Number worship: Race, place and numbers in ‘the San Francisco America pretends does not exist’

ABSTRACT

Numerical knowledge is critical to global public health, but numbers often emerge from contexts that require serious attention to historical and contextual factors shaped by race and place-based marginalisations. Without such attention, even progressive anti-racist public health projects can cause harm in neighbourhoods like Bayview Hunters Point in San Francisco. The central role of numbers in governance has received substantial scholarly attention and ethnographic studies often foreground the subjective stakes and selective erasures that result from privileging numerical ways of knowing. Based on 12 months of ethnographic research conducted in 2003–2004, we track different interpretations around the closure of an HIV/AIDS medication adherence program in San Francisco’s only majority Black neighbourhood. Drawing on Du Bois’ double consciousness, we show the entanglement of numbers, race and place through the story of how a program’s closure was interpreted by clients in terms of racialised neglect and experimental violence. Twenty years later, in the context of the COVID-19 pandemic, the quest for numbers continues to evoke community tensions.

KEYWORDS:

• Numbers
• HIV/AIDS
• race
• place
• Du Bois
• medical anthropology

Introduction

By the 1960s, the neighbourhood was predominantly home to African-Americans and other minorities, and became isolated from San Francisco. In 1963, the writer James Baldwin called it ‘the San Francisco America pretends does not exist’. (Bayview Hunters Point Foundation, n.d.)

Shortly before the COVID-19 pandemic, while on an international flight, I (Liu) watched ‘The Last Black Man in San Francisco’ (Talbot, 2019). The scenes of the Bayview Hunters Point neighbourhood reminded me of an HIV program there that had closed in 2004. At that time, in San Francisco’s only majority Black neighbourhood, a few progressive HIV/AIDS researchers sought to demonstrate that people with HIV who were poor, marginally housed, multiply diagnosed, and African American could adhere to the complicated medication regimens that the new highly-active antiretroviral therapies (HAART) required. The advent of HAART marked a substantial turning point in HIV/AIDS treatment and management and was instrumental in the reconceptualization of HIV from a fatal disease into a chronic one. In public health circles, some worried that giving these powerful drugs to people with low rates of adherence might lead to, or at least hasten, the development of drug resistance in the virus. This research team was motivated by a moral commitment that such treatments should not be withheld. They proposed that, using a modified directly observed therapy approach with regular meetings and incentive payments, they could show that even San Francisco’s most marginalised residents could adhere to, and benefit from, the new therapies.

As part of a qualitative research arm of a larger study, I was asked to investigate what was going on at some of the program sites, including the one in Bayview Hunters Point (BVHP). The team informed me that they were concerned about this site. Specifically, it seemed that the clients’ bloodwork numbers were getting worse and so they were planning to close the program there. This is the first of two stories we present that illustrate how numerical ways of knowing and being intersect with race and place. We engage with the burgeoning literature on enumeration in global public health to consider the lived experience of those most affected. In this first story, numbers pertaining to HIV medication adherence are pitted against the lived experience of the adherence program’s clients who suggested that the program was helping and interpreted its closure through an experiential analytic of race and place. In the second, the quest for numerical and scientific knowledge reinvigourates tensions in the same neighbourhood in the context of the COVID-19 pandemic.

Through these stories we suggest that an overreliance on numerical knowledge may elide other ways of knowing, such as those informed by history and experience. We highlight the harm these elisions may cause. Numbers represent an important kind of knowledge that is critical to public health projects. They can identify issues in need of attention such as health disparities and can suggest the success, or failure, of interventions in producing their desired impact. Numbers, however, require deeper contextual consideration in order to understand the lived experiences of those they impact and, thus, also to design any kind of meaningful attempts at intervention. Without this, public health programs risk doing harm especially in marginalised and racialised communities.

A century after W.E.B. Du Bois (1994 [1903]) introduced the concept of double consciousness to characterise the experiential and psychological insults of being Black in America, a kind of double consciousness appeared to be intact in the space of a public health HIV clinical site in San Francisco’s most marginalised neighbourhood. Twenty years later still, in the context of a different pandemic, COVID-19, similar dynamics of enumeration and marginalisation endure there (Whitacre et al., 2021). Public health numbers identify sites and populations or groups that are in need of intervention or care. In the context of racialised epidemics, however, they can lead to the depiction and perception of specific groups as problematic and dangerous. The resultant stigma may enhance public blame and often gives rise to questions of medical deservingness. It places the suffering individual face to face with Du Bois’ provocative question, ‘How does it feel to be a problem?’.

Below, we introduce the broader public health HIV medication adherence program, the setting of the story, and how numbers operate therein. Next, we address the BVHP program site and the ways that both place and race shape a local and particular interpretation of the program closure. We address how this interpretation refutes the dominance of numbers and shows, rather, the subjective insults that a reliance on numbers often silences. Then we reflect on the relationship between HIV/AIDS, COVID-19 and enumeration, emphasising how broader contextual factors shape experiences in spaces of racialised marginalisation.

The adherence program

San Francisco was an early hotspot of both the HIV/AIDS epidemic in the US and related political activism. From 1993 to 1997, I worked in traditional Chinese medicine community clinics there as desperate patients sought care and hope wherever they could find it. By the mid-1990s, the advent of highly-active antiretroviral therapies (HAART) seemed to provide a foundation for controlling the disease; if, that was, patients would adhere to their medication regimens. At the turn of the millennium, a diagnosis of AIDS in San Francisco often resulted in privileged access to social services such that those living in the margins often experienced an improved quality of life in the aftermath of an AIDS diagnosis. Ehnographic work reports people wishing for such a diagnosis and bemoaning that they weren’t, ‘lucky enough to get AIDS’ (Crane et al., 2002; Liu, 2002). At the same time, clinicians and researchers found access to enhanced funding and new research and career avenues (Benton, 2012; Crane, 2013; Nguyen, 2009), underpinning what is now often referred to as the AIDS industry (Altman, 1998).

In 2003–2004, San Francisco was a leader in the provision of social and medical services to people living with HIV/AIDS which, by then, was largely viewed as a manageable chronic disease rather than a death sentence. A team of scientists and medical doctors assembled to address a central research question, articulated by one of the principal investigators (PI) as, ‘Why, in a city with the best HIV/AIDS services, were people still dying of AIDS?’ The research team was largely driven by politically progressive and service-oriented researchers and clinicians, working, in part on an anti-racist platform. One of the PIs spoke of a project he had launched in Uganda, where, against what he saw as racist predictions that poor Africans would be unable to adhere to HIV medication regimens, in fact, their adherence was near perfect. In San Francisco, many physicians were reluctant to prescribe HAART to the city’s most marginalised people because they worried that low rates of adherence might lead to increased public health risks, especially through the development of viral resistance and the transmission of multi-drug resistant viral strains. This team wondered if a form of modified directly observed therapy might work amongst San Francisco’s most marginalised groups.

An excerpt from the adherence study protocol describes the problem:

Decreases in mortality due to ART are relatively modest among injection drug users (IDU), African-Americans, women, and other marginalised populations. As a result, the death rate among the HIV-positive urban poor in (San Francisco) is comparable to their death rate prior to the advent of ART … The ongoing high mortality rate is probably due to low rates of initiating and maintaining ART (Pilot study protocol).¹

These ‘relatively modest’ decreases are in contradistinction to relatively significant decreases among ‘HIV-infected people in the general population’ (Pilot study protocol). The practice of public health relies upon making distinctions between populations or segments of populations, often in the language of numbers, in order to identify sites of intervention. In this case, the homeless, the non-adherent, the African American, the woman, the HIV-positive are constructed as both deviant and dangerous, since they pose a risk to the general population as both vectors of disease and as contributing potentially to the development of viral resistance.

While the larger project was generally successful, this is a story of one of the program sites that was closed because it was determined to be failing. A program envisioned as one of social justice and anti-racism came, nonetheless, to reinforce racial tensions. No team member would have advocated the withholding of effective HIV/AIDS treatment. In fact, they were actively seeking ways to prescribe effective treatment, but found themselves in the quandary of providing and subsequently withdrawing treatment that appeared to be worsening client health. They seemed both sensitive to, and uncomfortable with, the racial component of the quandary. In the university’s meeting room, and in strictly medical and numerical terms, the proper course of action was clear; the program was not improving, and potentially harming, public health. The client perspectives and experiences, however, were not considered in the calculus of programmatic assessment and the medical mandate of doing no harm.

In 2003, I arrived at the program site we call BAP, aware that there was already talk of closing the program. Shortly after my first visit, a program extension was granted for another six months. My fieldnotes recount my first impression.

It (the program) is housed in an old brick building that used to be a police station. In fact, the sign on the front of the building is still, ‘Bayview Police Station.’ Underneath is a painted sign indicating that it is a Health Center.

Nayan Shah (2001) documents how public health policies and discourses shaped processes of space and race construction, specifically of Chinese as epidemiologically and sexually dangerous outsiders in San Francisco’s Chinatown during the eighteen and nineteen hundreds. While processes of racialisation are distinct, Shah’s analysis draws on a shift in governmental political approaches from disciplinary measures to those of self-government and self-care that the BAP site both belied and defied (Foucault, 1995). The police station, now repurposed as a site of clinical and social services, stood as an architectural reminder of links between carceral and medical modes of disciplinary power. BAP clients lined up in the former police station to collect their $10 stipends from the employee who – still – interacted with the clients from behind a barred window.

Nonetheless, this twenty-first century adherence project, we suggest, was ultimately interested in inculcating a sense of responsibility and self-care in its clients. The program relied heavily on numbers; it used epidemiological numbers to identify groups and places to target, and its methods were organised largely on the counting of pills and blood counts. The numbers indicated that BAP clients were not, in fact, doing as they ought. This was a problem.

Writing in 1903, W.E.B. Du Bois addressed the question, ‘How does it feel to be a problem?’ (1994[1903], p. 1). He used the concept of double consciousness to describe, ‘this sense of always looking at one’s self through the eyes of others, of measuring one’s soul by the tape of a world that looks on in amused contempt and pity’ (1994[1903], p. 2). Joseph and Golash-Boza (2021) characterise double consciousness, in part, as a way to theorise African Americans’ status as legal citizens but sociocultural non-citizens at that time. Although Du Bois expressed a refusal to answer the question directly (‘I answer seldom a word?’) he offered a sense of the cultivation of the Black self as a subject always already aware of their status as other, and as a problematic object. We do not suggest that a century later, in a San Francisco neighbourhood, things were the same as in Du Bois’ America of Jim Crow and public lynchings (although it is worth noting that it was only in 2020 that a federal bill making lynching illegal was finally promulgated). We do suggest, however, that sedimentations of a kind of Du Boisian double consciousness remain alongside a longstanding familiarity with being regarded as a problem. These appeared in clients’ understandings of the closure of their public health program in relation to longer histories of neglect and experimentation.

Place: Bayview Hunters Point

BAP is located in the Bayview Hunters Point (BVHP) neighbourhood, the place that James Baldwin, during a 1963 visit, called ‘the San Francisco America pretends doesn’t exist’ (Moore, 2013). Forty years later, BVHP remained multiply marked as problematic. It existed in the city’s imagination as a space of crime, disease, poverty, and racialised distress and danger. In a city famous for its breath-taking landscapes, its iconic bridge, and its self-perception as being on the cutting edge of so much, BVHP was an aberration. In a city that prides itself on its diversity, tolerance, and worldliness, BVHP was clearly at the margins. BVHP evokes Meyers and Hunt’s global health argument that precarious American cities like Detroit ‘should count as part of the global South, not only in statistical but also in historic and affective terms’ (2014, p. 1922). Even as gentrification spread throughout the city, this neighbourhood remained somewhat ‘protected’ by its reputation for poverty, crime, and environmental hazard.

BVHP (combining the South-Eastern Neighbourhoods of Bayview and Hunters Point) is San Francisco’s only historically majority Black neighbourhood. It developed into a thriving community when tens of thousands of African Americans migrated to San Francisco during the 1940s-50s as part of a large-scale migration from the rural South to the industrialised North and West (Dillon, 2014). The US Navy’s expansion of the BVHP shipyard for wartime production drew thousands of such migrants, effectively tipping the racial balance of the neighbourhood. However, much of the economic base disappeared when the shipyard ceased active operations in 1974 (Brahinsky, 2012). Bourgois and Schonberg (2009), highlight how the job losses and broader economic restructuring that resulted from the closure of the shipyard shape the lives of people experiencing homelessness in neighbouring ‘Edgewater’, as they also do in general for those living in BVHP. The shipyard was subsequently designated as a Superfund site, a result largely of nuclear contamination from the Naval Radiological Defense Laboratory as well as a range of toxic pollutants from both the Navy and subsequent companies sited there. Approximately 60 ships were decontaminated at the shipyard after atom bomb tests, and reports suggest that the lab was also used to study biological and chemical weapons (Fimrite, 2015).

Disproportionately high rates of various health problems show the cumulative and enumerated effects of poverty, racism, and toxic environmental exposure:

African-American women in BVHP have breast cancer and cervical cancer rates that are twice the average rates of women living in the entire Bay Area (all nine counties surrounding the San Francisco Bay) and are among the highest in the entire country. Among all ethnic groups, African-Americans in San Francisco have the highest rates of ischemic heart disease, stroke and all types of cancer. (Hunters Point Family, n.d.)

Significantly, some of these chronic and acute conditions overlap with HIV/AIDS and COVID-19 risks and outcomes.

In 2000, in one of the country’s most expensive cities, 30% of BVHP families earned less than $10,000 per year, and the median household income was $29,640.² More recently, environmental clean-up and gentrification seem to offer some promise of revitalisation, and new buildings have been built following nuclear remediation (Fimrite, 2015). Given San Francisco’s extremely expensive housing market, however, such changes threaten to make the neighbourhood too expensive for many of its long-time residents. The 45% drop in BVHP residents identifying as African American from 1980 to 2020 (Brinklow, 2020) alongside high homelessness rates in the area indicate this is exactly what is happening. Bringing this fraught history into the present, a 2018 lawsuit representing 40,000 BVHP residents charged that Tetra Tech, the company contracted by the U.S. Navy to clean up the toxic waste, had falsified records and left radioactive material in the ground. This placed many residents in a conundrum; they could advocate for their health and put themselves at risk of displacement through further gentrification or stay in a contaminated environment that is nonetheless home (Connolly, 2021).

Number worship

Across town from BAP, at another program site located in the heart of downtown San Francisco, I met Max, a forty-year-old Black man who had been going to the program for three years.³ Site administrators told me that Max was the star client. When the administrator prompted him to tell me how he was back when he first started at the program, he told me he was in pretty bad shape. He elaborated, ‘Oh yeah, I had a beard and all that.’ The site staff found this depiction funny because, as I learned later, the beard had been the least of his problems. Apparently, Max had been in terrible shape, very smelly with a heavy lice infestation. Three years later, he was healthy, clean and beardless.

I asked him what motivated him to take his medications, and if the $10 incentive payments made any difference. ‘No,’ he said, ‘it helps, but it doesn’t make the difference.’ What made the difference? ‘The numbers,’ he exclaimed emphatically and without hesitation. ‘To see the numbers keep going up.’ His CD4 counts had become the gauges of his success, the symbols of his future. ‘Now’ he said, ‘I think more about being around tomorrow’. (fieldnotes 2003)

Max is exemplary of the self-governing ethos of public health. He was the star client because he did as he ought, adhering not only to his prescribed medication regimens, but also to responsible self-management more generally. He kept a daily journal of his diet and took vitamins, both of which he credited with keeping the side-effects of his medications manageable. He thought and more importantly, planned, for the future. A few months after our first meeting, he received his high school equivalency certificate and was planning a trip to visit his parents in St. Maarten. He has become the self-regulating prudential individual taking proper care of the self (Foucault, 2003[1978]; Rose, 2006).

I first heard the term ‘number worship’ at a research team meeting. The team had read my fieldnotes and we discussed the apparent motivation provided by numbers for some clients, like Max. One of the MDs suggested that ‘number worship’ was common, taking on a kind of religious or magical element for some clients. More broadly, numbers in global public health have been taken increasingly as a way to communicate objectivity and facts in the language of science. Counting and statistics have long been central techniques of modern government (Ma, 2020; Rose, 1999; Scott, 1998), even as this has not been always a smooth adoption. Even as numbers show themselves to be particularly suited for decontextualisation and invaluable for the communication of authoritative knowledge, science studies scholars have emphasised that social statistics are never merely neutral statements of fact (Briggs & Martini-Briggs, 2003). Rather, they are deeply tied to the projects in which they are used, the institutions and governments from which they arise, the categories invoked and the conditions of their production (Alonso & Starr, 1987; Kohrman, 2005; Lambert, 2006).

Numbers, even as they are used to represent ostensibly objective facts, also have subjective stakes. Sangaramoorthy and Benton (2012) show that numbers are taken up by those they seek to represent, and by both counters and counted. Sangaramoorthy shows how HIV/AIDS numbers come to mould new ways of ‘being and belonging,’ which are experienced and communicated in the language of numbers – a ‘numerical subjectivity’ such that an HIV client can exclaim, ‘I am 400 CD4 and 250 viral load’ (2012, p. 292). Biruk's (2018) writing on global health metrics shows how on-the-ground information acquires statistical value through social relations which in turn both mediates the production of global health statistics and impacts local economies by influencing power relations. Adams (2016) notes that numbers tell certain kinds of stories and that there are other ways to tell stories and other stories to be told.

This is an account of another such story. Indeed, ‘number worship’ – including its elements of magical and religious thinking – can also be used to describe the research team’s approach to adherence. It is through numbers that the program staff could know and demonstrate program efficacy. This project of medication adherence treatment and research is predicated on numerical knowledge. Statistical measures supported the claim made in the pilot study protocol document excerpted above that certain populations had yet to benefit from significant advances in ART and thus served to justify the research. In this sense, numbers supported the creation of categories of people, divided into those who were and were not beneficiaries of new medications. Specifically, significantly reduced rates of morbidity and disease measures were reported in some groups and not others. This lack of improvement in disease measures among the ‘marginalised’ form the foundation for adherence research, based upon the presumption that improved medication adherence will lead to improved numerical measures of morbidity and mortality, as well as the reduction of the development of viral resistance and transmission.

Furthermore, once constructed as such, the non-adherent individual becomes both subject and object of calculation, measured by: pill counts, viral counts, CD4 numbers, rates of hospitalisation, and others (Bamberger et al., 2000). Adherence is measured in various ways including by counting pills and translating this count into a percentage rate that is later compared with numeric counts from the individual's blood samples. Numbers, in effect, are crucially linked to all aspects of the adherence project: in the identification of the non-adherent individual (by counting pills and cells); the conceptualisation of a ‘problem’ population (by counting individuals and categorising them); at the molecular level (by counting cells and viruses). At these multiple scales, cells, bodies, and populations are rendered both knowable and calculable through numbers.

Indeed, it is through numbers that stories about the clients are made and understood. These numbers underpinned the decision to close the BAP program site; they demonstrated that the program was not working even as clients’ reports indicated an opposing truth. Clients’ experiences were silenced by numerical truths: blood counts over time, hospital visits, deaths. But the clients insisted that there was another story, one in which the program did work. In this alternate story, the closure comes to be understood not as a matter of scientific numbers but as one of historical and ongoing racialised discrimination and experimentation.

How does it feel to be a problem?

At the BAP program site there was no star client that administrators insisted I meet. At the research team meeting, the BAP clinic numbers told a story of non-adherent subjects who, collectively, were failing to improve. This was a story of individual, collective, and programmatic failure. At BAP, however, clients told a different story. There, the story of programmatic failure was read in both racial and spatial terms, and clients’ stories showed their interpellation as a problem, but also as not necessarily a problem worth resolving.

When I arrived early one morning, two of the older male members of the group were already there. Wilson was reading a neighbourhood newspaper and he commented on a story about a political prisoner. He told me the man was arrested and locked up at San Quentin Prison in the 1970s. ‘We all were,’ he said, ‘we were incarcerated together.’ Wilson said that the prisoner had been found not guilty by a jury, but was in prison anyway and had been for decades. I asked, a bit sceptically, for clarification, ‘You mean he was acquitted, but is in prison anyway?’ He said without hesitation, ‘Yes.’ He let out a low, gentle laugh and said, ‘See, you’re surprised by that, but it doesn’t surprise us. We’ve lived in this neighbourhood for fifty years.’ Fifty years of injustice, incarceration, poverty and neglect.

I asked how they felt about the program closing. Wilson responded with a rhetorical question of his own, ‘Yeah, why are they closing this one?’ He continued, saying that they are always closing programs ‘down here,’ that they often open programs in BVHP, then close them down. He showed no anger, but rather a kind of calm resignation. He said they use this area as a testing ground for programs then shut them down. In this moment, I find myself thinking of the notorious Tuskegee syphilis study conducted by the US Public Health Service, in which African American men were deprived of treatment in an effort to study the natural course of the disease, and the longer history of medical experimentation on African Americans (Washington, 2006). Wilson continued, explaining that if the programs work, then they open them elsewhere. In Wilson’s view, the neighbourhood was treated as an experimental site for social and clinical programs which, if they work, are then exported to other neighbourhoods – higher priority areas, areas where people count. This is what it means to live in this neighbourhood for fifty years. A hundred years after Du Bois’ The Souls of Black Folk, is this what it feels like to be a problem?

At the meeting that followed this exchange, John, the facilitator, reminded the group that the program was being discontinued. He explained that the group would continue to meet on Mondays at a new site, but there would be no $10 stipends and they would have to arrange to pick up their medications elsewhere. There was grumbling throughout the group: ‘I just want to get my meds. It ain’t about no ten dollars!’ And the question, already asked so many times, hung uncomfortably in the air, ‘why are they closing this one?’ But no one seemed to expect an answer.

One of the staff members later acknowledged to me that, ‘Of course, Jason (the MD overseeing the program closure) is just looking at the numbers.’ I already knew that this was Jason’s frustration; the numbers showed no benefit. He worried that the stipends went for crack cocaine, thus contributing to harm and violating the first principle of clinical practice.

It seems likely that some clients did use their stipends for crack. Gina received her supply of ‘Boost’ – a high calorie nutritional drink – at the clinic. She bragged to Ben, another client, that she had managed to get an extra case, which she had sold for two rocks of crack. Ben also saw another client exchange his $10 on the corner in what looked like a drug transaction. After one meeting, I accompanied David as he used his $10 to buy a monthly bus pass. He had wanted to buy one a few days earlier but did not have the money. His stipend had bought him a month of transportation. The previous week he had used the money to buy himself gloves. It was November, and gloves were especially important for him because he had diabetes and related circulation problems. I asked David if he thought people would still come to the group even without the stipends and he replied that several had told him they would. When I asked him if he knows what people do with the stipends, he said:

Well, I know four or five use them to buy food, like me. But, I also know four or five use them for crack. Because they come to group and say they’re clean, but then I’ll see them from the bus, buying crack on the corner … Mostly, it’s food, crack, or wine. (fieldnotes 2003)

Another client, Paul, was scheduled to go in for surgery. He was scared but said that, in general, his problem was more depression than AIDS. He would get depressed and stay in which made it hard to stay sober. He liked the program and the people, ‘I like coming the group and I learn a lot from the other members, from their stories.’ Of the $10 stipends, echoing star client Max, he said:

They don’t mainly make the difference, but they’re just nice for people. Once a week they could do something nice, like get a cup of coffee or a sandwich … It was a little thing, but really nice. (fieldnotes 2003)

Even for those who used the stipends for drugs or alcohol, he said, ‘At least they would know that they wouldn’t be sick that day. And at least they didn’t have to get money some other way.’ For Paul, even stipends that were used for crack offered a form of palliative care and relief. A statement that I heard countless times during my visit was, ‘We’re a family.’ The group provided a space of kinship and support. More generally, however, the program provided a space to share experiential knowledge, and a community in which to share stories and care. In these ways, the clients reported that the program was working, even if not by the numbers.

COVID-19 – still a problem?

Twenty years later, COVID-19 has disproportionately impacted BVHP communities. As with the HIV adherence study, public health numbers are used to identify groups and sites perceived as being in need of intervention. These numbers can also help to identify health disparities and, indeed, the government website notes that its COVID-19 maps reflect institutional racism and structural inequalities (SF.GOV, 2022b). BVHP has the highest cumulative case rate for COVID-19 of all San Francisco neighbourhoods.⁴ This is not explained by differences in vaccination rates as over 90% of the neighbourhood population had received a complete vaccine series (SF.GOV, 2022a). Furthermore, risk factors for COVID-19 morbidity and mortality are overrepresented in BVHP:

Studies have confirmed this local knowledge, revealing that rates of asthma, breast cancer, leukaemia, and chronic diseases are considerably higher in District 10 than in other regions of the city. This is a list that is almost identical to the CDC’s list of ‘serious medical conditions’ that make a person at risk for severe morbidity and mortality from COVID-19. (Whitacre et al., 2021, p. 1398)

HIV infection is also one of these ‘serious medical conditions’ deemed to increase COVID-19 vulnerability (Mirzaei et al., 2020). Explanatory gaps regarding high rates of COVID-19 infection, morbidity and mortality show how enumeration can point to areas that require further investigation, care, and justice (Kang, 2021). However, it can also miss important social and scalar dynamics and may have unanticipated impacts given the sometimes complex interplay of care with surveillance and social control (Ma, 2020; Shah, 2001; Zola, 1972). In the current pandemic, tensions between researchers and residents persist.

Early in the COVID-19 pandemic, residents complained that their hard hit neighbourhood was not receiving adequate attention. In an open letter to the mayor, they asserted that this lack of attention continued the longstanding pattern of neglect in the area:

While the City of San Francisco has taken dramatic steps in the face of the COVID … very little of the response has been directed towards the Bayview. This is part of a longstanding pattern of ignoring the Bayview unless there are condos to build or a sewage treatment plant to locate. (BayView News & Views, 2020)

Herein, residents draw attention to spatial and racialised neglect and, in doing so, evoke narratives similar to those of the BAP clients twenty years earlier.

Subsequently, researchers arrived in BVHP. Public health projects require specific forms of data to inform decision making and resource allocation and, as with HIV/AIDS, progressive researchers and BVHP residents had different priorities. Residents objected that the proliferation of data collection to meet research goals failed to align with their needs. Whitacre et al. (2021) recount a meeting between researchers and BVHP community members that was held to determine where in the neighbourhood to locate a COVID-19 study and testing site:

One attendee remarked, ‘We don’t need the coronavirus map to tell us the problems that we have in the community because it’s the same map that shows diabetes, it’s the same map that shows unemployment, it’s the same map that shows, you know, disconnection to the rest of the city because of the poor transit. It’s all the same information.’ (Whitacre et al., 2021, p. 1400)

Maps, like numbers, have long served as central technologies of governance and public health (Koch, 2014; Scott, 1998). While critically important for certain kinds of information and representation, both maps and numbers often elide local knowledge and priorities. This attendee refuses the flattening and separability of COVID-19 metrics insisting, rather, on their imbrication in multiple social and biomedical processes.

Given the dearth of local COVID-19 testing sites, this service was much needed, yet the community remained largely ambivalent since they did not share the study priorities: ‘ … what you’re doing is setting something up to serve the institutions and the institutions do not serve or protect us’ (Whitacre et al., 2021, p. 1400). Like Wilson at BAP, residents articulate the difference between counting and caring, between research and resources for the community. The community does not need researchers to study them nor to tell them where the problems are; they need improved access to key resources. Wilson’s perception that ‘they’ used BVHP as a testing site for programs remains relevant in the context of COVID-19 with its seemingly endless research and testing (Hatch, 2022).

Experiences of epidemics and related public health responses are shaped by histories of marginalisation, racialisation and racism. Thus, when COVID-19 testing numbers show higher rates of infection in BVHP, we are reminded of Wilson’s comment, ‘It doesn’t surprise us. We’ve lived in this neighbourhood for fifty years.’ It also doesn’t surprise the current residents who claim that they do not need more research to identify sites of community need. As Whitacre’s interlocutors indicated, they know about the community’s disparate and multiple experiences of poor health and they know how these, along with COVID-19, map onto longstanding structural and intersecting challenges (Bowleg, 2012).

Conclusion

At a 2004 research group meeting for the adherence project, as the closure of the BAP program seemed imminent, an epidemiologist asked Jason, the MD, the same questions that had been circulating at BAP. ‘Why is this site destined for closure? Why does it not work?’ Jason replied, ‘Because people are getting sicker and dying. It seems to be having a negative effect.’ The epidemiologist continued to push, and my impression was that the subtext to his questions was, ‘Why is this program site that serves a primarily African American clientele being closed?’ He said he knows that Jason is uncomfortable talking about this, to which Jason replied, ‘It makes me uncomfortable because people continue to suffer out there and I can’t fix it.’ Later, Jason told me that the site was slated for closure because ‘most of the people, most of the time, got sicker’ in terms of CD4, death, and hospitalisation rates. When I suggested that according to the clients, the program was working, it was doing good for them, he raised the question of cost-efficacy. Perhaps, he suggested, a support group would do as much good for them as the more comprehensive and expensive program. Herein, moral questions of medical helping and harm butt up against other moral questions, such as those of cost-efficacy, medical deservingness, and social justice.

It is through numbers that stories about clients, communities, programs, and public health are often made. The clients at BAP were enumerated, and in a kind of Strathernian accountancy, read in terms of accountability (Strathern, 2000). For Jason, the numbers also held a moral accountability, evidence of doing good or doing harm. Thus, the program and its doctor also were judged, and the closure justified. The clients, the program and the doctor all failed according to the numbers. Enumeration, counting and accountability occur at multiple levels in such adherence programs and include also the moral question of iatrogenesis, itself arising from numerical reasoning. In an increasingly evidence-based world, we are asked to account for our results, to justify our effects and our efficacy, to be accountable – and increasingly, in ways that are measurable.

The adherence program did not work in the way that it was intented. It did some good for the clients, but its closure was read differently by the researchers and the client community. The numbers are important, but they are not uniquely so and this story raises the possibility that even programs that seem to be working well may be doing harm in other registers. Anthropologists are well positioned to attend to other ways of considering subjective experiences and criteria of programmatic ‘working,’ even as subjectivities, especially in relation to HIV/AIDS – and now COVID-19 – are themselves increasingly configured by numbers. Sangaramoorthy’s (2023) recent collection of life histories of older Black women living with HIV in the US clearly shows the complexity and radically different experiences faced by each of these women. Their priorities and lives defy the enumerative logics of global public health and assumptions that might accompany their placement within a public health demographic group. A reliance on numbers serves to silence other kinds of values – like palliative care in the face of addiction – and other forms of evidence found in client stories.

Furthermore, as Hatch suggests, the consolidation of data that erases local and individual contexts enables depictions of communities like BVHP as ‘racial spectacles’ that shape ideologies that may work to maintain, rather than merely reveal, inequities (Hatch, 2022). Such depictions also risk to compound the psychological and experiential insults of implicated individuals. Inclusion must go beyond research programs in order to ensure the sociocultural citizenship that was elusive in the accounts of W.E.B. Du Bois and James Baldwin. In the realm of public health, ethnographic attention can help to identify other ways in which programs might ‘work’ (or not), and remind us that programs are not experienced as discrete. They are not discontinuous from longer histories in the neighbourhood. These accounts from BVHP show that experiences with such programs are shaped by historical relations, such as those that underpin a Black man’s understanding of his program’s closure through the idiom of experimentation and himself as an experimental subject – good enough for the research, but not for the subsequent care that it might produce.

Acknowledgements

The authors owe our deepest thanks, as always, to our interlocutors who share their thoughts and stories with us. Additional thanks go to the research project team as well as to Thurka Sangaramoorthy and Adia Benton who read very early drafts of this paper. More recently, two excellent anonymous reviewers engaged very deeply with the paper and their specific and productive feedback was extremely helpful. We thank you!

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This work was supported by Mitacs; Office of Extramural Research, National Institutes of Health: [Grant Number MH064388].

Notes

1 This is excerpted from an internal document of the research group. It reiterates language from the grant proposal that resulted in funding for the research project.

2 From hunterspointfamily.org (our history) and compared to $65,000 for whites and $55,221 median household income citywide. (Accessed 26 April 2016).

3 All names are pseudonyms.

4 From March 2020 to July 2022 BVHP reported 12,851 cases for a population of roughly 38,000. In comparison Excelsior, one of San Francisco’s most ethnically diverse neighbourhoods with a similar population of 40,000, reported 9,450 cases within this same time period (SF.GOV, 2022b).