The Epitaphs in the Annual Report

Seen from above in early 1921, the city grid spread out, its streets once again filled with movement and crowds free of masks, the normal rhythm of life restored. The pale stone of commercial buildings gave way to rowhouse blocks, and the distant river lined a horizon where recent construction had replaced the structures of the recent past. From such a height, the epidemic appeared in the official mind as a concluded visitation, its passage marked only by an administrative residue left behind. By mid-1920, the pandemic was largely considered to be ‘over’ by the public as well as governments, with deaths from pneumonia and influenza ‘very much lower than for many years’ in the United States.

The residue had weight. It occupied three inches of municipal buckram, the color of wet sand, stamped in black with the seal of Philadelphia. The Bureau of Health had commissioned its printing in late 1920. The contract went to a firm on Callowhill Street that handled city documents: tax rolls, court reports, the annual summaries of the Water Department and the Bureau of City Property. The bindery workers had handled epidemic materials before—the emergency ordinances of October 1918, the closure orders, the bulletins on prevention.

Now they assembled something more permanent. They gathered the mortality returns from forty-eight wards, the hospital admission ledgers, the burial permits from seventeen cemeteries, the sick reports from the Navy Yard and the Army depot. They fed these into a system of collation and abstraction that had been refining itself since the city charter of 1854. The result was a volume that would outlast the marble markers at Holy Cross.

The book weighed four pounds. Its title page announced it as the Annual Report of the Bureau of Health of the City of Philadelphia for the fiscal year ending December 31, 1919, though its narrative reached back to September 1918 and its tables extended through the winter of 1920. The discrepancy between fiscal year and biological event was itself a kind of statement. The city recorded time by budget cycles. The epidemic had erupted in the middle of one, and the report contained it within the administrative unit that happened to be open when the dying stopped. The epidemic first hit Philadelphia through the Philadelphia Navy Yard on September 19, 1918, from sailors returning from Europe.

A clerk in the statistical division had begun the work in early 1919. He faced shoeboxes of paper that had accumulated faster than any previous event in the bureau’s history. The death certificates arrived from registrars who had worked through nights in October 1918, their desks piled with forms that recorded temperatures of 105 degrees, hemorrhages from nose and ears, cyanosis that turned young adults the color of plums. Each certificate named a specific failure of the body’s defenses: lobar pneumonia, bronchopneumonia, “epidemic influenza” in the handwriting of physicians who had signed dozens in a single day, their script slanting sharply rightward from haste and exhaustion. The certificates came from the Navy Yard hospitals, from the emergency wards set up in school gymnasiums, from private homes where no doctor had visited for three days because none were available.

The clerk’s task was transformation. He sorted the certificates by ward, by week, by cause coded according to the International List of Causes of Death adopted by the American Public Health Association. The raw documents smelled of mold and coal dust. They recorded that a woman of twenty-four had died on October 10 at 2232 South Sartain Street; that a motorman of thirty-one had died on October 12 at the University Hospital; that an infant at St. Joseph’s Orphanage had died unnamed on October 14. The clerk transferred these particulars to ledger books, then to summary sheets, then to the forms that would become printed tables. The table listed “Week Ending October 5” beside the figure 289, and “Week Ending October 12” beside 4, 597. The names vanished into the column heading “Excess Mortality Over Previous Years.”

The number settled at approximately 17, 000 dead, though the exact figure shifted depending on whether one counted only certificates marked “influenza” or included the secondary pneumonias that finished what the virus began. The report settled the question by counting both, then presenting the total as a column of integers without names. The clerks had calculated the “years of potential life lost,” a new metric borrowed from actuarial science. The figure ran to six figures. It appeared in a footnote. The main text moved on to discuss diphtheria immunization rates in 1919.

The narrative occupied seventeen pages near the front of the volume. Its prose stated that “the disease visited the city” in late September. It noted that “the situation became acute” during the second week of October. The verbs were intransitive. The report recorded that “cases multiplied,” that “hospitals became crowded,” that “the supply of caskets ran low.” The grammar performed a careful operation. It transformed a series of human choices into a meteorological event.

The document mentioned that ‘public gatherings were prohibited’ on October 3. It did not mention that this order came five days after two hundred thousand people had mingled on Broad Street for the Liberty Loan parade, a parade city officials had projected would draw only around 10, 000 spectators. The narrative acknowledged that ‘the epidemic demanded the most strenuous efforts’ without identifying who had demanded them or from whom. It noted that ‘many physicians succumbed to the disease,’ a passive construction that avoided stating how many had died because they lacked masks or because they had been summoned to overcrowded wards without antiseptic supplies.

Wilmer Krusen contributed a preface. He wrote that the year had presented ‘unprecedented conditions.’ He wrote that ‘the epidemic demanded the most strenuous efforts.’ He did not write that he had assured the public on September 28 that the parade posed no danger, a parade the city saw as a way to raise the $259 million for which it was in charge. The report’s syntax allowed him to appear as a man managing a natural disaster rather than as an administrator who had misjudged a biological threat. The volume listed every hospital by name and every nurse by title, but it assigned no blame for the delay in closing the schools.

The bindery finished its run in March 1921. Copies went to the Mayor’s office, to the State Board of Health in Harrisburg, to the Surgeon General’s library in Washington. One copy traveled to the Philadelphia Free Library’s reference room, where the librarian stamped it with a date due slip and shelved it under 614.518. The book sat beside reports on typhoid and tuberculosis, its spine uncracked for months. Future historians would find it there, dusting off the buckram to discover that the city had officially acknowledged the scale of the catastrophe while erasing its mechanism.

The report’s mortality tables showed that ninety-nine percent of the dead had been under sixty-five. Nearly half were between twenty and forty. The demographic anomaly appeared as a dip in the curve of the city’s vitality, a valley between two peaks of normal life expectancy. The clerks had never seen such a distribution. Influenza was supposed to kill the elderly and the very young. This virus had sought out adults in their productive years, the wage-earners and parents, the nurses and the gravediggers. The report presented this fact without commentary, as a statistical feature rather than a social catastrophe. In 1918–1919, 99% of pandemic influenza deaths in the U.S. occurred in people under 65, and nearly half of deaths were in young adults 20 to 40 years old.

In the Potter’s Field at League Island, where the city had buried its unclaimed dead in trench graves during the worst weeks, no marker recorded the epidemic. The Health Bureau’s report noted that ‘burial difficulties were experienced,’ and that ‘prisoners and seminary students assisted in interments.’ It did not describe the wooden coffins stacked three deep in open pits, or the lime scattered by lantern light. The report’s description of the burial crisis occupied three sentences. The sentences were factual. They stated that ‘the usual facilities were inadequate.’ They did not state that the usual facilities had been designed for a city that died at its normal rate, and that no contingency plan existed for a city that died all at once.

Even in areas where mortality was low, so many adults were incapacitated that daily life was hampered. There were reports that healthcare workers could not tend the sick nor the gravediggers bury the dead because they too were ill. Mass graves were dug by steam shovel and bodies buried without coffins in many places.

The volume contained no photographs. It did not reproduce the death certificates with their jagged handwriting. It did not print the telegrams from the Navy Surgeon General warning of contagion among sailors. It presented instead a smoothed curve on a graph, a line rising in September, peaking in October, falling in November. The line suggested order. It suggested that the city had experienced a wave, like a flood or a heat wave, and that the wave had receded, leaving the municipal structure intact. The evening bulletin of The Philadelphia Inquirer had described the parade as ‘the first premonition of – victory’ and as ‘a great day in Philadelphia.’ Twenty-four hours after the parade had ended, 118 Philadelphians were described as coming down with ‘a mysterious, deadly influenza.’

The report’s final section discussed “Lessons Learned.” It recommended better ventilation in schoolrooms. It recommended a reserve corps of nurses. It did not recommend changes in the protocol for responding to novel infectious diseases. It did not suggest that health directors should possess the authority to cancel patriotic demonstrations. The pattern that had operated in September 1918—where official optimism suppressed alarming reports, which then made optimism seem justified until catastrophic proof broke the cycle—found its memorial in these pages. The report recorded statistics while dispersing the narrative that had produced them into passive constructions and intransitive verbs.

In the years following publication, the volume served as the primary source for every subsequent account of the epidemic. Journalists cited its figures. Historians copied its chronology. The passive voice of the original became the passive voice of the record. When future writers described the events of October 1918, they wrote that “the flu struck Philadelphia,” using the intransitive verb the report had supplied. They did not write that Philadelphia, through its elected and appointed officials, had failed to protect itself.

The International List of Causes of Death, adopted by the American Public Health Association in 1909 and revised in 1920, provided the clerks with a taxonomy that predetermined what could be seen. Influenza appeared as Code 11, “Influenza with respiratory complications,” or Code 12, “Influenza with other complications,” while the pneumonias that actually filled the morgues scattered into Codes 30 through 33, “Bronchopneumonia” and “Lobar pneumonia,” as if they were separate events rather than terminal stages of a single assault. When the clerk at his high desk ruled a line through a certificate marked “Spanish Flu” in a physician’s shaky hand, he translated the specific into the generic, the epidemic into the endemic.

The International List had not anticipated a pandemic; it had been designed to track chronic disease in industrial populations, to calculate insurance risks for factory owners and railway companies. By forcing the October deaths into these pre-existing categories, the report presented the catastrophe not as a singular event requiring civic soul-searching, but as an unfortunate clustering of routine pathologies.

The calculation of “years of potential life lost”—that six-figure footnote—represented the final transmutation of flesh into figures. Developed by actuaries for the Metropolitan Life Insurance Company to assess the economic impact of industrial accidents, the metric treated each death as a subtraction from the city’s productive capacity. A twenty-four-year-old woman who died at 2232 South Sartain Street became not a mother, not a daughter, but 41.3 years of lost labor potential; the motorman became 28.7 years of absent utility. The clerk multiplied these remainders across the ward totals, producing a sum that appeared in small type beneath Table XIV.

This was the only place in the volume where the report acknowledged that the dead had been young, that they had possessed futures. Yet by expressing that tragedy as a depreciation of municipal assets, the Bureau performed an accounting operation that paralleled the burial crews at Potter’s Field: it interred the particular beneath a layer of lime-white mathematics.

The Bureau’s choice of grammar reflected pressures that reached beyond the statistical division. In the budget hearings of spring 1919, council members had questioned whether the Health Bureau deserved its appropriation increase given its evident failure to prevent the spread of contagion during the autumn emergency. Krusen, defending his department before the Finance Committee, emphasized “natural causes” and “unprecedented virulence” rather than administrative lag. The Annual Report, compiled under this political scrutiny, became an extension of that testimony—a bound brief for the Bureau’s continued existence.

Every passive construction served a defensive purpose: if the disease “visited,” then the Bureau could not be said to have “failed to bar the door.” If hospitals “became crowded,” rather than “were allowed to fill beyond capacity,” then the lack of emergency planning disappeared into the syntax. The document was printed not merely for the historical record, but for the file cabinets of Harrisburg bureaucrats who controlled state health funding and for the reference shelves of insurance commissioners who set the city’s mortality rates—and thus its bond ratings.

The wards themselves disappeared into the abstraction. The certificates had arrived bearing postal addresses that mapped a geography of neglect: the dense boarding houses of the waterfront, the African American neighborhoods south of South Street where medical facilities were segregated and physicians scarcer, the immigrant blocks of Kensington where families shared single water taps. These specifics dissolved in the ledger books. Table VII showed “Mortality by Ward,” but the wards appeared only as numbers, One through Forty-eight, stripped of their social topography. The clerk knew that Ward Seven had contributed 847 deaths in three weeks, but the report did not note that Ward Seven was the river ward, home to the Navy Yard and the immigrant docks, or that its mortality rate per thousand was triple that of the affluent districts near Fairmount Park.

By presenting the epidemic as a uniform wave that had washed equally over all districts, the report erased the municipal decisions—about housing inspections, about hospital placement, about the distribution of nurses—that had made some neighborhoods into channels of death while others remained relatively dry.

The fiscal year ending December 31, 1919, imposed a false finitude on an ongoing trauma. The virus had not consulted the city’s budget calendar; it continued to kill through the winter of 1920, particularly among the tubercular and the cardiac-weakened survivors of the first wave. The report’s narrative stopped at December 1919 not because the biological event had concluded, but because the accounting period closed.

This chronological violence meant that the deaths of January and February 1920—hundreds of them, recorded on certificates that sat in the same shoeboxes—entered the next Annual Report as ordinary mortality, disconnected from the epidemic narrative. The bound volume thus presented the catastrophe as a contained episode, a spike between September and December 1918 with a minor echo in early 1919, rather than as a prolonged attrition that weakened the city’s population for two years. The graph’s smooth curve, rising and falling within the neat boundaries of the fiscal year, suggested a complete story with a beginning, middle, and end, when in reality the dying had merely slowed to a rate that the statistical tables could absorb without comment. Seasonal influenza began to be reported again from many places in 1921.

The physical specifications of the volume reinforced its epistemological boundaries. The Bureau had ordered a print run of five hundred copies on twenty-pound bond, the same stock used for ordinances and tax codes—paper designed for permanence but not for intimacy. Unlike the cheap newsprint of the October 1918 bulletins, which yellowed and crumbled within months, this paper was sized with alum and rosin to resist moisture and decay. The bindery used library buckram with a thread-sewn binding that would hold for centuries, ensuring that the data would outlast the memory of those who supplied it.

This material durability served as a kind of institutional confidence: the city assumed that future readers would need to consult these figures, but would not require the death certificates themselves, or the photographs of the mass graves, or the letters from nurses describing wards where patients died on stretchers in hallways. The report’s very permanence announced that the crisis was over, filed, finished—available for reference but no longer for mourning.

The term “visitation,” which appeared twice in Krusen’s preface and once in the narrative summary, carried theological weight borrowed from biblical plague narratives—the Lord’s visitation upon Egypt, the visitation of divine wrath—while simultaneously serving a secular bureaucratic function. By framing the epidemic as a visitation, the report invoked a tradition that interpreted mass death as either punishment or inscrutable providence, thereby removing the event from the realm of preventable civic failure.

The word appeared in the same paragraph that discussed “Lessons Learned,” creating a paradoxical structure: the city acknowledged that something had happened that required learning, yet described that something as an act of God or nature rather than as a consequence of specific human decisions made on September 28, October 3, and October 12. This linguistic strategy allowed the Bureau to recommend practical improvements—better ventilation, nursing reserves—while avoiding the admission that different choices by identifiable officials might have altered the column of integers printed on page 147.

The cemetery at Holy Cross remained. Its stones weathered. The names carved there—those that could still be read—faded slowly under acid rain and lichen. The municipal report did not fade. It sat in climate-controlled stacks, its buckram binding stiff, its pages uncut in some copies. The paper was acid-free, manufactured to last centuries. When researchers requested it, the librarian produced it with white gloves, handling the artifact of an event that the city had already begun to forget in the only way it knew how: by transforming it into data, filing it under number, and moving on to the next annual cycle of births and deaths that fell within normal parameters.

The bound volume, with its cold final tally, became the official epitaph. It sat on shelves in three cities, waiting for future hands to open it and find what the city had chosen to remember and what it had chosen to leave between the lines. Seen from above, across the whole map of the administrative record, the official epitaph took shape as a bound volume, cold in its final tally. It sat on shelves in three cities, waiting for future hands.