Whose Global, Which Health? Unsettling Collaboration with Careful Equivocation

ABSTRACT

The recent push for multidisciplinary collaboration confronts anthropologists with a long‐standing ethnographic problem. The terms we have to talk about what we do are very often the same as the terms used by those with whom we work, and yet we are often doing very different things with these terms. I draw on over a decade of “awkward collaboration” with scientists working in highland Guatemala to explore how challenges of equivocation play out in research focused on improving maternal/child nutrition. In the interactions I describe, epidemiologists undertake ethnography, anthropologists study scientists, and a Mam–Spanish translator works for projects organized around English‐language funding structures and aspirations. I detail situations in which methods, interests, and goals coalesce and diverge to argue for the importance of careful equivocation, a research technique attuned to unsettling binaries that does not result in sameness or unity. I offer suggestions for how this technique might productively reshape the emerging global health imperative to work together. [global health, controlled equivocation, co‐laboring, material‐semiotics, Guatemala]

2016. Xela, Guatemala. “What will we ask them?” María asked me.¹ We were in a café drinking sweet banana smoothies. Eventually, María and I would come to work together closely, but this was our first meeting and neither of us knew what to expect. A friend had introduced us because María was the Mam representative for Guatemala's Academia de Lenguas Mayas de Guatemala (Guatemalan Academy of Mayan Languages) and had worked for years as a translator for regional health improvement projects, which have proliferated alongside Guatemala's postwar, CAFTA‐era poverty (Chary and Rohloff 2015; Maupin 2009; Rahder 2014). I had planned to interview María about her work translating for these projects, but she quickly assumed the role of translating for me and wanted to make sure she understood what I wanted to know from women in the region before we began our work.

“I'm not sure yet.” I answered her, writing down both her question and my response, my phone also recording the conversation.

“Well, what will we do?” she asked while I scribbled.

There was silence as I thought this over, and then I responded: “I'd like to learn from the women who live around San Juan about how the maternal health projects in their communities are affecting them, but I think this [learning] will be difficult. And so I'd like to learn more about these difficulties.”

María leaned over to see what I was writing, so I set my pen down and tried to explain: “This, right now, is part of what I'm interested in. How do projects get set up? How do researchers and health workers know when projects are working well or failing? How do they decide which questions to ask?” She seemed unconvinced. At least, she offered no smile of comprehension. So I tried again: “The data I'm after aren't just out there in the communities. The data I am after relate to the process of assembly and the way this conversation, right now, comes to shape the future conversations we're able to have.”

“So you're investigating me?” María asked, dubious.

“Yes. Sort of. At least, I'm interested in how our conversation now fits into what will be and can be spoken and what—and who—remains silent. Although, really, ‘investigating’ may no longer be a good description of what I'm doing.” After a long pause we switched to an easier topic, but the questions of what I was doing—and how to explain it—persisted in our work together. I return to it below.

****

The recent push for multidisciplinary collaboration (cf. Barry and Born 2015; Benezra 2016; Callard, Fitzgerald, and Woods 2015; Konrad 2012) confronts anthropologists with a long‐standing ethnographic problem. The terms we have to talk about what we do are very often the same as the terms used by others with whom we work—and yet we often use these terms in very different ways. Already I am writing around the problem, having organized a difference through the categories “we,” the anthropologists, and “others with whom we work.” What this distinction is taken to be is at the heart of the problem of crossing disciplinary borders. In the stories that follow, I will tell you about anthropologists doing science and scientists doing anthropology while collaborating in the practices of global health. Yet, even as we collaborate, we often do not have the same goals. This article is about the tension of working together amid various and shifting kinds of differences.

I draw upon more than a decade of collaboration with scientists in Guatemala. I use the term collaboration here less to connote the general circumstances of interaction than to name a specific institutional politics encouraging unity that preconditioned our meeting. In 2006, I received a grant for “international collaboration,” jointly funded by the US‐based Social Science Research Council and the Ford Foundation (an interesting collaboration in itself). The purpose of the grant was to unite social scientists with experts from other fields on topics where there was overlapping interest. The grant emerged at a time when traditional techniques of comparative anthropology were being displaced by a new—and often suspiciously neoliberal (Canfield 2018; Riles 2015)—emphasis on collaborative improvement. I was beginning research into how the diagnostic category of obesity was adopted and refused in Guatemala's highlands. The funding allowed me to spend three months in Guatemala City interacting with scientists at a research center that I will call “Center C,” whose almost entirely female staff includes numerous internationally recognized experts in public health nutrition.² The logic driving the collaboration was as follows: Epidemiologists and anthropologists are each studying the problem of obesity in Guatemala. If we put our different perspectives together, we will have a richer understanding of how to tackle the problem.

But then I did some things that messed with this logic. I took notes about what the scientists were doing in addition to the knowledge they shared, and I asked questions about the context rather than simply the content of their expertise. As time passed, it became clearer (to us all) that my study of obesity in Guatemala encompassed the study of scientific cultures and, by doing so, showed that “the problem” of obesity was not, after all, a singular or shared problem (Yates‐Doerr 2017). On several occasions, the scientists called me a spy. At other times, they positioned me akin to a journalist investigating a scandal. But because I was not writing for a newspaper—and because I was working very slowly—it was also clear they did not quite know what to make of me. “What is it that you are doing?” they wondered, even as I was asking the same question of them.

Amid these uncertainties, we forged interest in one another's work and lives. I ended up basing my research in Quetzaltenango, a state five hours west of the capital. But during sixteen months of fieldwork in 2008–2009, and over the years that followed, I returned frequently to Center C's modest office building to share what I was learning and receive feedback from the scientists. They would ask me questions that would cause me to change the questions I asked of them and others. This was not me “giving back,” which presumes a clear—and colonial (cf. TallBear 2014)—divide between research and researcher; instead, working with them changed the way I listened and what I chose to analyze and write.

After my book on obesity was published (Yates‐Doerr 2015a), I returned to Guatemala for a new project focusing on a maternal nutrition policy that would have me working daily with Center C scientists. This time, with their help, I wrote them into the project's description from the outset. I would not just study Guatemalan understandings of nutrition, nor would I study the laboratory and field life of the scientists (e.g., Latour and Woolgar 1979), but I would study our own practices of collaboration. For indeed, the question of “how to collaborate”—across political sectors and parties and between disciplines of medicine, education, development, and so on—was everywhere. It was a question asked by politicians, by international institutions, by nurses, and by people in communities where the nutrition policy I wanted to study was instigated. It was asked by the scientists in Center C, and now it was asked by me.

Yet it turned out that again we were often not asking the same question. This article offers the idea of awkward collaboration to emphasize the differences that shaped our interactions, and it presents careful equivocation as an alternative goal for working together than that of collaborative unity.

AWKWARD COLLABORATION

In a classic article, Marilyn Strathern (1987) describes the relationship between feminism and anthropology as “awkward.” She notes that both feminism and anthropology are committed to difference, but rather than finding themselves at a point of convergence in this commitment, they find themselves relating to difference in potentially conflicting ways. It is telling that Strathern takes up the theme of collaboration to make the point. She writes that for anthropologists, collaboration is often used to reflect that the Other's input has shaped the anthropologist's experience. The example Strathern gives is the collaboratively written ethnography, which was used by anthropologists to allow the divergent voices that have always shaped ethnographic practice to not become submerged by one another (she calls this a “working ethic of humanism”). Meanwhile, for the feminists she describes, this idea of collaborative writing remains a delusion because relationships are structured by power, and dialogue is unavoidably asymmetrical. The giving of credit and voice become a means of taking them away.

Strathern calls the tension between anthropologists and feminists “awkward” because these are not simply two analogous communities to be compared; feminists and anthropologists are relational communities who influence each other as they interact. As she writes, the tension between them is one “that must be kept going” rather than a tension whose aim is reconciliation (1987, 286). She would later write in Partial Connections (2004) that the awkward relation can pose a problem for those who see difference through a binary logic, in which a definitive choice between sides must be made. What she pushes us to see is the feminist‐anthropologist's mode of relating, which acknowledges the productive aspects of difference.

Marisol de la Cadena (2015) mobilizes concern for productive difference in her analysis of the co‐laboring of translation. Co‐laboring, as she frames it, entails working besides one another, but not necessarily for the same ends. The labor of co‐laboring reflects the effort that it takes to attend to spaces of difference without collapsing these differences upon themselves. In co‐laboring, instead of focusing on what is “lost” in translation across languages, the focus becomes holding onto the generative effects of not understanding.

Concern for colonial erasure motivates de la Cadena's work. She notes that making unlike things equivalent—or even comparable through the same scale of valuation—often serves as an exclusionary tactic to silence “the Other” (2015, 27). Translation, when treated as a process of making differences equivalent, functions to make nondominant languages and practices disappear. Moving away from attempts to communicate univocally, she rather encourages the strategy of communicating by differences. To make this argument, de la Cadena draws upon Viveiros de Castro's (2004) discussion of “controlled equivocation.” Whereas the Euro‐American project of translation has sought “to find a synonym” between words, the kind of translation Viveiros de Castro is after seeks to “avoid losing sight of the difference” concealed within equivocal homonyms. His ambition is not the discovery of a common referent (the quest to ascertain meaning that has underpinned the field of anthropology) but rather to make equivocations explicit.

Controlled equivocation is an especially useful technique for my work at Center C because the scientists and I have so many keywords in common—maternal health, nutrition, obesity, development, culture, and their various Spanish counterparts³—even while we spend much of our time together looking over another's shoulders, noting that we do not use these terms in the same way (see also Hardin 2015). Yet, when I went to practice the technique of controlled equivocation in my own work, this technique of comparison fell apart. I could not control equivocations, which evaded ethnographic capture.

Viveiros de Castro is clear that equivocation does not “belong to the world of dialectical contradiction” (2004, 11), but he still leaves his readers with the impression that the control of misunderstanding is possible. Even while his writing is playful—for example, he says controlled equivocation “is controlled in the sense that walking may be a controlled way of falling”—the language of control risks reinstating the binary logic of an “us” who is against a “them.” Setting aside the ambition to control equivocations by emphasizing how binaries are heterogenous and mobile, I offer a mode of collaboration based on a feminist commitment to careful equivocation.

If equivocation orients us to difference, caring for equivocation asks us to attend to how differences will themselves move and restabilize. Caring for equivocations is not a radical departure from the work of Viveiros de Castro, but caring for instead of controlling equivocations places attention on how equivocations will transform through awkward collaborations.⁴ Crucially, this does not suggest that differences cease to matter—or that all differences transform in similar ways or at similar speeds. It suggests, rather, that analysis of equivocations must be continuously readjusted to the context at hand. Determining how to situate context and how to grapple with actors’ heterogenous positionalities requires the engagement and temporal attention asked of “care.”

Care, as Annemarie Mol uses it, relates not to kindness or compassion but to a willingness to eschew generalities so as to attend to the situated contexts in which some problems come to matter as others come to disappear (see also Pols 2015). Mol develops her thinking around care through long‐term observations of the “tinkering” in medical practice in the Netherlands. She notes that care is not a matter of control (Mol 2008, 39). This is not a sentimental claim, but one culled from extensive participant observation in a Dutch hospital, where she learned that the ability to be responsive to illnesses’ contingencies was crucial for good doctoring. She draws from doctoring to offer a caution for the practice of anthropology: the aspiration to control, she warns, risks producing harmful theory. “Theory,” in her framing of care, is something to be handled not something to be resolved. Adopting her approach to care, I propose that caring for equivocations might be a useful way forward for anthropologists—and, perhaps, for others engaged in the shifting fields of global health.

Insofar as careful equivocation offers an intervention into the anthropology of global health, it is one that acknowledges that a strength of this domain of scholarship lies in its commitment to holding itself open—not knowing at the outset of fieldwork what counts as “global,” “health,” or their disciplinary conjuncture “global health” (e.g., Brada 2011; Fassin 2012; Pigg 2013). One aspect of the care work of careful equivocation entails teasing out which interests should be emphasized among the awkward collaborations of medicine, obstetrics, humanitarianism, development economics, finance, or other kin‐making or life‐ (or death‐) making fields that anthropologists are likely to encounter. A second aspect of the care work lies in a temporal commitment or engagement with others as cross‐disciplinary interactions unfold. To put this otherwise, this is a technique of taking fieldwork as a practice of care work, while caring comes to both constitute and transform “the field.”

UNSETTLING TERMS

At the start of 2016, I arrived in Guatemala to spend three months in the highland state of Quetzaltenango. I was studying an initiative to improve global development by improving fetal and early life development. The initiative was in the process of being crafted by scientists and policy experts who were mostly based in urban centers, and I had been tracking these activities online, at conferences, and through policy documents. Now, in Guatemala, I wanted to learn how policymakers’ plans were unfolding in a rural region where I had previously worked.⁵ Before traveling to the highlands, I presented my own research plans to Center C's scientists in Guatemala City to get their feedback.

Early on in our conversation, I explained that I was interested in the question of how “global health”—a term that is not locally intelligible to most Guatemalans in its English form—has come to drive so much of the aid and development work in the region. I offered the example of the nearby offices of the Pan American Health Organization (PAHO), referring to this as a “global health” organization. In applying this term, I had been considering how PAHO's governance structures have developed through international networks: it is today part of the United Nations, holds the title of the Latin American affiliate of the World Health Organization (WHO), and advocates a “universal right to health,” as do many health organizations that operate globally (Farmer et al. 2013). I was also relating to the writings of historians and social scientists who note how UN health organizations, which had been previously organized to coordinate between different national interests, now coordinate a dizzying proliferation of philanthropic agencies, nongovernmental organizations, and pharma/nutraceutical industries (e.g., Biehl and Petryna 2013; Quirke and Gaudillière 2008). I was further thinking about trends among policy experts who work in global health's paradigmatic “resource‐limited” spaces (Brada 2011, 286) to assess health through the indicator of childhood stunting—as defined by being two standard deviations below normal growth. The World Bank had recently deemed Guatemala's stunting rates to be the third worst in the world, and PAHO was responding by incorporating stunting‐prevention strategies into its action plans. It seemed obvious to me that PAHO was involved in the production of “global health.”

I was therefore surprised when the Center C scientists objected to my use of this term. They explained that “global health” was a title belonging to activities associated with the central branch of the WHO and not to PAHO. One person in the room tightened the parameters even further, claiming “global health is Africa.” To clarify this position, they pointed me to dates of origin. PAHO, whose founding charter responded to the need to keep workers who were building the Panama Canal from catching malaria, has existed since 1902. When the WHO was founded in 1948, PAHO had already been tasked with caring for Latin American health for a half‐century. The scientists explained that when it came to making health policy, the two organizations have never quite known how to work together. They wanted me to recognize how health policies designed around Africa might fail to address the health concerns of Latin Americans. They also wanted me to see that “Latin American health” predated “global health,” so as to re‐center Latin American peoples in the history of global development.

A semiotic analysis of the conversation might take this as an occasion of misunderstanding: for me, “global health” named an expansion of health governance; for them, it named an ongoing exclusion. Yet a potential problem with the framing of misunderstanding is the risk of keeping alive the possibility of a correct understanding and with this granting authority to a single definition. A virtue of Viveiros de Castro's “controlled equivocation” is that it upends the aspirations to a single, correct meaning. As he explains, an equivocation is not an error or an illusion. The opposite of misunderstanding is not understanding but univocality and its compression of differences into similarity.

As I co‐labored through these exchanges, however, I found that different meanings did not only resist being compressed into similarities; they also resisted being stabilized into sides. For example, when the scientists interrupted my use of the phrase “global health,” they were not telling me to take their term as my own.⁶ Nonetheless, this act of interruption offered a space for learning, and with this, unsettled my terms. When analyzing global health agencies and their granting structures since this encounter, I have noticed how many leave Latin America off their agendas while claiming to do global work, and I have begun to consider how my own use of the term (in conversations with other experts, in grant applications, in articles, in teaching) might resist or exacerbate these exclusions. In turn, one Center C scientist began to use air quotes around “global health” when using the term in conversations of which I was part—signaling my presence in the room, at least, if not the term's instability.

Working to care for, rather than control, our equivocations reframes the units of collaboration (the “me,” the “them”) as awkward. Drawing from Strathern's “awkward relation,” I use awkward collaboration to imply an uneven and unfinished interaction across differences that changes the relations between those in a place of collaboration. There are many such awkward collaborations in the conversation above. There are those between the (not‐so) global interests of the WHO and the Latin American interests of PAHO, who work together while at odds. There are also those between PAHO's policy and Center C's science; although PAHO is explicitly charged with forming health guidelines and Center C focuses on research (and is frequently critical of PAHO's activities), the institutions maintain close ties, often presenting their research together in international conferences, debating and learning from one another even though they do not always agree. Then there are awkward collaborations between me and the scientists at Center C. We are neither entirely separate nor entirely together as we shape and draw from—without quite dissolving—each other's terms. Referents are thought to hold steady, but in these awkward collaborations, we find them changing forms.

In contrast to the focus on commonality that drives many of global health's political, scientific, or even theory‐making projects, a focus on equivocations produced out of awkward collaborations helps to maintain attention on the differences that continue to persist while working together. This attention to difference is especially crucial given the asymmetry of many global health projects, where some actors have more (political, scientific, theoretical) power than others—and where an emphasis on “unity” routinely becomes a way of eliding or ignoring those (typically with less political, scientific, or theoretical power) who do not fit. Also crucial, however, is that a focus on differences risks pushing nondominant actors to the outside of systems in which they play an active part. Showing collaborations to be awkward is a means of honoring differences between actors while also refusing to lock these differences in place.

ARTICULATED FUTURES

Often, when I do the work of co‐laboring on the spot—pointing out variation in seemingly like terms that arises in the process of using them—professionals in other scientific disciplines have dismissed this work as immaterial. “Just semantics,” is a common response. Strathern is again helpful because of her insistence on the double work of articulation, being at once semiotic (articulate: to speak) and material (articulate: to join together) (see also Choy 2011; Haraway 1992; Langwick 2008). In the section above, Center C scientists and I are sitting together in a meeting room, in clear dialogue about global health. In the next section, I'll move from dialogue to an analysis of long‐term, interactive engagement. I move from conversation to practice to make explicit how words articulate worlds—to cite a refrain of material‐semiotic scholarship, which points to how language gives shape to materiality. I describe actors in highland Guatemala who seem to share a future goal of “better health.” Yet, by unpacking a series of interactions that span a decade of awkward collaborations, it becomes clear that health is not, after all, a shared project.

In 2009, at the end of an intensive phase of fieldwork, I sent a file that listed the institutional contacts I had made while working in the state of Quetzaltenango to the Center C scientists. These scientists had once introduced me to public health networks in Guatemala City, and now I was in a position to return the favor. My file was passed along to a newly affiliated researcher, an epidemiologist from Canada—I'll call her Anne⁷—who was preparing a study of the effects of nutrition and stress on prenatal growth.

Among my contacts was a health extension project that operated in a mountainous region outside of San Juan Ostuncalco, an area reported to have the country's highest rate of maternal mortality and second‐highest rate of stunting—each a national indicator of poverty. The government had contracted the project to provide health coverage to a cluster of ten communities, each with roughly thirty to seventy homes (Figure 1). I had traveled with its health practitioners for several days a week over a four‐month period in 2008–2009 as they administered weekly vaccinations and medications, doctor consultations, health education, and nutrient supplements to long lines of women who arrived with large US‐flag towels wrapped around their handwoven huipiles, the towels offering an added layer of protection against the cold mountain air. These towels were a nonmetric indicator of poverty; emigration rates to the United States from this region are among Guatemala's highest, and they were a sign that nearly everyone has extended family living in the United States.

Figure 1.

This map shows the houses in one of the communities served by the health extension program, with newborns, unvaccinated women, pregnant women, and women who have recently given birth marked by differently colored pins. (Photograph by author)

My contact with the extension project proved useful for Anne. Over the next few months the staff helped her to gain access to the communities they served. Shortly after I concluded fieldwork, she began a three‐year cohort study that assessed cortisol and anthropometry through the period between pregnancy and six months postpartum—a window global experts have deemed as crucial for fetal and global development alike. The health extension program's employees helped Anne to find a pool of people (now termed “population”) in the communities to study. They also introduced Anne to the region's network of health promotoras. The Mam‐speaking promotoras helped the extension project recruit and monitor women who were pregnant or nursing; they could now help Anne recruit and monitor women for her longitudinal study. Anne would eventually train the promotoras in the particular techniques of her research, but much of the difficult work of explaining why she was there had already been done—at least partially. During the decade of the extension project's operation, women in the communities had become accustomed to outside organizations arriving to their homes with the stated aim of studying and improving their bodies and their health. When Anne arrived expressing interests in health and development that were (or sounded) similar to those made by the extension program, the women allowed her to enter their communities.

Shortly into Anne's research an embezzlement scandal caused the Guatemalan government to abruptly end the extension services. Before long, several NGO's moved into the empty buildings that the extension service had left behind. One organization, Paisanos, which was funded by the UK‐based Save the Children and run by Guatemalan health and development experts, contracted the same promotoras who had previously been employed by government funds. The US‐government‐funded USAID also supported Paisanos by providing promotional materials and a corn‐based protein supplement that was similar to what the government‐funded project had provided (although this new supplement, which looked the same as the previous one, was sourced from US surplus corn). USAID also provided the funding for Paisanos to train teams of técnicos—the title given to the men they hire to run the pop‐up clinics and collect physiological data. Every three months, at a minimum, técnicos entered weight and height measurements of all women and children from each community enrolled in their project into databases. This raw data, as they called it, would eventually be used to produce data that would, in turn, be used to monitor and evaluate their project's impact. If enough babies grew longer and mothers weighed in as heavier, they would call the project a success (see also Olson 2012).

Proyectos, the local term for the broad range of interventions in the region, is used by both Spanish and Mam speakers. The term, which translates readily enough into English as “projects,” captures well the notion that the future—a projection—is in the process of being built (see also Law 2002, 87). It might seem that these various projects are aligned in the effort to improve community health. But when teasing apart the work undertaken by epidemiologists, nutritionists, promotoras, and even anthropologists, it becomes evident that the collaborations are awkward. Even as we work together, the futures we are projecting—the healths that we are seeking—take divergent directions as they form.

Let us consider Anne's work, which began as a PhD investigation of cortisol and then expanded and transformed as it was funded by follow‐up grants. One of these, from a Western government, emphasizes in its mission statement “bold ideas with big impact” in global health and “helping the poorest and most vulnerable populations.”⁸ Indeed, Anne hoped her work would “shed light on important strategies to increase resilience and empower women and communities to break the intergenerational cycle of poor growth and reduced lifetime health and opportunity” (Chomat 2015, viii). This language is resonant—even homonymous, we might say—with that used by USAID, which champions the importance of “fighting poverty to end the cycle of preventable child and maternal deaths so as to change the lives of women and their families.” Both Paisanos and Anne produce statistics of morbidity and mortality. They also employed the same women as assistants while interacting with the same households of marginalized women in their collection of data.

Still, the healthy futures Anne and Paisanos are working toward are not the same. The directors of Paisanos champion the value of unity (Figure 2). “Unity” might very well connote a general condition of comradery or togetherness, but in this case it advances an idea of singularity that allows Paisanos's directors to treat health‐related knowledge as something that can move from place to place, unchanged. The language of unity holds in place a trickle‐down model of intervention, in which a regional coordinator trains several técnicos, who in turn give skills to promoters, who in turn give skills to mothers. There is a relevant equivocation here: the phrase used by Paisanos to describe skill building is dar capacitaciones, literally “to give skills.” This could be translated as “capacity building”—a long‐standing technique of development work promoting community‐driven changes. But the model for dar capacitaciones put in place by Paisanos operates with unidirectional arrows. With singularity as an organizing principle, the content of skill is imagined to exist independently from the context of the skill, such that experts bestow skills upon those previously lacking expertise. The model both reflects and establishes the notion that solutions should be replicable and can be carried out at scale. Lost in the model is attention to the specificities of the health problems faced by people on the receiving ends of the arrows, the sense that people may already have their own expertise, or a recognition of how skills are site specific such that it simply may not be possible to scale them up (Kenworthy and Parker 2014; Yates‐Doerr 2015b).

Figure 2.

Women from one of the San Juan communities gather near a message from the NGO, Paisanos, which reads, “We are Paisano [country‐folk] united for a healthy country.” (Photograph by author) [This figure appears in color in the online issue]

Anne, meanwhile, was critical of a “top‐down and utilitarian approach to health and policy making” (Chomat 2015, 188) and has worked to make the arrow bidirectional by carrying out research that aims to be “horizonal, with both sides contributing equally and operating as equal partners” (Figure 3). I attended one gathering of women she helped facilitate, where she sat at the side of the circle of women and neither interjected nor imparted her knowledge. She considered creating spaces for women to gather together and talk as a success—even without a quantifiable “health” outcome like child length to justify the occasion. “They tell me they feel better having each other's support,” she explained, shifting her attention away from universal standards. She hoped her interventions would encourage self‐efficacy, agency, self‐esteem, communication skills, and mental well‐being, and she placed emphasis on the “outcomes” of play and laughter.

Figure 3.

Women, including Maria (second to the right), gather as part of Anne's intervention project, which is community led and designed to address community‐specific needs. (Photograph by author) [This figure appears in color in the online issue]

While the resonance between her project and other Guatemalan global health projects may have initially made her work intelligible to San Juan's communities, thereby facilitating her entry, it also posed a problem. The communities where she worked were used to extractive research and unfamiliar with the participatory action research frame that had inspired Anne's work. As she explained, “it took a long time to build trust because there was a strong sense that such research never served them in the end. This is something that I have been trying to counter and do otherwise—through participatory methods and long‐term engagement with communities, which includes co‐designing projects and their content.”

Anne and Paisanos share a claim to “global health.” They share a claim to doing science, both collecting and evaluating anthropometric data. They even share the seemingly equivalent goals of improving maternal health and, with this, improving women's lives. But they do not share the same practices. Anne's data was intended for academic journals and a community of expert scientists, and it was often discussed with the communities to assess the cultural acceptability of her work; the data Paisanos collected was taken to far‐away experts to evaluate the validity of their ongoing work (if people grew significantly taller, the projects would continue). Anne's women‐led health initiatives entailed facilitating women's gatherings and used a method of listening to women to encourage change in their lives. Paisanos teams of male técnicos sought to deliver education and supplements in the hope of making them taller and healthier. Their different activities, though both undertaken in the pursuit of “a better future,” articulate different kinds of futures.

BINARIES IN RELATION

This brings me, finally, back to my conversation with María, for the awkward collaborations in the project of studying and intervening in maternal health projects do not just happen between the scientific and political projects I have been following. They also extend into my project of following them.

Six years after Anne first arrived in Guatemala, as she was concluding data analysis, I returned to Guatemala to learn about the maternal health projects taking place in the region where she was working. If Anne had once made use of my fieldwork contacts, I was now making use of hers. This was not only because I planned to study maternal health projects carried out in the San Juan region where I had once worked, which now included her research, but also because I employed one of her primary field assistants to do so.

Prior to beginning her research, Anne had enrolled in Mam‐language classes, where she became María's pupil. Though María was a first‐language Mam speaker born in Guatemala, she was, in some ways, a foreigner as well. Her family had fled Guatemala's escalating violence in the 1970s, leaving their small town in the northwestern state of Huehuetenango when she was young. As with many refugee children in southern Mexico, she spent her school‐age years in and out of hiding. She returned to Guatemala, setting up residency in Xela—a new city for her—after the Peace Accords were signed in 1996. Lacking the proof of schooling necessary for higher education, she made use of her linguistic skills, which she could demonstrate without any documentation. When she met Anne, she had worked in Quetzaltenango for many years as a translator for regional health projects. Before long, María was traveling regularly, and often independently, to recruit and monitor the participants of Anne's research.

When I called María, she was without employment. Anne's three‐year study had ended and other projects in the region required that técnicos drive motorcycles. María didn't mind the extra time it would take to bus and walk to rural communities, but like most of the women in the region, she did not have a motorcycle license, and the projects that were hiring would not change this requirement (they said the rules were to protect her; she pointed to the obvious sexism).

At first, I had simply planned to interview María about her work, but she saw before I did that she'd be a valuable assistant, and we soon fell into an easy agreement: I'd pay her what Anne had, and we could return to many of the communities they had visited before, where she would help with matters of translation. The apparent ease of this setup was, however, destabilized well before setting out because what I wanted to do with María was so unlike what she had done with Anne.

Anne, as with many global health experts, holds many different disciplinary alliances (e.g., Kim et al. 2002). Whereas the methods of her health interventions prioritized listening to people, her work as a public health scientist required measurement. Her cortisol study evaluated causal relations, using statistics to ascertain whether maternal cortisol mediates the vertical transmission of what she referred to as “stress” (a word without a corresponding term in Mam). To account for stress, she delimited three types of stress experience: nutritional stress, infection‐based stress, and psychosocial stress. She then transformed each of these types of experience into measurements. For example, she counted the number of people that fell into a respondent's social network, calculated a dietary diversity score, translated group membership into a binary variable, and evaluated social harmony and trust through a thirteen‐item questionnaire. When María traveled with Anne, there were participants to check up on, illness to catalog, urine, saliva, and feces to collect, and the length and weight of mothers and babies to track.

Meanwhile, I had neither measurements to record nor surveys to fill out. I asked María to help me collect something arguably far more intimate than blood or feces: I wanted stories. Here, however, is yet another site of awkward collaboration. Anne, in her work as a public health practitioner, was also interested in stories. As she writes, “critical to the goal of improving health of marginalized populations is an improved understanding of their lived experiences and how they define and understand notions of stress and wellbeing” (Chomat 2015, 187). Like many people working within public health, Anne was versed in anthropological texts and social theory, citing Paulo Freire and using the language of decolonialism in her own publications. Acting on her commitment to understand “the local point of view,” Anne facilitated monthly meetings where women would discuss, in their own languages, their problems while Anne mostly listened.

The apparent similarity again risks masking a divergence. Anne eventually worked to translate the women's beliefs into English. To do so, she would record the conversations, later working with María to make transcriptions, which would be coded into thematic topics through the words vulnerability, resilience, nature, animals, tradition, and so on. She counted the frequency of these words, later reporting these numbers. Meanwhile, the stories that I was after were not so much their stories but rather our stories. Where Anne sought stories about the women, I sought stories about practices of equivocation. At a moment that many health workers were placing emphasis on the discourses and practices that they have in common, I wanted to follow the feminist‐anthropologist tactic of paying attention to what is not shared, what may never be shared, and what may even be co‐opted in the name of sharing. My focus was on the productive conjunctures between policymakers, scientists, care professionals, and mothers as they variously handled the technologies, from scales to supplements, entering their lives.

It was my first impulse to critique Anne's deployment of coding and measurement as harmful toward Indigenous systems of thought. Initially, I objected to how complex social worlds were stabilized into statistical enumerations, with statistics standing in for, re‐presenting—and, I feared, thereby displacing or even erasing—these worlds. I also objected to how quantitative epidemiology treated women's conversations and bodies as sources of health data, studying them as if the science was not part of this practice (see also Ceron 2018).

Undertaking the practice of co‐laboring has helped me slow down this impulse to criticism. Recall that co‐laboring entails the labor of attending to differences that arise in projects that seem, on the surface, to be shared. This labor requires not only attending to differences in seemingly similar words that are spoken but also the audiences for whom they are spoken and the practices in which these words are embedded. It is, in other words, the labor of attending to communicative contexts. To co‐labor with Anne was to pay attention to whom she was speaking and how she became attuned to how her words were taken up or, alternatively, failed to travel. The technique is well suited to the awkward collaborations of cross‐disciplinary work, where some methods and languages come to dominate others even as they might seem to be unified. It can also be used to draw attention to divisions within a discipline that might otherwise appear stable.

For example, I would learn by watching Anne interact with her different global health audiences that she explicitly used the epidemiologist's conversion from story to number to challenge existing agendas of global health, which routinely deprioritize maternal health (Storeng and Béhague 2016). She measured the bodies of Mam women, but she also listened to them. Over years of interaction in their communities, and in conversation with other scientists, she developed the position that improving nutrition typically required improving social and health infrastructures—not making better supplements.⁹ She deployed the reductionism offered by numbers not because her numbers were true in an objective, disinterested sense of the term but because they were true in the pragmatist sense of being useful. Converting Mam to English and again converting English to a number (for example, “X percent of people report symptoms of stress”) was a strategy to convince a global health audience that “nutritional health” related more to women's social ecology than to isolated nutrients. As she writes, numbers were important for her project of improving women's conditions because numbers “allow local needs to be represented and spoken for at a national or global level” (Chomat 2015, 208). Numbers made people in political and institutional power—people whom she wanted to listen to her work and people whose work she wanted to change—pay attention. Although she couldn't yet expect a seat at their table, by articulating her concerns through a language that seemed to be theirs, she might get them to at least crack open the door.

In this way, her numbers did not—at least did not only (Taguchi 2017)—report upon how the world was but were tools to be deployed to impact an interested audience. Though I make use of the power of concepts more than I do with the power of numbers, I was familiar with this move. Recall my first meeting with María, over banana smoothies, when I had spoken with her about the data I was interested in: “The data I want to collect isn't just out there in the communities. The data I am after pertains to the process of assembly and the way this conversation, right now, comes to shape the future conversations we're able to have.”

A dictionary would tell us that data comes from Latin for “the given.” But here I was using the term performatively. I was trying to fit my interests into a frame of science that María had learned to expect through her work with other health researchers, where data lies out there in the world—raw, waiting to be collected by a disinterested investigator—while at the same time unhinging the term from its definition so that it could become something else, something interested, dependent, wily, and social. Despite attempts to locate meaning in etymology, not even the concept of the given is given (Morita and Gergely 2013).

As with my “data,” Anne's statistics were at once reductive and performative, allowing her to speak in a language that is partially intelligible to the global health community while at the same time shifting the terms of the conversation. “My work is an intervention,” Anne once told me. This was familiar. After all, anthropology has long wrestled with the fact that the stories we tell are not neutral but performative (Dunham 2002; Hurston 1942). As with Anne's statistics, the anthropologist's story redescribes so as to intervene. Anne's statistics seek to make a difference; my stories also seek to make a difference—in this case, by probing the very conjunctures of difference. In a moment in which global health policymakers routinely draw from what is shared to “form the basis for collective action” (cf. Clarke and Star 2008), my stories of awkward collaboration hope to inspire the call for the development of modalities for action, and especially collective action, that are not preconditioned on sameness but which recognize, allow for, and have the capacity to even welcome dissimilarity. As a method, it not (only) asks to be included in many of global health's political processes but also asks if there are other, better ways of doing politics.

CAREFUL EQUIVOCATION

The feminist commitment to care for objects, categories, and terms as they are done in practice unsettles the binary often drawn between one object, category, or term and another. After all, when binaries are shown to be relational, they are at once binaries and, by making each other move, not binaries at all (Sanabria 2016). When putting Anne's work alongside mine, her concern for measurement and coding is striking. When putting Anne's work alongside the work of USAID, what stands out is her focus on people's stories. It might be tempting to see this as relativism—in which meaning develops relative to an individual's (or group's) own worldview. But what I have pointed to is how we operate through engagement with one another, conversing in different languages in ways that also transform one another's terms—and, with this, one another's practices, as well as the very constitution of “the Other.”

Viveiros de Castro (2004, 20) posits “controlled equivocation” as a means of treating difference as a condition of signification and not a hindrance to translation. I share his desire to do away with translations that act as if they rest on a natural referent, but I have suggested that careful equivocation might be a better way to approach the relational and temporal space of awkward collaborations in which “sides” refuse to stay put. Drawing on well over a decade of grounded engagement with an ever‐changing field of fields, I have worked to articulate the lively sociality of our disciplines alongside our terms and practices, showing how the differences between actors are emergent and slippery—however unevenly so (Law and Lien 2012; Roberts 2017).

To claim that boundaries are slippery is not to claim that we are, therefore, united. Care, in Mol's redescription of the term, entails the work of coordinating across differences rather than the work of dissolving boundaries entailed in the push toward holism. Differences in expertise matter. The future of nutrient supplementation and taller bodies that Paisanos works toward is not one I would advocate. My own desire to slow research down, to ask questions and listen to responses “again and again” (Holtrop 2017), is an approach I share with Center C's scientists yet is one global health's policymaking structures only minimally value. The “good nutrition” sought by health workers who fight stunting, by regional medical workers who treat acute illness, or by development workers who seek greater crop yields is often divergent from that of the women they claim to serve—who spend their days farming and cooking but rarely speak of nutrition at all. Still, these actors are in conversation, frequently sharing multiple subject positions and wanting different things at once. Controlled equivocation helps to make evident that we are not just coordinating different forms of expertise toward a shared common goal; we are after different ends. Careful equivocation helps make evident how these ends will change as we set out to achieve them (see also Nelson 2009). To care for equivocations is to acknowledge that we do not have to know “the Other” to be involved in spaces of learning and exchange.

CONCLUSION: UNCOMMON OPENINGS

Articles published in the journals of “the expert field now self‐identified as global public health” (Pigg 2013, 127) tend to conclude with repetition. Findings are restated so as to solidify the knowledge that has been presented. I could do this here as well, but to emphasize the divergence in our methods as well as in our aims of research, I conclude not with closure but another story. For rather than move forward through problem closure, the version of anthropology I practice moves forward by opening up problems, turning them around, asking how they've come to be articulated as such—and how they could be articulated otherwise. The claim I make to doing global health is not made by evaluating who is healthy but by asking what happens when different articulations of evaluation come together and by tracing the effects these articulations have upon their worlds.

I noted this difference when talking with a scientist from Center C at an international meeting of nutrition experts on the Spanish island Gran Canaria. We were sitting on a pavilion overlooking the North Atlantic sea, not far from a spot memorialized as the last place Cristóbal Colón (Christopher Columbus) anchored before setting sail on his first voyage to the Americas—a voyage that would foment a future of violent, captivating stories about the truth of the world out there to be discovered. In the pale marble building behind us, hundreds of people had gathered for the World Congress on Public Health Nutrition, a meeting held directly beneath a banner with a quote from the French‐American scientist Jean Mayer: “Nutrition is not a discipline, it is an agenda.” Here, there was no need for anthropologists to mess with a modernist division between nature and culture; the scientists gathered were well aware their work had politics. My role was rather to draw attention to how many different forms of politics were at work—how many agendas were held within the sign “nutrition”—and how many agendas had not found a place there at all.

My friend responded that he found this approach interesting but worried that we risked talking past one another: “If health is not health, is not health, is not health, then on what grounds might we have a conversation, let alone develop the much‐needed interventions to improve the current, often‐tragic, state of affairs?” he asked. His is an important question, but it seems to presume that conversation—and the interventions they enable—necessitate common ground. Meanwhile, the search for common ground too often privileges, from the outset, methods skilled at making differences disappear. It is a search that too easily prioritizes dominant languages, and a search where claims to a universal truth (seem to) come out ahead.¹⁰

I responded to him with questions meant to shift the foundation of his question: What happens if we start our much‐needed conversation without insisting on the importance of commonality? What might be the outcome of engaging in collaboration from places of difference, learning to recognize—and also to care for—equivocations?

I end by drawing attention to an absence in this paper. I have written about my interactions with scientists, public health workers, and a translator—perhaps better called an equivocator—whose work brings these different disciplines together. I have not told you what is desired by women in the communities—women whose homes and bodies have been marked by global health as impoverished, who spend many hours of their days waiting in clinic lines holding their babies with that mixture of tender anticipation and boredom felt by mothers almost everywhere, and who must select from three common brands of nutrient supplements when feeding their families because three different projects running nutrition interventions in their communities do not, it turns out, coordinate very well. I have not told you what they think of these projects. I have not translated their understandings for you.

I hope you find this to be a loud silence, a present absence (M'Charek 2005). Too often, global maternal health projects carry on without soliciting input from the people whose lives they are meant to impact. Yet, in this article, I have not written their knowledge as a way of caring for the equivocations that would run through any attempt on my part to tell you what “they” want. I want to make evident that their knowledge is not knowledge I can make easily palatable, like a global health nutrient supplement, served up in a package of roughly ten thousand English words. In spending time—across and through material‐semiotic divides—with hundreds of San Juan women, it has become obvious to me that there exists no unified answer to what they think about the maternal health projects in their communities. Some wanted to be consulted; they wanted their techniques of expertise—from midwifery to childrearing, to a mode of quantitative assessment that we might term “Indigenous statistics” (Walter and Andersen 2013)—to be incorporated into the apparatus of global health. Many wanted their kids, so many of whom had become separated from them while traveling north, to be able to come home at night, to gather their families around them to chat and banter as they ate food pulled from their soils. But as for a single, steady outcome from the projects? The answer would depend on what was being offered, who was doing the offering, and what the offerings would displace. To think I could ever stabilize their understanding would ask for people to not be social, taking me along a path of inquiry that is as harmful in the social sciences as it is well trodden.

I have also not written their “knowledge” because I want to care for the fact that even as the divisions within and between the actors involved in Guatemalan global health shift in form as we learn from one another, there are some that hold strong. It is not possible to capture San Juan women's beliefs in my languages, but even if I could, it is not my place to share with my audiences stories that are theirs to tell.

It matters that the scientists I write about have read drafts of this article (or in María's case, have heard me describe it) and have offered their opinions and critical feedback. I have heard them say, for example, that they do not understand parts of it—a critique I expected because it was written for an audience of anthropologists and to shape anthropological expertise. Still, I have drawn them into the work of this awkward collaboration, and I have responded to some of their comments.

I have not made similar requests of women in San Juan—who are also experts in feeding and trauma, though they do not mark their expertise with the language of “nutrition” or “stress.” While I have learned considerably from them, the act of naming them as collaborators—even awkward collaborators—risks understating the economic and military borders separating their knowledges and lives from those of Euro‐American anthropology.

Of course, it can be argued that translating Mam worlds into English academic texts facilitates an opening, a space for inclusion. I leave open the possibility that this could be carefully done. But because I have seen such minimal structural change in the decade I have worked in these Mam communities—despite the many scientists and policymakers carrying out projects that they say will improve women's lives—I fear this act of ostensible inclusion too easily facilitates closure and erasure, much in the same way that the feminists cited by Strathern (1987) cautioned it would decades ago. Before bringing Mam to English readers, I would also want to see English translated into Mam. Before calling Mam women my collaborators, I would want to see them economically compensated for their work in ways that most Western institutional funding structures that champion collaboration do not presently accommodate. It is my hope that I can do more to change these uneven spaces between us by acknowledging them, not writing over them as if they were not there. Here, I am in agreement with Viveiros de Castro that learning to attend to equivocations instead of translating knowledges might make for a better anthropology.

The normativity implied by a “better anthropology,” however, returns us to the care work that is asked of careful equivocation. For rather than seek out stable or prescriptive answers to the question of what makes anthropology “better,” careful equivocation asks for responses that are situated empirically and that reckon with positionality: Better when, how, for whom? In moving from the aspiration to discover or uncover knowledge (once and for all, knowledge that will be true for all of us), it also moves toward a commitment to follow along as things come together and fall apart. While it follows transformation, it also follows the effects of talking about transformation to make evident that too often the very language of transformation can be used to uphold long‐standing nontransformative techniques that continue to pin certain kinds of people down (Ahmed 2006). It asks, in other words, for social theory to develop the capacities to attend to the different and shifting kinds of differences that are often erased by the universals of some powerful kinds of science, but that nonetheless remain a powerful part of daily life.

It would surely be possible to enrich the terms of “investigation” or “discovery” when writing about relations between anthropologists and the various groups of people with whom we interact, to make evident that our knowledges are more process than product. But as I noted to María at the start of our time together, it might also be time to let these terms go, shifting what we are after. The hope is that careful equivocation can help to articulate uncommon futures—futures where the awkwardness of our collaborations is more explicit, futures at once more imaginative and more real than a future prioritizing unity.