On the night of November 19, 1957, at 11:18 p. m. , an ambulance belonging to the city of New York turned off 125th Street onto Lenox Avenue and stopped outside a five-story tenement. The call had come in at 10:50 p.
m. The distance from the receiving hospital to that address was 1. 6 miles. The ambulance had taken 28 minutes to cover it.
It was cold, 34 degrees, with a thin rain falling. A small crowd had gathered on the sidewalk outside the tenement—women in housecoats with overcoats thrown over them, men who had come down from upper floors in undershirts and trousers. Nobody was shouting. That had stopped some minutes earlier.
Inside the third-floor front apartment, a 34-year-old woman named Odessa Apprentice Vaughn had been in labor since shortly after 8:00 p. m. Something had gone wrong with the delivery. The bleeding had begun around 10:10.
A 15-year-old boy had been sent down the stairs at a run with a nickel and an address, and had placed the call from the pay telephone in the luncheonette. The ambulance arrived at 11:18. The attendants went up and came down at 11:41 carrying Mrs. Vaughn on a stretcher.
She was alive then. She died at 12:04 a. m. at the hospital of hemorrhage, having lost an amount of blood that a transfusion begun 40 minutes earlier would very likely have made survivable.
The child, a girl, lived. Among the people standing on that sidewalk was a woman named Lavinia Aurelia Tarrant, 41 years old, who lived on the fourth floor of the building next door and had been standing there for nearly an hour. She was a practical nurse who had trained at a hospital in Norfolk, Virginia, and had come north in 1939. She worked days doing private duty for two elderly patients and nights, three nights a week, on the ward at a small proprietary hospital on 135th Street.
She had been up in the apartment with Mrs. Vaughn from 9:00 until the ambulance came. She had done what could be done without equipment, which was not much, and she knew it. She had come down the stairs behind the stretcher, wearing a gray wool coat over a housedress, with no hat, her hair wet, blood on her sleeve.
Standing near the curb, surrounded by several men in overcoats, was Ellsworth Raymond Johnson, the man everyone in that neighborhood called Bumpy. He was 61 years old. He had come because someone had telephoned him, a thing that happened on that avenue when something went badly wrong, and because he had a habit of arriving at trouble. He had arrived 6 minutes after the ambulance—22 minutes too late to be of any use whatsoever, and he knew it.
A man from the city, not a police officer but some kind of supervisor who had come with the ambulance in a separate car, was standing near the tenement door with a clipboard, asking questions in the flat voice of a man completing a form at the end of a long shift. He wanted to know who had placed the call and at what time. Mrs. Tarrant told him.
She gave the time as 10 minutes to 11 and gave the name of the boy who had made the call. Then she said that the delay had been 28 minutes and that the patient had been bleeding for the last 30, and that this was the fourth time in 14 months she personally had waited over 20 minutes for an ambulance on that stretch of Lenox Avenue. The man with the clipboard did not look up. He said, “Are you family?
” She said she was not. He said, “Are you a physician? ” She said she was a practical nurse. He said, still not looking up, “Then you don’t have information for me.
Step back, please. ”
Then, to one of the attendants, in the same tone but slightly louder, in a way that carried across the sidewalk to perhaps 20 people: “Every time everybody up there hears a doctor. ”
There was no laughter. It was worse than laughter.
There was the particular silence of 30 people who have all heard something and have all decided at the same instant not to have heard it. Lavinia Tarrant did not answer him. She looked at the man for a moment, then past him at the doorway of the building, then stepped back exactly as she had been told, two paces onto the wet sidewalk, and stood there with her hands at her sides. Bumpy Johnson had heard it.
He took a step toward the man with the clipboard and stopped. After a moment he turned and said something to one of the men with him, and the man went and got the car door. There was nothing for him to do. That is the whole significance of his presence that night.
A man who could settle almost any dispute on that avenue within an hour, who had access to money and influence and a network of obligation running through half of Harlem, had arrived 6 minutes after an ambulance that had itself arrived 28 minutes after a telephone call, and the entire apparatus of what he was could not compress those 28 minutes by a single second. The thing that had killed Odessa Vaughn was a dispatch schedule. Before he left, he crossed to where Mrs. Tarrant was standing and said something to her—one sentence, and an offer of money for the family, which she declined on the family’s behalf and later told them about.
They accepted it, and it paid for the burial. Lavinia Tarrant went back upstairs at close to 1:00 in the morning. She washed the coat sleeve in cold water in her own kitchen sink, hung the dress over a chair, and sat down at her table. She did not go to bed until nearly 4.
She did not cry that night and did not the following day. What she did instead was to stop thinking about the man with the clipboard. She understood something clearly. The man with the clipboard had been rude and dismissive, but if he had been courteous, Odessa Vaughn would still have been dead.
The insult was not the injury. The injury was an arrangement of vehicles, stations, telephones, and schedules drawn up somewhere by people who had never stood on Lenox Avenue in the rain. And it was killing people at a rate that nobody had ever counted. That last part was the thought that kept her at the table until 4 in the morning.
Nobody had ever counted. She knew the ambulances were slow. Everyone knew it. It was among the most thoroughly known facts in Harlem, and it had precisely the standing that thoroughly known facts have when nobody has written them down, which is to say none at all.
She had said “fourth time in 14 months” on the sidewalk, and it had been true, but she could not have proved it. She had been watching this for 18 years and had never once written down a time. She began the following night. The first thing she made was crude, and she replaced it within 2 months.
But the principle never changed. It was a small notebook, and in it she recorded for every emergency call she had any knowledge of four items: the time the call was placed, the time the vehicle arrived, the address, and the source of her information. She added a fifth column within a week: the outcome, because she realized that a delay without a consequence attached to it would be dismissed as an inconvenience. She understood that a record kept by one person is an opinion, and that a record kept by 40 people using the same form is a body of evidence.
She spent the next 14 months building the second thing. She began where she had standing, which was among nurses. There were by her own eventual count somewhere over 300 women working as practical nurses, registered nurses, midwives, and nurses’ aides in the 20-odd blocks she considered her neighborhood. They worked at four hospitals, two clinics, a health center, and in private homes.
No organization of any kind connected them. She did not attempt to found one. What she did instead was considerably smaller and worked. She had a printer on 132nd Street run off 200 cards, 4 by 6 inches, with five ruled columns and a heading.
It did not say anything about a campaign or a committee. It said: Call Record—Time Call Placed, Time Vehicle Arrived, Address, Outcome, Reported By. She gave them out one at a time in hospital corridors and at bus stops and in the back of the drugstore on 130th Street, where nurses coming off nights bought coffee. She did not ask for membership or agreement or attendance.
She asked each woman to fill out a card whenever she happened to know the two times and to leave it with one of four people whose names were on the back. Between 60 and 70 percent of the women she approached did nothing at all, which she had expected. The rest returned cards. By the summer of 1958, she was receiving between 15 and 25 cards a month.
By the spring of 1959, it was over 60. She was rigorous about the cards in a way that struck several people as excessive at the time, and turned out to be the reason the entire thing eventually could not be dismissed. She discarded any card that lacked a source. She discarded any card where the time of the call was an estimate rather than a known figure.
She wrote to the reporting nurse asking for verification on any card where the delay exceeded 40 minutes, on the ground that the most extreme figures would be attacked hardest and had to be the most solid. She was building, though she did not use this language, a data set—built to a standard she had arrived at by asking herself repeatedly what a hostile person would say about each entry, and then removing every entry about which anything could be said. The four collection points were chosen with more care than they appeared to require. One was the drugstore on 130th Street, whose proprietor kept a cigar box under the counter.
One was a beauty parlor on 135th Street, which caught the women who did not work in hospitals at all. One was the vestry office of a church on 132nd Street, accessible to women who would not have gone into a drugstore to hand a stranger a piece of paper about a death. The fourth was Mrs. Tarrant’s own mailbox, for the women who wanted no one at all to see them do it.
That fourth point mattered more than the others. A substantial number of the cards came from women employed by institutions that would not have been pleased to learn that their staff were recording response times. She also, from the beginning, kept a second and entirely separate record that she showed to nobody for 4 years. It was a list of every call she learned of in which the response had been prompt.
There were a good many of them. By 1962, the prompt column held over 400 entries. She kept it because she had worked out that a record of only the failures would be described as a collection of complaints. When the submission was assembled in 1962, the prompt entries were included in full, and their presence is the reason the average figure of 21 minutes and 40 seconds could be presented as an average rather than as a selection.
The second ally came to her in the ordinary course of the work. His name was Dr. Amos Kellaher Brath, 56, a general practitioner with an office on Edgecombe Avenue who had been in practice in Harlem since 1931. He was not an activist.
He was a physician with a large practice and a habit of grumbling. His value was threefold. He could attach a clinical judgment to an outcome, converting a lay person’s assertion that someone would have lived into a physician’s assessment, which is a different category of statement. He had access through the county medical society to physicians at other hospitals, and could ask questions that Mrs.
Tarrant could not have asked without being reported to somebody. And he was respectable in a way that could not be undone. When a physician of that description states a figure, the burden falls on the person who wishes to dispute it. He agreed to review cards.
He reviewed them for 4 years. He was paid nothing and complained about the work continuously and never missed a month. The third ally was an attorney named Corin Deline Aubrey, 34, one of a small number of Black women practicing law in the city at that time. She mattered because she understood something that neither Mrs.
Tarrant nor Dr. Brath had grasped, and she explained it at their first meeting in September of 1958, and it redirected the entire effort. Mrs. Tarrant had assumed that the object was to demonstrate that Harlem’s ambulance service was slow and to present that demonstration to the Department of Hospitals and ask for improvement.
Miss Aubrey told her that this would accomplish nothing. The department would say that emergency response times varied with traffic, distance, call volume, and vehicle availability, that it was operating within its appropriations, and that it would give the matter consideration. The reason this reply would work, she explained, was that there was no standard. Slow is a comparison, and a comparison requires something to compare against.
The department had never published a target response time. It had never published response times at all. Without a published figure, 28 minutes was not a failure. It was simply a number.
The object, therefore, was not to prove that the service was slow. The object was to compel the creation of a standard and to compel the publication of the figures measured against it. The fourth ally was an alderman’s clerk, a young man named Isidor Portius Marquetti, 26, whose function was to explain the calendar. He told Mrs.
Tarrant that the Board of Estimate held budget hearings each spring, that the Department of Hospitals appeared before them, that any citizen or organization could request to be heard under an existing procedure, and that the department’s appearance included the submission of a statement of operations. Any figure the department stated in that submission became a figure it had to account for in every subsequent year. The fifth and least likely ally was a man named Selwyn Portius Yearley, 63, who owned two funeral homes and had buried a considerable proportion of the people whose deaths appeared on the cards. He had his own business records, which established for a large number of individuals the date, time, and place of death.
Where a nurse’s card said that a woman on 133rd Street had died in October of 1958 after a delay, Mr. Yearley could confirm that a death had occurred at that address in that month, which converted a nurse’s recollection into a documented event. He also mattered because he could say, when asked by people who found the whole effort distasteful, that the families were not opposed. Eleven of them eventually gave written permission for their circumstances to be described in the submission.
Every one of those permissions was obtained by Mr. Yearley personally, sitting in front rooms over a period of 7 months. The strategy, as it stood by the middle of 1959, had four elements. The evidentiary element was the foundation.
The cards accumulated. Dr. Brath annotated them. Mr.
Yearley confirmed the deaths. Mrs. Tarrant maintained the master ledger, which by 1960 was three bound volumes, and computed monthly averages by hand. The administrative element was Miss Aubrey’s, and its purpose was not to obtain anything but to establish in writing that certain things did not exist.
She wrote to the Department of Hospitals in March of 1959 requesting the department’s published standard for emergency response time. The reply in May stated that no such standard was published. She wrote again requesting response time data for the receiving hospitals serving Manhattan north of 110th Street. The reply in August stated that such data was not compiled in the form requested.
She wrote a third time asking whether it was compiled in any form. The reply in November stated that individual call records were maintained at the receiving hospital level. Those three replies were worth more than the cards. They established, over the department’s own signature, that the city did not know how long its ambulances took, had no standard for how long they should take, and had never looked.
The economic element was Mrs. Tarrant’s own. In the spring of 1960, she approached the three insurance agents who wrote the great majority of the burial and life policies sold in those blocks. She asked whether their company’s underwriting for the neighborhood reflected mortality experience, and whether that experience included deaths in which emergency response time had been a factor.
A letter from a regional office in September of 1960 stated that mortality experience in the district was materially above the borough figure and that the differential was reflected in premium rates. This converted the matter into money. Residents of those blocks were paying higher premiums for burial insurance, and the city was collecting the same taxes from those blocks as from any other. The public element came last and was held deliberately for 18 months against considerable pressure from within the group.
Mrs. Tarrant told one pastor that anything published before the department had stated in writing that it had no standard would be answered by the department announcing a standard the following week, ending the matter with a press release and nothing else. The department’s third reply, stating that individual call records existed at hospital level, arrived in November of 1959. That reply was the thing she had been waiting for, because it established that the underlying data existed and had simply never been assembled.
The execution took from November of 1959 to April of 1962, and it consisted very largely of tedium. Mrs. Tarrant worked days and three nights a week throughout. The ledger work happened on the four remaining evenings and on Sunday afternoons in the church basement on 132nd Street, at a table with a fountain pen.
Between the department’s third reply in November of 1959 and the beginning of submission preparation in December of 1961—a period of 2 years and 1 month—the effort produced no event of any kind. Cards came in, cards were sorted, entries were made, confirmations were sought from Mr. Yearley and were sometimes obtained and sometimes not. Nothing was announced, nothing was demanded, and nothing changed.
During those two years, the group lost the beauty parlor collection point when the proprietor sold the shop. It lost Dr. Brath for 11 weeks in the spring of 1961 when he was ill. It lost at various points four of the seven core women, three of whom returned.
Mrs. Tarrant’s own contribution during those two years was almost entirely clerical. She said once to Miss Aubrey that she had come to think of the ledger as a thing that had to be fed on a schedule whether or not anyone was watching, and that the whole difficulty was that nobody could see the difference between feeding it and doing nothing. There was a particular evening in March of 1961, recorded because two people mentioned it independently, when she arrived at the basement to find that no one else had come.
She worked alone for 3 hours, entered 41 cards, locked the ledger in the vestry cabinet, and walked home. The submission preparation that began in December of 1961 was 4 months of work of a different character. Miss Aubrey required every figure in the ledger to be traceable to a card, every card to a source, and every death to a confirmation. She sampled 120 entries at random and traced each one completely.
Two failed the trace. Both were removed. She then traced a further 140 entries before satisfying herself that the failure rate was low enough that the department could not use it. By April of 1962, the ledger contained 1,147 verified entries covering 52 months.
The average interval between call and arrival across all entries was 21 minutes and 40 seconds. Dr. Brath had annotated 39 entries in which a death had occurred and had identified 141 in which, in his professional judgment, the delay was clinically significant. There were three attempts to stop it, and none of them succeeded.
In February of 1960, Mrs. Tarrant was informed by the proprietary hospital where she worked nights that her services would no longer be required. No reason was given, and she asked for none. She was without night work for 11 weeks, then obtained private duty work that paid slightly less.
She did not mention the dismissal to anyone in the group for over a year. When it emerged in 1961, several of the women wanted to make an issue of it. She refused, on the ground that the moment the campaign became about her employment, it would become a personnel dispute, which was a category of thing the city knew how to handle. In the autumn of 1960, an official from the Department of Hospitals described the card collection as unscientific and the figures as anecdotal.
Mrs. Tarrant’s response was to write to that official courteously, enclosing a blank card and inviting him to specify what additional field would render the record scientific, stating that she would add it to the form and to all future collection. She received no reply. She kept the copy of her letter, and it was submitted with everything else in 1962.
The word “anecdotal” was not used again by anyone in an official capacity. The third attempt was an offer, made in the summer of 1961 through an intermediary, and it was the most dangerous of the three. The proposal was that an additional ambulance would be stationed at a hospital serving the district effective within 90 days, in exchange for the group’s withdrawal of its request to appear before the Board of Estimate. The offer was real.
An additional vehicle would have measurably reduced response times immediately, for actual people. Four of the seven women in the core group wanted to accept. The discussion took two evenings. Mrs.
Tarrant’s position, recorded in the group’s minutes in a hand that was not hers, was that a vehicle assigned by decision could be reassigned by decision; that no one outside the department would ever know if it were; and that the group would have traded the only thing it had—the ability to compel a published standard—for something it could lose in a year without being told. The vote was four to three to decline, having been three to four to accept two days earlier. The appearance before the Board of Estimate took place on April 24, 1962. The submission ran to 61 pages and had been prepared by Miss Aubrey from Mrs.
Tarrant’s ledgers over 4 months. It contained the 52-month response data with monthly averages, Dr. Brath’s clinical annotations, the 11 family statements obtained by Mr. Yearley, the three departmental replies establishing the absence of any standard, the insurance letter, and a two-page statement of what was requested.
What was requested was in three parts: establishment and publication of a maximum target response time for emergency ambulance calls citywide; compilation and quarterly publication of actual response times by receiving hospital; and review of vehicle and station allocation by an authority outside the Department of Hospitals. Mrs. Tarrant spoke for 7 minutes. She read no account of the night of November 19, 1957, and did not mention Odessa Vaughn by name.
Miss Aubrey had advised against it, and she had agreed without argument. She presented the 52-month average, the distribution, the department’s three replies, and the request. Dr. Brath spoke for 5 minutes and stated the clinical finding.
Mr. Yearley spoke for 2 minutes and said that he had personally obtained the 11 family permissions and that no family had declined. The department’s representative spoke for 9 minutes. He did not dispute the figures, because there were no departmental figures to dispute them with.
The department had spent 3 years establishing in writing that it did not know how long its ambulances took. It was therefore in no position to say that they were faster than the ledger said. The matter was referred that afternoon for administrative review. The review took 22 months.
Its report in February of 1964 found that the department maintained no response time standard; that call records at receiving hospitals were kept inconsistently and were not reported upward; and that the absence of any compiled data made the adequacy of vehicle allocation impossible to assess. It recommended a standard, mandatory uniform call logging, and quarterly reporting. A departmental target response time was established in September of 1964. Uniform call logging became mandatory at receiving hospitals in January of 1965.
Quarterly reporting by receiving hospital began with the quarter ending June 1965. The figures for the first published quarter showed an average response time for the two hospitals serving the district of 19 minutes and 11 seconds, against a citywide average of 12 minutes and 4 seconds. That table was the victory, and it did not look like one. It was a small block of print in a document almost nobody read.
Its effect was that the argument no longer had to be made. It had been made by Lavinia Tarrant for seven years with a fountain pen. After June of 1965, it was made four times a year by the Department of Hospitals in its own publication, over its own name, without anyone asking it to. Two additional vehicles were assigned to the district in 1966.
The gap between the district average and the citywide average narrowed from 7 minutes and 7 seconds in 1965 to 2 minutes and 50 seconds in 1971. It narrowed because it was visible, and it was visible because it was printed. Miss Aubrey used the same structure in 1966 in a matter concerning building heating violations, which produced a published inspection standard in 1968, and again in 1969 concerning school nurse coverage, which failed. Her own account of the failure identifies the reason as the absence of anything corresponding to Dr.
Brath. There had been no professional willing to attach a clinical judgment to the records, and without that the figures remained a description of an inconvenience rather than of a harm. Dr. Brath spent the rest of his career telling younger physicians that the most useful thing a doctor could do in a poor district was to sign his name to other people’s records.
Lavinia Tarrant continued to keep cards until 1968. She stopped because the department’s own quarterly figures had become more complete than hers, which she regarded as the correct outcome, and said so. She worked as a nurse until 1974. She was appointed in 1967 to an advisory committee on emergency services, on which she served for 6 years.
In 1969 she was among those consulted during the planning that led to the citywide emergency telephone number. She gave one interview in 1970 to a nursing journal, in which she was asked what had made the campaign work. She answered that they had never asked for anything the city could give them as a favor. Odessa Vaughn’s daughter was raised by an aunt on 126th Street.
She became a hospital administrator. It is not recorded that she ever knew the particulars. Bumpy Johnson died in July of 1968. He had no involvement in any of it after the night on the sidewalk, and there is no evidence that he followed it or knew it was happening.
His presence in this account is not incidental. He was in November of 1957 very close to the most powerful private individual in Harlem, and his power was of a kind that worked on people. It reached those things quickly and reliably, and it was real power. What it could not reach was a dispatch schedule.
Because a dispatch schedule has no interests, cannot be persuaded, cannot be threatened, does not want anything, and does not know that anyone is standing in the rain. The 6 minutes by which he was late that night were not the point. He arrived at the scene of a failure that his entire apparatus was structurally incapable of addressing, and he stood on the sidewalk and understood that, and offered money for a burial, which was the only instrument he had. The woman he had watched being told to step back, who had blood on her sleeve and no standing to speak, did in seven years what he could not have done in 30.
She did it with index cards, a ledger, a physician’s signature, an undertaker’s date books, and a great deal of patience about not being listened to. Power that works on people stops at the edge of systems. What is required past that edge is documentation, and the willingness to keep producing it for years while being told that it is anecdotal, and the discipline to refuse the vehicle that could be taken back in exchange for the number that could not.
The number was published in June of 1965, and it has been published in some form ever since.