian School in Nevada. Carson had successfully quarantined against influenza throughout the fall and winter of 1918–19, but “when it was thought that nearly all danger was past the disease started in the school, on March 14, 1919, and continued for several weeks.” The outbreak at Carson was severe. Two hundred eight children fell ill, which represented more than two-thirds of the student body. Five of these died, including the child of a white employee. Twenty-two school employees and their family members also contracted influenza, with three resulting fatalities. Among the dead were Superintendent James B. Royce and the school matron, which left a leadership vacuum at the school. The Indian Office sent Special Agent L. A. Dorrington to Carson to take charge of the school. Dorrington’s “untiring efforts” to organize the remaining staff and students for the care of the sick finally brought the epidemic under control.⁶³

The Fourth Wave of the Pandemic, Winter–Spring 1920

Influenza returned for a fourth wave during the late winter and early spring of 1920. Although many of the nonreservation boarding schools recorded renewed outbreaks of the disease in their annual reports, the Indian Office did not require superintendents to submit official morbidity and mortality statistics, which means that data for the fourth wave at these schools is incomplete. For the most part, the fourth wave of the pandemic seems to have been milder, whether due to a mutation of the virus or to the population’s increased immunity through prior exposure. Superintendents remarked on this change. The superintendent of the Wahpeton Indian School insisted that many of the influenza cases that appeared at his institution in late January 1920 “were so exceedingly mild that it [was] very doubtful if they were cases of influenza” at all. Nevertheless, to be safe he put “all pupils to bed as soon as they complained in any manner.” All recovered.⁶⁴ At the Flandreau Indian School, influenza also returned in “a very light form” with just a few cases and “no pneumonia.”⁶⁵ The Albuquerque Training School, meanwhile, experienced just six cases of influenza among students in 1920. A school employee whose daughter contracted the disease at a local public school apparently infected the children; all survived.⁶⁶

A few schools, however, did experience major outbreaks that winter. At the Mount Pleasant Indian School, 235 pupils fell sick with influenza in early February 1920, which represented over two-thirds of the student body.⁶⁷ L. L. Culp, a special physician assigned to the school, described the epidemic as “raging.” More than twenty students developed pneumonia and four died. In addition, a pregnant mother who came to the school to tend her sick child “developed flu, aborted, and died after about 30 hours.”⁶⁸ The outbreak at Mount Pleasant that winter may have been particularly severe because the school had avoided the fall 1918 and spring 1919 waves of the pandemic through a strict quarantine. Thus, it is likely that many of the students had not developed immunity to the virus through prior exposure.

The Chilocco Indian School in Oklahoma, which had avoided influenza in 1918 and 1919, also experienced a large outbreak in January 1920: 195 pupils and 35 employees, or their children, fell sick. “There were no very serious cases,” however, “and no deaths resulted.”⁶⁹ The Tomah Indian Industrial School likewise experienced an outbreak in 1920 after evading the disease in 1918 and 1919. Although around one hundred students contracted the virus, “it was in a very mild form, and all recovered without any serious complications.”⁷⁰

A few schools experienced renewed influenza outbreaks in 1920 despite having gone through epidemics in 1918–19. The Pipestone Indian School in Minnesota, which had weathered a relatively mild outbreak in 1918, was struck by “successive epidemics of influenza and measles which occurred and reoccurred throughout the [academic year of 1919–20] from October to May.” Although there were “no deaths directly of these diseases,” one child died of “tuberculosis induced by influenza.”⁷¹ The Haskell Institute also suffered a new attack of influenza in late January 1920. Although most of the 135 children who fell sick had mild cases, one student died.⁷² The Hayward Indian School experienced an outbreak in early February 1920, which infected between thirty and forty students. Two of these patients developed pneumonia with fatal results.⁷³ A particularly serious outbreak occurred at the Pierre Indian School, where “practically every pupil in the school and many of the employees were smitten with the disease.” Five students died in the school hospital, and another girl passed away shortly after returning to her family.⁷⁴

Mitigation Efforts: Quarantines, Isolation of the Sick, and Classroom Closures

There were attempts to mitigate the spread of influenza at the nonreservation boarding schools, although these in general did not include complete closure. Only the smallest of the schools—the Fallon Indian School in Nevada—took this approach during the height of the fall 1918 wave of the pandemic. Serving thirty-six Indigenous pupils and employing just one teacher, the school did not have the resources to care for sick students.⁷⁵ For this reason, Superintendent Harry M. Carter, who also oversaw the nearby Fallon Reservation, shut the school for a “three weeks vacation” while he dealt with an outbreak on the reservation that attacked “practically every Indian man, woman, and child.” The school reopened on 11 November 1918. Although some of the pupils sent home contracted influenza, “all except one” had recovered sufficiently to reenter school by 18 November.⁷⁶

Instead of closing the schools, many superintendents endeavored to protect pupils by quarantining their institutions against outsiders. One of the most successful quarantine efforts occurred at the Chilocco Indian School in Oklahoma, where “not a single case developed” during the fall of 1918 and spring of 1919 even though “the entire country was caught in the death grip of that awful scourge known as Spanish influenza.” Superintendent Oscar H. Lipps attributed this victory to the fact that he maintained a “strict quarantine” from October until February and gave “special attention” to the “proper exercise and entertainment” of his students while ensuring they received “wholesome food in ample quantity and variety.”⁷⁷ The Tomah Indian Industrial School in Wisconsin also maintained a rigorous quarantine that protected its students against infection during the fall of 1918 and spring of 1919. Superintendent Lindley M. Compton was inspired to take this step after one of the Indian Office’s traveling physicians, L. L. Culp, was called away from Tomah to take care of influenza at the Rapid City Indian School “and wrote back advising a very strict quarantine.”⁷⁸ The success of the Tomah School’s quarantine was due in large part to the cooperation of its Indigenous students. Compton remarked that his pupils “were very anxious to observe the regulations as they knew how much trouble other schools and agencies were having and they could also see how the influenza was affecting the neighborhood as they saw three or funerals go by the school grounds each day during the month of October and part of November.”⁷⁹ At both Chilocco and Tomah, seemingly positive relationships and open communication between school administrators and students appear to have contributed to the success of their quarantines.

Yet quarantines were only effective if they were strictly adhered to, and not all superintendents were so successful at fostering a cooperative spirit among students and staff. Superintendent E. H. Hammond of the Cushman Trades School, for example, supposedly put his institution under “a partial quarantine against Tacoma, Camp Lewis and Seattle” from 11 October 1918 to 15 March 1919, with but “one short interval,” yet his institution experienced successive waves of influenza, scarlet fever, chicken pox, and smallpox.⁸⁰ Hammond admitted in his annual report that the school employees had been “overworked” during this period “and were in no condition to undergo the additional nervous strain of protracted quarantine,” which suggests that he permitted at least some of them to leave school grounds.⁸¹ Superintendent John B. Brown of the Phoenix Indian School quarantined his institution against both the city of Phoenix and the adjoining reservation in mid-October 1918.⁸² Despite his decree, two pupils slipped out of the school later that month to visit relatives, which reveals the limits of Brown’s authority as well as perhaps his inability to impress the seriousness of the situation on his pupils. When the runaway boys returned to school, they brought influenza with them.⁸³ Other superintendents, like Charles T. Coggeshall of the Fort Bidwell School, tried to prevent infection by suspending the enrollment of new students but did not effectively isolate those already present at their schools.⁸⁴

Superintendents had a particularly difficult time keeping Indigenous parents away from the schools, especially when these parents feared for the safety of their children. Superintendent Peyton Carter of the Wahpeton Indian School complained that several parents from the White Earth Reservation “appeared to take their children home” during the outbreak at his institution. Although he advised them that they were not permitted to do so, some took their children anyway, “making the statement that they had public school facilities at home” and no longer needed the educational services of the Indian Office.⁸⁵ The actions of these parents reveal both the strength of their kinship ties as well as their rejection of federal officials’ right to make unilateral decisions about the well-being of their children.

Unfortunately, some of the parents who came to collect their children from Wahpeton contracted influenza while at the school. Carter reported that three adults, “parents or relatives of pupils or employees,” fell ill and two of these developed pneumonia. “All three adults were taken care of in the school hospital.”⁸⁶ Similarly, family members of a sick child at the Salem Indian Training School in Oregon traveled to the school in late October 1918 to see their son. The mother, father, and two younger children all contracted influenza during the visit. One of the small children died within days, and the mother was also “not expected to live.”⁸⁷ Some parents also traveled to the boarding schools to collect the bodies of children who perished during the pandemic. “The parents or other immediate relatives” of children who died at the Hayward Indian School came to the school “to take charge of the remains” even as influenza continued to rage among the rest of the student body.⁸⁸ For these Indigenous families, ensuring that their children received proper burial was worth the risk of infection.

In some cases, quarantine efforts came too late; by the time they were implemented, influenza had already infiltrated the school. The Salem Indian School maintained a month-long quarantine but only after students had already contracted the disease. Ultimately, there were over five hundred cases in the school, “resulting in the death of a number of students.”⁸⁹ The Springfield Indian School, meanwhile, only implemented a quarantine after four students had already fallen ill.⁹⁰ Such efforts nevertheless may have helped to prevent influenza from spreading to new enrollees. After the virus appeared at the Wahpeton Indian School during the second week of October 1918, Superintendent Carter advised reservation superintendents “not to send any more pupils” until the epidemic ran its course.⁹¹ Similarly, Superintendent Conser reported that the epidemic at the Sherman Institute “came at a time when there were a large number of students ready to enter school.” To spare these enrollees from infection, he notified “as many as possible not to come until the danger of contagion had passed.”⁹²

Once influenza entered the schools, some superintendents endeavored to limit its spread by separating the sick from the well and adopting other preventive measures. Flandreau superintendent Charles F. Pierce, for example, established sick wards in the school dormitories and kept uninfected students “out of doors as much as possible.” He also ordered that all school buildings be “sprayed every other day with disinfectant” and that every well pupil be “carefully watched so as to catch every case before it had reached a severe stage.”⁹³ Superintendent Mollie V. Gaither of the Springfield Indian School also isolated sick children “in a room by themselves” and requested that the local physician give “a talk to pupils and employees, telling them how to prevent and to take care of the disease.”⁹⁴ Employees at the Rapid City Indian School surrounded sick pupils with “sheets wet in formaldehyde” in an effort to limit the spread of the disease.⁹⁵ Superintendent Coggeshall of the Fort Bidwell School, meanwhile, ordered “masks made from gauze” for all employees and students after the first cases appeared at his institution. Although well intentioned, this effort did not ultimately prove successful since “it would appear that those sick had already been subjected to contagion” before the mask order took effect.⁹⁶

Superintendents sometimes also chose to suspend classroom work during the outbreaks, both as a way to limit the spread of the disease and so that employees could devote themselves to the care of the sick. At the Flandreau Indian School, Superintendent Pierce suspended academic work and detailed the teachers “to care for sick pupils and to keep others from being sick if possible” as soon as “it became evident that the disease was going to be general.”⁹⁷ Similarly, “every school activity was discontinued” during the outbreak at the Wahpeton Indian School so that “all the employees and students” well enough could tend the sick.⁹⁸ At the Fort Bidwell School, meanwhile, classroom work was suspended and “all other activities” were limited “to the absolutely necessary work” until “the pupils had made recovery and were again strong enough to attend School and take up the other details.”⁹⁹

Not all superintendents, however, were so rigorous in their prevention methods. Superintendent C. J. Crandall of the Pierre Indian School, for example, continued to enroll new students even as influenza tore through his institution. He admitted thirty-three new pupils between 5 October and 10 October, six of whom ended up dying at Pierre in the following weeks. Crandall only ceased this practice after receiving the Indian Office’s 11 October telegram that explicitly warned of the dangers of influenza.¹⁰⁰ Reportedly, Crandall also failed to separate the sick from the well at the school, which resulted in the infection of “practically every pupil there.”¹⁰¹ Superintendent William E. Thackrey of the Fort Mojave Indian School in Arizona, meanwhile, spent the better part of October 1918 traveling across southern Nevada, Utah, and northern Arizona “soliciting new pupils” despite the virus sweeping the country. His goal was to “exceed the attendance” of 150 at the school, which was the appropriation limit designated by Congress. When he returned to Fort Mojave in late October, he found “pupils and employees sick with influenza.” He seemingly did not make efforts to shield his new recruits from the disease. Although Thackrey temporarily succeeded “in raising the attendance a little above the 150 mark,” “death and other matters” lowered it again.¹⁰² Four students died at the school.¹⁰³ Concerned about how their enrollment numbers would impact school finances, these superintendents prioritized the well-being of their institutions over that of their students. Indigenous children paid the price.

The Albuquerque Training School had mixed results with its quarantine and isolation policies. When influenza first invaded the school in early October 1918, Superintendent Reuben Perry took immediate steps to enforce a “strict quarantine” against outsiders and to isolate each of the afflicted eight students and five employees. This effort succeeded in stemming the spread of the disease to the rest of the student body for several months. ¹⁰⁴ Over time, however, Perry grew less rigorous in his enforcement of the quarantine: he permitted the school band to participate in a Liberty Day parade, and he allowed students to go to town to shop “in the mornings just before Christmas.” These activities aligned with the federal government’s goal to turn Indigenous children into patriotic, Christian, consuming citizens. In late January 1919, twenty-five “small boys” at the school came down with influenza. Although Perry endeavored to isolate the cases, the virus made its way to the rest of the student body. In part, this result may have stemmed from Perry’s decision to “go ahead with class room work and other activities rather than to close the school work” during the epidemic. He reasoned that if he suspended schoolwork the children would “become restless” and would “gather in bunches on the grounds,” which he believed would facilitate transmission of the disease as much “as if we proceed in the usual way.”¹⁰⁵ This statement hints at the stress students experienced during the outbreak and their desire to find solace among friends, as well as the lengths to which superintendents went to maintain order. The Indian Office agreed with Perry’s assessment, although Assistant Commissioner E. B. Meritt advised him “to keep the pupils who [were] ill away from those who [were] well to such an extent as may be practicable.”¹⁰⁶ Ultimately, 138 students fell sick. Fortunately, only three cases were severe, and no pupils died at the school.¹⁰⁷

Medical Aid

In the early twentieth century, the best treatment for influenza was prevention, but when this failed, superintendents endeavored to provide sick students with medical aid. This was sometimes easier said than done since wartime conditions had depleted many boarding schools of their medical staff and other key employees. The Genoa Indian School in Nebraska, for example, had “neither nurse nor physician” in 1918 since the former ones had left for “war work.”¹⁰⁸ The Albuquerque Training School, which served nearly four hundred Indigenous students, also had “no regular nurse” in 1918; the position was “filled temporarily by a number of nurses during the year” but was “vacant for a good part of the time.”¹⁰⁹ The Cushman Trades School, meanwhile, lacked numerous key employees, such as hospital nurse, assistant matron, manual training teacher, teacher of ironwork, and principal teacher. The remaining force consisted mostly of “makeshifts” since Indian Service salaries could not compete with wartime wages in the nearby industrial city of Tacoma.¹¹⁰ The same was true for the Salem Indian School in Oregon, which experienced “the resignation of so many employes [sic] who left to accept higher salaries outside.”¹¹¹ Superintendent Miller of the Greenville Indian School protested in October 1918 that the “severest handicap” at his school was “the want of employees.” Since Greenville served some of “the youngest and frailest part” of the state’s Indigenous population, detailed supervision was especially critical. Miller even wondered if it would be better to abandon the school altogether and transfer students to other schools “until such a time the employee force could be kept regularly and we get a class of employees that could be interested in giving the students the close attention they so much need.”¹¹²

Despite deficiencies, superintendents relied on their employee force to care for the sick during the pandemic. Given the limited medical interventions available for influenza in 1918, such care mostly consisted of nursing work—keeping students comfortable, clean, hydrated, and fed (Crosby 1989: 7). To this end, schools made special purchases of aspirin, ice bags, and hot water bottles, as well as disinfectants like chloride of lime.¹¹³ Employees sometimes also prepared special diets for the afflicted students. At the Phoenix Indian School, sick pupils were “furnished special diet in the way of eggs, milk and citrus fruits.”¹¹⁴ Given the high prevalence of lactose intolerance in Native American populations, this regimen may not have been the most beneficial for afflicted students (Newcomer et al. 1977). The Haskell Institute converted three domestic science departments into kitchens where employees could prepare “suitable food” for convalescents.¹¹⁵ School physicians also experimented with various herbal remedies and patent medicines.¹¹⁶ Although these products could not cure influenza, they may have helped relieve some discomfort among the patients.

For the most part, superintendents were favorably impressed by the dedication of their labor force during the outbreaks. The Sherman Institute superintendent praised his “efficient and faithful corps of employees” who taxed themselves “to the utmost” to give students proper care during the fall 1918 wave. He also commended the “splendid assistance” given by students who helped care for their sick classmates, which reveals that Indigenous children also participated in this nursing work.¹¹⁷ The few employees at the Hayward Indian School well enough to work were up “early and late” and did “everything possible to have all details as carefully attended to as possible.”¹¹⁸ Special Physician L. L. Culp, who assisted with the fall 1918 outbreak at the Rapid City Indian School, remarked that “every employee” at the school “was worked to death, but was whole heartedly interested and trying to exert every ounce of energy for the welfare of the sick.” Some relief also came from volunteer nurses from the town of Rapid City, who ensured that “no pupil was neglected in any way.”¹¹⁹

Sometimes, however, the strain of epidemic work proved too much for school employees. Cushman superintendent Hammond noted that “a few of the teachers were panic stricken and thoroughly demoralized” by the experience, which led to “some friction” among the employees. After he had a frank discussion with his work force and took “a firm stand . . . against gossip,” the situation improved.¹²⁰ At the Fort Mojave School, meanwhile, tensions reached a boiling point when the school clerk slapped the school cook in full view of a detail of students. This violent outburst must have come as quite a shock to Indigenous students who had been taught to look at school employees as role models for “civilized” behavior. The cook in turn consulted a lawyer and demanded that the clerk be arrested. When a sheriff arrived at the school to investigate, Superintendent Thackrey intervened by protesting that there was “then too much sickness to allow employees to waste any time over personal affairs.” “An arrest would have greatly interfered with the care for the sick,” he later explained to the Indian Office, “and it seemed best to stop the whole matter with a strong hand.”¹²¹ At the Pierre Indian School, meanwhile, one nurse grew so desperate with epidemic conditions that she fled the school in the middle of the night, leaving her patients to their fate.¹²²

Part of the stress of school employees came from the fact that they were not immune to influenza; many also fell ill during the school outbreaks and some succumbed to the disease. At the Wahpeton Indian School, “ten or eleven employees” contracted influenza at the same time that 112 pupils were down with the virus. Ultimately, the school principal died along with two of the students.¹²³ At the Fort Bidwell School, “many of the regular employees were taken sick with the ‘flu’ and they had to be looked after and cared for in addition to the pupils and Camp Indians.” The school disciplinarian had such a serious case that he remained in bed for more than a month.¹²⁴ Similarly, twenty-three employees at the Sherman Institute came down with influenza concurrently, which left them “not fit for work” at a time when more than 400 pupils were also sick.¹²⁵

In cases when the employee force was insufficient to manage an outbreak, securing outside aid was essential. Fortunately, the Indian Office entered an agreement with the US Public Health Service in October 1918 that provided funds for the hiring of additional medical personnel on reservations and in boarding schools during the influenza crisis. Special epidemic employees were paid out of US Treasury funds rather than from Indian Office appropriations. This financial support was especially critical since the Indian Office had already depleted its annual appropriation for Relieving Distress and Prevention of Disease among Indians to help forest fire sufferers on the Fond du Lac Reservation by the time influenza hit.¹²⁶ School superintendents eagerly marshaled these extra funds to hire doctors, nurses, and other caregivers whenever possible. The Fort Bidwell School, for example, employed eight practical nurses at a salary of $50 per month plus a $4 per diem allowance for a total of $595.¹²⁷ The superintendent of the Hayward Indian School and Hayward Reservation employed two doctors at $200 a month, two graduate nurses at $90 a month, and thirteen nurse assistants at $50 a month, all of whom also received a $4 per diem allowance.¹²⁸ The Fort Mojave Indian School also employed two graduate nurses and eight nurse assistants to help care for students during the epidemic.¹²⁹

Yet outside help was not always easy to find, especially when the surrounding population was also in the grips of an epidemic. Cushman superintendent Hammond informed the Indian Office that it had been “impossible to procure” special epidemic employees “owing to the prevalence of the Spanish Influenza in Tacoma, Seattle, and, in fact throughout the state during the time it was raging at the school.”¹³⁰ Superintendent Carter of the Wahpeton Indian School complained that during “the time when the epidemic was at its height” in mid to late October 1918, he “could procure no outside help of any sort whatever.” The virus was “so prevalent in the community and for hundreds of miles distant that every well person was needed to take care of those sick.”¹³¹ Similarly, the Sherman Institute found outside assistance “almost impossible to obtain.”¹³² Eventually the superintendent managed to hire one graduate nurse and one nurse assistant to supervise the care of more than four hundred sick students, but, for the most part, he had to rely on regular school employees and students to care for the sick.¹³³ Superintendent Miller of the Greenville Indian School could get but one woman to help at the school. He noted that he probably “could not have gotten her if she was not a relative of our school physician, who prevailed upon her to come in our time of sore need.”¹³⁴

Influenza Vaccines

In addition to implementing social isolation protocols and securing medical aid, some superintendents also experimented with newly developed vaccines to protect the health of their charges. Superintendent Perry of the Albuquerque Training School obtained a “mixed respirator vaccine” in late October 1918 and immunized both pupils and employees following a mild outbreak of influenza at his school.¹³⁵ Cushman superintendent Hammond likewise inoculated “all suspicious cases” of influenza at his school “to prevent pneumonia.”¹³⁶ Hayward superintendent McQuigg, meanwhile, reported in mid-December 1918 that a doctor named E. C. Rosenow of Rochester, Minnesota, had “furnished free of charge vaccine virus for influenza-pneumonia epidemic,” which McQuigg used to inoculate employees and students alike.¹³⁷ In early 1920, Haskell superintendent Peairs also used the “Rosenow influenza-pneumonia vaccine” to immunize approximately three hundred pupils who had not previously contracted influenza.¹³⁸ Superintendent Brown of the Phoenix Indian School, meanwhile, obtained “a mixed vaccine prepared by the Pathological Laboratory of Phoenix,” which he used “rather freely” among students and staff.¹³⁹

The Indian Office did not endorse or pay for these vaccination efforts, but neither did it prohibit them. As Assistant Commissioner Meritt explained, the “Office received many requests for influenza vaccine, but, upon the advice of the Surgeon General of the Public Health Service, it was not supplied to the Service because of the impossibility of obtaining a reliable product, early enough, in sufficient quantities to be effective.”¹⁴⁰ The consensus of the surgeon general of the US Public Health Service, the surgeon general of the US Army, the surgeon general of the US Navy, and the editor of the Journal of the American Medical Association was that “such a procedure [would] very likely prove useless.” The Indian Office did not object, however, if superintendents procured their own supply of vaccine. Indeed, the office asked superintendents to report on the outcome of such efforts.¹⁴¹ It is unclear whether superintendents obtained any sort of parental permission before administering these experimental serums to the children under their supervision.

We now know that the vaccines developed during the pandemic were not effective against influenza, in no small part because researchers at the time had yet to identify the true cause of the pandemic as a virus (Eyler 2010). Yet recent studies have suggested that at least some of the vaccines may have helped to prevent secondary bacterial infections, which could have lessened the severity of the outbreaks (Shanks 2018; Roth 2023). The superintendents who immunized their students spoke confidently of the efficacy of these vaccines. Flandreau superintendent Pierce insisted that the vaccine the school physician obtained from Boston and Rochester was “very successful” since “not a person was attacked after the second inoculation, and in fact, none of those inoculated had any sort of respiratory organ trouble during the remainder of the school year.”¹⁴² So pleased was he that he had the entire student body vaccinated with it again the following year.¹⁴³ Hayward superintendent McQuigg insisted, “It was proven very clearly at this school that such vaccination obtained good results.” Some of the pupils and employees who received the vaccine “did not take the disease at all,” and in others “the disease was taken but in a very light form.” McQuigg believed that “if the vaccine had been obtained a couple of weeks earlier that the benefits could have been yet more.”¹⁴⁴ Just one superintendent expressed doubt about the vaccines. Superintendent F. T. Mann of the Pipestone Indian School noted that the “entire pupil body was vaccinated” against influenza in early 1920. This effort “did not appear to prevent contagion,” although it did seem to lighten “the attack.”¹⁴⁵ In most cases, superintendents administered vaccines only after influenza had already infiltrated their institutions. It is therefore possible that the observed positive effects of the vaccines were simply due to the natural abating of those institutions’ outbreaks rather than the properties of the vaccines themselves.

Downplaying Mortality Rates

The high mortality rates experienced at many of the nonreservation boarding schools were upsetting to superintendents not only due to the lives lost but also because of the negative effects that these deaths might have on their reputation. Superintendents worried that these losses reflected poorly on their management skills as well as on the sanitary conditions of their institutions. Thus, they frequently endeavored to downplay influenza deaths by blaming students for their own poor health or for their behavior during their illness. Superintendent Pierce of the Flandreau Indian School, for example, insisted that extenuating circumstances explained the two deaths at his school during the second wave. The first fatality, a “large boy,” “had been excused from a reservation school for two years past, and had lived in a tent at home on account of threatened tuberculosis.” The boy had only attended the school for three weeks before his demise.¹⁴⁶ Meanwhile, the nineteen-year-old girl who died “was inclined to be nervous and hysterical.” When she learned of her mother’s death while sick with influenza, she experienced “hysterical and crying spells,” which the superintendent believed “contributed to deficient heart action and was partly responsible for her death.”¹⁴⁷ Pierce protested that “had the boy and girl been good strong physical specimens . . . they would have been saved,” in which event the school “would have lost no cases.”¹⁴⁸ Wahpeton superintendent Carter complained that one of the pupils who died at his institution had failed to reveal “the fact that he was feeling ill” until his case was already beyond help. His death, therefore, was “no reflection upon anyone of the employees.” The other pupil who died at Wahpeton “was peculiar in that she gave up immediately after being taken sick showing no disposition to help herself or fight the disease.” Had she fought as hard as had other sick students, he argued, she might have lived. ¹⁴⁹ Carson superintendent Frederic Snyder, meanwhile, opined that two of the four Indigenous pupils who died from influenza at his institution would have met early deaths anyway “as they were tubercular.”¹⁵⁰ Greenville superintendent Miller likewise argued that he would “not have lost” the student who died at his school “if his physical condition had been anywhere near good when he took the disease.”¹⁵¹ These comments reveal much about the abject preexisting health conditions of Indigenous children in the early twentieth century as well as the suffering and trauma that they experienced as they battled influenza far from kin and the comforts of home. For superintendents, though, the supposed physical and mental “deficiencies” of the students served to exculpate school staff for the deadly outcomes at their institutions.

Another way that superintendents diminished mortality rates was by sending sick students home or to other institutions. If a student died elsewhere, superintendents did not submit that death as part of their school’s official mortality count. While they might mention such deaths in passing in correspondence with the Indian Office, they did not take responsibility for the outcome. Superintendent Pierce of the Flandreau Indian School, for example, sent home several pupils who had not made “a real satisfactory recovery” following their bout of influenza in 1918.¹⁵² The Pierre Indian School also sent several students home “in poor physical condition.” At least two of these “died at their homes from bad effects of ‘flu.’”¹⁵³ When influenza struck the Pipestone Indian School for a second time in 1920, Superintendent Mann sent two children home “in poor condition after the attack.”¹⁵⁴ The superintendent of the Albuquerque Training School, meanwhile, sent one of the “severe cases” of influenza to a tuberculosis sanatorium, where the boy later died.¹⁵⁵ Those children sent home may have found comfort among kin in their final days, but such a return of weakened and dying children must have weighed heavily on Indigenous parents who had entrusted their care to the federal government.

The Indian Office was also invested in downplaying mortality statistics at the boarding schools. At certain institutions, the high number of student deaths was too shocking to ignore and the Indian Office launched special investigations to see what had gone wrong. The Pierre Indian School suffered a particularly high mortality rate, with 16 out of 219 students in attendance at the school dying within a few short weeks.¹⁵⁶ The Indian Office sent Special Supervisor Lawrence Michael to Pierre to conduct “an investigation in a quiet manner.”¹⁵⁷ Michael uncovered a series of missteps at Pierre, including the ongoing enrollment of students at the school after the arrival of influenza, the indifference of employees who cared for the sick, the lack of proper nourishment provided to convalescents, and inadequate bathing and toilet facilities, which put “an extra tax upon the strength of the student patients.”¹⁵⁸ Similarly, the office sent the chief supervisor of Indian schools, O. H. Lipps, to the Cushman Trades School after the school reported multiple student deaths. Lipps found the school “turned into one big hospital.” Although he commended the efforts of employees and older students to care for the sick and insisted “Superintendent Hammond seemed to be doing everything possible to prevent further spread of the epidemic,” he nevertheless pointed out serious deficiencies in the school’s staffing and facilities. Indeed, Lipps closed his report by arguing that in the future the school should “either be improved or abandoned.”¹⁵⁹

For the most part, the Indian Office chose to ignore or excuse mistakes uncovered at the nonreservation boarding schools and to bury negative reports. Despite the clear deficiencies that Special Supervisor Michael had uncovered at Pierre, for example, Commissioner Sells concluded that circumstances were such as could not “have been corrected in time to have effected [sic] the situation that developed.” In an intraoffice memorandum, Sells defended Superintendent Crandall by insisting that no one who knew him “would believe for a moment that he would have admitted pupils in the face of danger.” The suddenness of the influenza outbreak had thrown everything into “chaotic confusion,” but Crandall and his employees had done “the best they could.”¹⁶⁰ The commissioner did not dwell on the fact that general poor conditions at the school had left Crandall ill equipped to weather the crisis.

In the Indian Office’s annual report of 1918–19, Sells chose to focus almost exclusively on the positive. He bragged about the successful quarantine efforts of the Chilocco Indian School in Oklahoma, where “not a single case developed among 600 pupils and employees,” and he praised the “great personal effort and sacrifice” of Indian Service employees during the pandemic. Eager to forget the hardships of the past year, Sells looked to a future of “work and prosperity” for his Indigenous charges, which he hoped would lead to their “health and happiness” (Secretary of the Interior 1919: 73). In the years directly following the pandemic, the office did close some of its least successful boarding schools, including the Cushman Trades School. This effort was more the result of postwar financial retrenchment than an admission that these institutions could not properly care for Indigenous pupils (Roberts 1987: 221).

Quantitative Analysis of the Role of Different Factors in Pandemic Outcomes

As indicated by the preceding discussion, a number of factors may have contributed to epidemic outcomes at the schools in general, as well as to variation in outcomes between schools. The school reports contained sufficient information for several variables (tables 1 and 2) to permit statistical analyses testing the relationships between them and the estimated morbidity, mortality, and case fatality rates during the second and third waves of the pandemic. We analyzed the second and third waves together since the Indian Office collected statistical data on the pandemic during the late spring and summer of 1919, which covered both of these waves. As previously mentioned, most of the boarding schools experienced outbreaks during the second wave, with only Carson experiencing a true third wave outbreak. The Indian Office did not collect similar statistical data for the first and fourth waves of the pandemic, although superintendents mentioned these outbreaks in their annual reports.

Demographic variables include region, which may correspond to larger patterns of geographic spread of the pandemic in the United States, as well as differences in regulations or practices for schools. We assigned schools to one of five regions—West, Southwest, Lower Midwest, Upper Midwest, and the Dakotas—based on their locations. The schools also educated students up to different grade levels, including vocational training, broadly categorized here into three groups of up to sixth, seventh or eighth, and tenth grades. As students may have been placed in different levels based on their academic abilities or previous schooling, grade level does not perfectly match age. However, the number of grade levels or the approximate ages of students enrolled in different schools may affect contact patterns, probability of transmission, and/or severity of illness due to immunological risk factors.¹⁶¹ Further, each school recruited students from different areas potentially including multiple Indigenous communities, which may have experienced population-level or reservation-level variation in factors such as distance traveled to school—and thus potential exposure—or previous epidemic history and general health. Finally, sources reported average attendance in the 1918–19 academic year, approximate capacity of the schools, and lists of employees. These data enabled calculation of density, which likely impacted transmission of the disease, as well as the student-employee ratio, which could be considered a proxy measure of the health care that may have been available for ill students. Several pandemic-related variables are also considered in these analyses. As discussed above, previous exposure to influenza during the first wave may have affected immunity during the second wave. Therefore, we categorized schools according to whether a spring 1918 wave was reported and broadly how severe it was. We also considered attempts to mitigate the pandemic’s impact, including efforts to quarantine or isolate the students and the use of vaccines.

The outcomes of interest should be taken as approximate estimates. In all cases, the population at risk is based on average attendance reported, not actual attendance during key months, so there may have been changes or loss of students throughout the year for both pandemic and nonpandemic reasons, such as the enlistment of older boys in the military. The number of cases used to calculate morbidity rates were sometimes reported in qualitative terms such as “only a few” students or “half the student body” that were then interpreted into quantitative values. If mortality calculations returned that more than 100 percent of the assumed population at risk died, this value was changed to 100 percent but the population at risk value was not adjusted. Further, death counts used to estimate mortality do not include students who were reported to have been sent home or elsewhere before they succumbed or children of white employees. Because case fatality is a function of morbidity and mortality, the values for this outcome are affected by the same data concerns. Nonetheless, with these caveats in mind, the estimates can be used to broadly evaluate the role of different factors on health outcomes.

Each independent variable was tested individually using either t-tests/analysis of variance (ANOVA) for categorical variables like region or grade level, or linear regression for the number of tribes represented, school density, and student-employee ratio. Further, we ran each test four times. The first set of tests included all schools with information for the relevant variables. The second set of analyses were limited to only the schools that experienced the fall 1918 and spring 1919 waves, so the average outcomes were not skewed by those that escaped. Finally, we repeated both these sets excluding the potential outlier of Pierre due to the unusually high mortality and case fatality rates estimated for this school. The vast majority of analyses returned nonsignificant results, perhaps at least partly due to the small sample size. In the interest of space, only results that were significant (p < 0.05) or approaching significance (p < 0.1) are presented and discussed here.

Significant associations were found most commonly with average morbidity during the fall 1918 and spring 1919 waves. Region was moderately significant when all schools (F = 2.76, p = 0.06) or all schools without Pierre (F = 2.52, p = 0.08) were included, but not for either analysis of only those schools that experienced the fall 1918 and spring 1919 waves. Post hoc t-tests suggest these results are driven by differences in the average morbidity between the Western (82 percent) and Lower Midwest (16 percent) regions and between the Dakotas (71 percent) and Lower Midwest.¹⁶² Attempting to quarantine or isolate was significant or close to significant in the all-school analyses with and without Pierre (t = 2.05, p = 0.054 vs. t = 1.91, p = 0.07, respectively) and for the fall 1918 and spring 1919 waves only schools with and without Pierre (t = 2.14, p = 0.048 vs. t = 2.34, p = 0.058, respectively). In all cases, those schools that attempted the mitigation strategies had lower average morbidity, by 22–29 percent, than schools that did not.

The remaining significant results for morbidity were found only in the analyses of schools that had experienced outbreaks in the fall of 1918 and the spring of 1919. In these analyses, average morbidity was significantly different when comparing schools based on maximum grade level, with and without Pierre (F = 4.58, p = 0.03 vs. F = 5.72, p = 0.02, respectively). Post hoc t-tests indicate this result is driven by differences between schools that went up to grade 6 (100 percent) vs. the other two types (68 percent and 54 percent for maximum grade of 7/8 and 10, respectively). Thus, to the imperfect extent that grade level correlated with age, the schools with only younger pupils experienced higher estimated morbidity. This result may reflect the difficulty of ensuring that younger children followed sanitary guidelines. Further, schools with fall 1918 and spring 1919 waves that attempted to vaccinate students also had lower average morbidity than those that did not, whether analyses include or exclude Pierre (t = 2.44, p = 0.03 vs. t = 2.34, p = 0.03, respectively). This result does not necessarily mean that the different vaccines used were effective in preventing transmission or illness, as it does not control for other characteristics of the schools including whether other mitigation strategies were attempted. However, there is not a significant association between schools that attempted to quarantine or isolate and those that attempted to vaccinate (Χ² = 0.202, p = 0.65). In summary, public health responses or interventions, regardless of their biological effectiveness, seemed to have the most meaningful relationships with variation in morbidity in the schools, while region and grade level also had some influence. Results are more marked when considering only the subset of schools that had an outbreak in the fall of 1918 or spring of 1919.

Average mortality rates at the schools were the only measures significantly affected by density. In analyses considering all schools except Pierre, regression analyses show a negative trend (B = −4.515, p = 0.045, r² = 0.187). A similar but stronger trend (B = −5.250, p = 0.03, r² = 0.262) is seen when fall 1918 and spring 1919 wave–only schools, excluding Pierre, are analyzed. Both results suggest that schools with more crowding had lower mortality. A potential explanation for this counterintuitive result is that schools where capacity and attendance more closely matched or perhaps were even overcrowded were better managed overall because the superintendents were more effective or efficient at recruiting and/or the school facilities were better so more students enrolled. Lower density might indicate problems with the school in general, such as insufficient funding or poor facilities. Indeed, after World War I, several boarding schools with low enrollment closed for financial retrenchment reasons and due to insufficient infrastructure or facilities. Finally, there were no significant results for any of the tests involving average case fatality.

Despite the caveats mentioned above regarding sample size and data quality, results of these analyses broadly suggest that several factors contributed to variation in mortality and especially morbidity among the schools. A lack of individual-level information prevents consideration of personal factors that may have influenced health outcomes, such as sex or preexisting health conditions. On the other hand, the factors that we considered here are largely related to management and response decisions that were made—and so could have been made differently—by the superintendents or school system.

Conclusion

As this article demonstrates, the detailed records kept by the Indian Office provide a rare opportunity to qualitatively and quantitatively evaluate the severity of the influenza pandemic of 1918–20 in institutional settings as well as to assess the varying success of mitigation efforts. Superintendents who invested in preventive measures were able to reduce morbidity rates at their institutions, which also reduced their mortality rates. Superintendents who failed to take such precautions as well as those hampered by insufficient resources witnessed severer outbreaks and death counts. Although it is possible to analyze these variations between schools, it is ultimately difficult to assess the Indian Office’s broader claim that Indigenous students were safer at school than at home since there is no comparable statistical data on the morbidity and mortality rates of similarly aged Indigenous youths who did not attend school. Influenza also struck the reservations hard, and available sources make it impossible to know if students would have fared better under the care of their families. What we can say is that in general the nonreservation boarding schools were not safe places during the influenza pandemic. Scholars have estimated the overall mortality rate for the United States as a whole during the fall 1918 and spring 1919 waves of the influenza pandemic as 0.64 percent (Crosby 1989; Ewing 2021). In contrast, the overall mortality rate for students at the nonreservation boarding schools during the fall of 1918 and spring of 1919 was 1.5 percent. When we exclude from our calculations the four schools that avoided influenza during the second and third waves, this rate increases to 2.5 percent. Certain institutions experienced extraordinarily high mortality rates—in particular, the Pierre Indian School (7.3 percent), the Hayward Indian School (5.6 percent), the Rapid City Indian School (4.8 percent), the Cushman Indian School (3.3 percent), and the Salem Indian School (3.0 percent). These outcomes fall into a general pattern of poor health at the nonreservation boarding schools in the early twentieth century—a pattern that reflects the insufficient funding that Congress set aside for these institutions in this period.

Notes

Works Cited