Incarcerated Pregnancies

“Dr. Sufrin, I gotta ask you somethin’.” Kima fidgeted with her jail ID wristband. As she continued, she looked down at her feet; each standard issue flip flop had the name of one of her children etched in blue ballpoint ink: on the right, Dante, and on the left, Koia, the baby I had delivered 4 months ago. “I was just wondering, did you take someone’s baby into your custody? Someone from jail? Because I heard that you did, and you know, you could have taken my baby when they put the CPS hold on her. But you said you couldn’t.”

Once I got over my shock, I responded, “No Kima, I didn’t. I can’t do that and I didn’t do it for anyone.” I could easily surmise the chain of people and events which led Kima to her misplaced jealousy that I had taken another enjailed woman’s baby. Here is what happened.

For the five and a half years before Kima’s question, I had been working as an Ob/Gyn at the local jail and the nearby county hospital where pregnant women in jail go to deliver their babies. I had also been simultaneously conducting ethnographic fieldwork for my dissertation in medical anthropology, officially for the prior year. Kima was in jail during her pregnancy. I had been her doctor and made a special trip into the hospital to deliver her baby on a sunny summer afternoon. Kima knew that I was also doing research on the experiences of pregnant, incarcerated women. After she had given birth, our doctor–patient relationship faded into the background, and she consented to be a subject in my study. I spent time with her in the hospital in her post-partum days of recovery—more time than I would have otherwise, since I happened to be post-call and was in need of a long nap. Outside her hospital room a uniformed guard sat, to ensure that Kima would not escape. He did not raise an eyebrow as I entered and exited the room of a prisoner, because I was wearing scrubs and had a hospital badge clipped to my shirt. Inside the hospital room, Kima wore a blue and white patient gown, ate three tasty meals a day, had unlimited phone access, and sang to her baby as she nursed this new life. In a corner of the room, her standard issue orange jail uniform was balled up on the floor, a barely perceptible reminder that she is still incarcerated.

She would soon be returning to jail, without her baby. Kima had arranged for her sister to take care of baby Koia until she got out of jail in 2 weeks. But, as it turned out, her sister had an open case with Child Protective Services (CPS) and could not take the baby. A CPS worker then put a police hold on baby Koia—this meant she could not stay in the hospital room with Kima; they could only spend time together in the nursery. Kima wailed. Her sobs pierced through the many cries of newborns and women in labor which are background noise on a labor and delivery unit. She was going back to jail in 24 h and wanted to spend every minute with her baby until the carcerally enforced separation. After their hospital farewell, the baby was to be placed into foster care.

One month later, Evelyn delivered a baby girl. Evelyn had been in and out of jail most of her pregnancy, and I was the only obstetrician she saw for prenatal care. She was released from jail to a residential drug treatment program 5 days before she went into labor. Although she was in a program designed for mothers and children, although she had not used drugs in 5 weeks, although she was talking the ‘change talk,’ these were not enough for CPS; the CPS worker decided the baby would go into foster care until Evelyn could prove herself. Since she was no longer officially my patient, she signed a consent form and became a ‘research subject.’ I had become close to Evelyn and felt optimistic about her dreams to recover from addiction and be a mother to her baby, the third she had given birth to and the first with any hope of retaining custody. So I was baffled and angry with CPS’s decision to intervene when Evelyn was doing everything she was supposed to be doing.

Evelyn also felt we had become close: she named her baby Carolyn, partly after her mother who had died when she was 5 years old and, so she told me, partly after me. I spent a lot of time with Evelyn: at her drug treatment program, accompanying her to supervised visits with her baby, advocating for her at CPS meetings, driving her to court appearances. Every week seemed to bring with it a new stipulation from the courts or CPS as to why Evelyn could not have baby Carolyn just yet. And then Evelyn relapsed. The stress of CPS’s continued deferral of reunification with her baby, she later admitted, pushed her over the edge. She called me one night from her familiar drug corner; with a tremulous voice, she declared “Dr. Sufrin, I just wanted you to hear it from me. I left the program.” I was in shock; I felt betrayed; I did not know what to do—until I decided I had to find her and try to help. Find her I did. I also found several other women I knew as my patients in the jail. One of them, Quianna, initially assumed I was in cahoots with the police. I explained I was there to help Evelyn, and eventually Quianna came around. “Wow Evelyn. You’ve got your doctor here as your guardian angel. You better go back to that program and raise that baby of yours.” She did neither.

And now, with 4- and 3-month-old babies in the world, Kima and Evelyn were both back in jail. Both of their babies had been placed with family members. Both of them were in a familiar struggle of navigating the legal terrain and society’s moral judgments surrounding motherhood and drug addiction. Kima was currently cellmates with Quianna, who had seen me on the street that night trying to rescue Evelyn. When Kima asked me if I had adopted someone’s baby, I quickly realized that Quianna either assumed or deliberately fabricated that my helping Evelyn on the streets meant I had taken baby Carolyn into my care. Kima was still scarred by the police hold that CPS placed on her newborn in the hospital. So when Quianna told her about me and Evelyn’s baby, Kima had to ask me why I had not spared her the pain.

Revisiting the Margin

These parallel stories of incarceration, addiction, birth, and child guardianship are fertile grounds for many concepts in social theory—which I explored in my doctoral dissertation. What I want to focus on here is a critical engagement with my strategic but limited role in Kima’s and Evelyn’s stories, and their role in mine: or more generally, the convenient and inconvenient positionality of being simultaneously both a physician and an anthropologist. How does such a dual identity frame our informants’ perceptions and expectations of us, and our expectations of them? What are the demands on ethical practice created by the intersection of the already fraught relationships of doctor–patient and researcher–subject? These questions are not new, and there is a rich legacy of past and current people who have practiced both medicine and anthropology. But just how physician–anthropologists practically and ethically have negotiated their dual roles and commitments has not yet been the subject of study. Moreover, scrutinizing the duality and the questions precipitating from it crystalizes fieldwork issues relevant for all anthropologists, for as Michael Walzer has noted, all selves are divided among a multiplicity of roles, identities, and ideals. The self, he writes, “speaks with more than one moral voice—and that is why it is capable of self-criticism and prone to doubt, anguish and uncertainty” (Walzer 1994, p. 85).

The moral entanglements of the physician–anthropologist duality are a case study in such divided selves, bound by professional conventions; the duality, as it plays out, highlights how the conventional ethics of anthropology and medicine challenge and enhance each other. Arthur Kleinman (1981, 1995) spatialized the tensions between medicine and anthropology into a margin in between them, a critical and productive space for understanding human experience. A margin, more dynamic than an intersection, is “bound to both, and at the point of their engagement, in order to discover that they are (or are not) and I am or am not” (Kleinman 1995, p. 3). It is a good place to begin when thinking about Kima’s and Evelyn’s childbirths, about the ethical reflections onto each field, and about how being an anthropologist led me to act in ways I might not have as a doctor, and vice versa.

In this paper, I delve into the margin to grapple with the methodological tensions and transformative potential of concurrent clinical practice and anthropological inquiry. Throughout, I draw on data from my fieldwork and clinical practice at an urban, county jail for women and its community environs. Examining the constantly shifting roles and obligations of practicing medicine and anthropology makes visible quandaries that both disciplines struggle to confront. In particular, reflecting anthropological and clinical ethics on each other reveals how each field addresses or avoids four overlapping challenges. These include how to manage public disclosure of information gleaned through intimate interactions, and the implications for privacy and trust. Second, it also includes how our divided selves influence our expectations of our patients and informants, and theirs of us; this can, thirdly, lead to an unsettling recognition that there may be limits to our compassion which we hesitate to admit.

Finally, what these all converge on is that, because both disciplines rely on human relationships, they both entail an ethic of care with certain practical, affective—perhaps more obvious in medicine—and political dimensions. The margin between anthropological and clinical ethics calls on the other to recognize care not only as taking care of another person, but as “taking care of relations” (Geissler 2013)—relations which emerge through fieldwork and healing endeavors. Here, I take the many registers of care fundamentally to encompass “the way someone comes to matter and as the corresponding ethics of attending to that other who matters” (Stevenson 2012, p. 595).

The Generative Clinic: Framing Anthropology and Data Dilemmas

In 2007, I began volunteering as an Ob/Gyn at the local jail in my urban community, motivated by a commitment to serving underserved populations and by a curiosity about the intersection between carceral and biomedical logics. Were the regulatory elements of medical practice integrated seamlessly into the classic disciplinary aspects of the jail’s everyday routines? How could the intimacy of caregiving occur in a punitive space presumed to be devoid of humanity? In the jail’s clinic, I cared for women with irregular periods, chronic pelvic pain, sexually transmitted infections, drug addiction, histories of sexual abuse, women desiring contraception, pregnant women seeking abortions, and pregnant women who needed prenatal care. I met women like Evelyn and Kima for whom involvement in the criminal justice system was a normalized part of their lives, not an exceptional moment of arrest into a punitive institution. After only a few months of doctoring in the jail, I was overwhelmed with the complexities of patients’ lives, and with the equally complex interactions among guards, medical professionals (including myself), and patients who were incarcerated. These clinical experiences compelled me to turn to formal anthropological training and informed the foundation for my dissertation research.

The writings of other physician–anthropologists have produced similar evidence that being a clinician-anthropologist provides a special framework for understanding phenomena they analyze anthropologically. Kleinman’s influential Patients and Healers in the Context of Culture (1980) used his patients’ stories to develop models, now critiqued, of how cultural and social context influence the experience of illness. Obstetrician–anthropologist Clare Wendland (2006) deftly weaves her obstetrical knowledge into an analysis of how evidence-based medicine has displaced the mother from the birth experience in the U.S.. This analysis is informed not only by her rigorous medical training in the subject, but also by her inculcation into clinical decision making processes and, undoubtedly, intimate knowledge of her individual patients.

Using clinical insights to frame anthropological inquiry is certainly useful, but bypasses the moral entanglements of duality, and can encourage a partial compartmentalization of roles. I attempted, once I officially began fieldwork for my dissertation, to compartmentalize my time at the jail: Mondays I would remain a doctor in the clinic, and Tuesdays through Sundays I would observe what transpired in the jail and its clinic outside of the exam room. This seemed easier, and less messy—but ultimately impossible. On my doctor days, I was listening to women’s heart-wrenching stories of violence, manipulation by their boyfriends, addiction, and dreams to transform with intimate detail that was enabled by privacy of the doctor–patient relationship. I was interacting with jail staff who controlled my access to patients, experiencing the carceral constraints on and possibilities for caregiving. And on non-doctor days, I also fielded clinical questions from nurses, inmates, and even deputies. One Sunday evening at the jail, for instance, Evelyn was having contractions; I examined her cervix and determined that it was false labor, saving her and the staff an unnecessary trip to the hospital. Even as they knew I was conducting research, people at the jail always viewed me as a doctor. Compartmentalization was not an option.

As another example of the clinic being generative for anthropology, to demonstrate structural violence, Paul Farmer (1996, 1999) narrates stories of people by providing clinical details that suggest his key informants must also be his patients—those suffering from HIV, or from a fungating breast mass. In Infections and Inequalities (1999), Farmer labels some of his stories as “ethnographic interludes,” yet he is written into these interludes as an anthropologist who is also a healer; his subjects are people who have come to him seeking medical care. The representational politics of categorizing ethnography as an “interlude” aside, that some of his rich stories of people who may have started as patients are labeled “ethnographic” belies the complexities of using our own doctor–patient encounters as analytic material.

This is the gray zone where the privileging of health information over other personal information, Institutional Review Board (IRB) interdictions, and the inherent power imbalance of the doctor–patient relationship come into play. In western societies, medical information about a person is enshrouded in a culturally specific value for the sanctity of the individual (see Cockroft et al. 2009), whose intimate bodily processes are revered as private knowledge. That value of exceptional privacy of health related information is operationalized in the confidentiality of the doctor–patient relationship: abiding by ethical conventions, doctors cannot disclose information about their patients publically, with very few exceptions (child abuse, for instance). Such agreements are understood as fundamental to the trust a patient has in her doctor—that she can be forthcoming in the exam room without others’ learning things she does not wish to share.

Codifying this secretness of medical information into regulation is the federal Health Information Portability and Accountability Act (HIPAA); the HIPAA Privacy Rule grants patients medical privacy as a right and penalizes institutions which disclose health information without a patient’s permission (Department of Health and Human Services 2003). HIPAA makes privacy protection part of the caregiving endeavor, a part of the conventional ethics of practicing medicine. Such cultural, professional, and legal veils all signal that the privacy of medical information is highly respected and thus potentially problematic when made public in research.

Although ethnographic research gathers deeply personal information about people’s experiences, it carries a less sacred—though still regulated—valence in institutional bureaucracies than does information about a person’s medical diagnostic tests and doctors’ visits. Numerous anthropologists have theoretically mused on their frustrations with obtaining “ethical approval” from IRBs that have limited abilities to understand the relational, open-ended, and intimate nature of ethnographic fieldwork (Strathern 2006).Footnote 1 While the American Anthropological Association (2009) has published core guidelines for the ethical practice of research, deciding what ethnographic information to include and to exclude is left to trust in the researcher’s best intentions, to a sense of responsibility to those we study (Battaglia 1999; Slattery and Rapp 2003). The process is always partial, representation is always incomplete, but we do our best (Lederman 2006).

This conventional ethic in anthropology is challenged when a physician–researcher is privy to “protected health information” gleaned from a clinical encounter or a medical chart. IRB rules prohibit the use of HIPAA-protected information, unless the subject–patient has given consent and signed a separate HIPAA waiver. Without such a waiver, medical details must be omitted, unless the informant volunteers such information in a non-clinical, research setting. But some of those details might thicken the description or illuminate key analytic points; excluding them might be more problematic for developing an accurate assessment. In a prenatal visit in the jail, for instance, Evelyn divulged to me deeply personal information about her childhood sexual abuse, the reason she started using drugs, and an important detail in piecing together her current experiences of reproductive abandonment. IRBs are designed for data about individuals or groups, but not about relationships (Strathern 2006); their rules are limited in their abilities to understand the centrality of intimate relationships to ethnography—and thus how, other than by prohibition, the intimacies of caregiving can also be understood as data.

In instances like the prenatal visit with Evelyn where clinically ascertained information is anthropologically illuminating, the conventions of anthropological ethics might compel us to include medical information using the usual tricks of concealment: de-identifying patient information, changing details, making composite patients—these are techniques which ostensibly eschew the privacy issues while still valuing the depth of insight gained from the interstices of clinical medicine and social analysis.

Whether guided by HIPAA or by good faith ethnography, the decision of what to do with medical information obtained in a clinical encounter reinforces the notion that some forms of data are regarded as more sacrosanct than others. When non-physician anthropologists make ethical decisions about what information to divulge in writing, part of that too involves implicitly asking what is too private to share. The regulatory and ethical tensions over what to do with “protected health” information for physicians wanting to use such data in research makes visible an ethic of care which is central to anthropology. In anthropology, the boundaries of data are less distinct than the door of a clinic exam room, but we learn to do our best with the informality of data gleaned from fieldwork relationships (Lederman 2006).

“Taking Care of Relations”

The intersubjective connections of care are not, of course, limited to medicine’s therapeutic attempts, or merely an anthropological object of study. Care, in its affective and practical dimensions, is also central to the relationships formed in anthropological fieldwork. We seek to know people’s lifeworlds by building relationships with them. We insert ourselves into aspects—often intimate ones—of their everyday lives in ways that cultivate concern for their well-being. Accounting for such ties, anthropological ethics are not concerned with “regulatory ‘protection’ of subjects but with ‘taking care’ of and with relations with multiple others” (Geissler 2013, p. 30). Practically, taking care includes being attentive to decisions about the data constituted by those relations. Although the privacy mandate of HIPAA, seemingly ill-suited to the relations of fieldwork, has mutated expressions in the decisions anthropologists make about what information to exclude from our writings. In war zones, for instance, anthropologists must take care not to reveal information that would compromise someone’s safety (see Nordstrom 1997).

The ethics of taking care of relations with others may at times compel the disclosure of information which could be damaging to certain people we have come to know in the field—for we come to know multiple others. Nancy Scheper-Hughes’ extended fieldwork in the global trade in organs, for instance, has involved her forming relationships with organ brokers, donors, and recipients (see Scheper-Hughes 2004). In taking care of those various relations, she has publically revealed the identities and activities of some of her informants who control the illegal organ trade, compelled by ethical commitments to expose the deeply exploitative nature of this phenomenon and its ties to poverty.

If the privacy and trust elements of the doctor–patient relationship have traces in fieldwork relationships, then we must also consider how anthropologists’ careful approach to relationships and disclosure reflects back on the protectionist dilemma of deploying medical information in research. Taking care as a physician–anthropologist outside the clinic means a commitment to understanding the importance of a clinically gleaned detail (including probing whether it is even important) to someone’s everyday reality. It entails a relational ethics which compels researchers to initiate conversations (Ellis 2007, p. 4). This is more than simply asking permission, seeking “informed consent.” It means asking more; bringing the clinical moment outside the clinic and intentionally probing it with the person, as I later did with Evelyn; and overcoming the protectionist anxieties by engaging the patient–informant in the process—in other words, being an anthropologist.

The Methodological Grammar of Observation and Participation

The cornerstone of ethnographic fieldwork classically involves the researcher engaging in a combination of observing and participating in the everyday realities of the people and phenomena we are studying. We call this “participant observation.” Its characteristics easily roll of the tongue of any anthropology graduate student: immerse yourself in a place and among people; observe; intimately participate in everyday life; take part in community rituals; attend important events; maybe even do some work alongside your informants. It is a dynamic between involvement and detachment (Powdermaker 1966, p. 9), a tension which Benjamin Paul (1953) deemed an oxymoron. The grammar of the term “participant observation” combines two nouns, the first identifying the ethnographer in a state, the second explaining what she does, though the sequence posits “participant” as a descriptor of “observation,” giving primacy to the latter. The ethnographer is in it, but not fully of it.

Doctoring in the jail while I was doing fieldwork there was grounded first in my participation at the site, and being observant of my own and others’ actions, interactions, and reactions—perhaps more aptly described as “observant participation.” I was a doctor at the jail before I was an ethnographer there. In fact, it was participating in the overwhelming complexity of administering care in a space of discipline which then led me to pursue rigorous social analysis of it. With observant participation, one is already an integral actor in the phenomenon she is studying, making direct contributions in real time. One is in it and of it.

There are numerous examples of non-physician anthropologists similarly using their established group membership opportunistically to do ethnographic research. Michael Oldani (2004) was a Big Pharma drug representative who then analyzed his experiences to theorize the gift economy of the pharmaceutical industry (although he left before writing about it). Brian Moeran (2007) became a participating advertising executive during the course of his fieldwork in a Japanese company when his subjects loved the tagline he created for one of his projects. Others deliberately become an integral part of the community precisely to better understand it. Loic Wacquant (2011), for instance, trained to be a boxer in order to acquire and thus better understand the habitus of those he was studying; he identifies this as observant participation, claiming that intimate understanding of what our informants experience can only be obtained through rigorous immersive practices. These examples of observant participation remind us that the grammatical sequence of participation and observation—and the attendant ethical concerns—holds a dynamic potential for all anthropologists.

The writing analog to the methodological dynamic of observant participation is a type of autoethnography. This genre of research and writing “seeks to describe and systematically analyze (graphy) personal experience (auto) in order to understand cultural experience (ethno) (Ellis et al. 2011, p. 1)”. There are times when one’s own experience—not merely the experience of being part of someone’s lifeworld—is a critical source of data (see among others Ellis et al. 2011; Chang 2008; Barton 2011). People who identify as writing autoethnography emphasize that every element of one’s life becomes data (Barton 2011, p. 243), with theoretical insights informed by interactions with everyone they encounter. Data from informants and the researcher’s own lived experience become ‘co-mingled’ such that demarcating the field becomes an impossible task (ibid). The questions raised by autoethnography encourage a general questioning of what counts as data.

While many authors have vexed over the dilemmas and taxonomies of autoethnography (Ellis et al. 2011), these tensions between the nature and source of data arise with any immersive fieldwork and with all ethnographic writing. Fieldwork is inherently informal, consisting of “undemarcated moments of ethnographic practice when ‘research’ and ‘daily life’ are inextricable” (Lederman 2006, p. 477). As the Writing Culture project taught us, we are always part of the story being told, and the stories necessarily incorporate our own experiences into what we are analyzing (cf Clifford and Marcus 1986). The dilemma of how our selves fit into experiences we analyze is the Heisenberg uncertainty principle of anthropology. In physics, the Heisenberg uncertainty principle affirms that because of the presence of the observer, one can never know both the position and velocity of a particle at a given moment (Wheeler and Zurek 1985). The quantum physicist is, in effect, a part of the experimental phenomenon he is measuring. Similarly, the anthropologist can only know about a phenomenon through her own participation in it. I cannot understand how Evelyn’s baby would have otherwise been processed through state systems had I not been speaking up as “her doctor” at meetings with Child Protective Services or at closely scrutinized 2 hour, biweekly visits she had with baby Carolyn.

To write oneself out of ethnographic narratives in which one is dispensing anti-retroviral medications to HIV patients who have climbed mountains to obtain them (Farmer 1999), in which I am delivering a baby from an incarcerated womb into state sponsored foster care is as much a political elision as the elite outside observer of a Balinese cockfight overlooking not only his colonially enabled positioning but also the surrounding eruption of genocidal civil war (Geertz 1973). Not bracketing out the active role a practicing physician might be playing at a site similarly acknowledges the politics and hierarchies which predetermine the research situation in the first place. After all, we all have multiple selves in a given situation (Walzer 1994).

What is different about being a doctor in one’s fieldsite from the usual undemarcated moments of fieldwork is the private status our western culture has ascribed to bodily, medicalized experiences; that is, we view the body and its medical encounters as the purview of individuals,Footnote 2 not of a community, consistent with western values of the sanctity of the individual. This value is the basis of HIPAA regulations, after all. Yet is not the birth of Kima’s baby, whom I caught with my own gloved hands, also my experience, and within my purview to analyze? After all, this embodied, tactile moment informs my understanding of how her incarceration and its social production made the birth both different and similar to non-carceral reproduction. As a doctor and an anthropologist, my clinical interactions with Kima, her baby, and even with the jail guard outside her hospital room were data. Yet using experiences from concrete clinical encounters as ethnographic data further challenges the conventional ethics of anthropology and of medicine—the ineffability of relationships with others and human insights that provides, versus the sacredness of IRB rules.

Observant participation is an apt description for the concurrent work of physician anthropologists at one clinical and field site. We are deeply engaged in relationships with other professionals and with patients, relationships that are rife with opportunities for anthropological inquiry. We are entrenched in the institutions we often critique, yet we also must maintain credibility and excellence in both fields. Every day in the clinic presents an opportunity for ethnography. As embedded insiders, physician–anthropologists are challenged to constantly balance our inculcation into two disciplines with different epistemologies. We are practitioners within the culture of biomedicine and have anthropological frameworks to analyze what we practice. Sometimes this produces an uncomfortable tension, and other times we might resolve the conflict by temporarily abandoning one framing for another.

In medical school, we learn to compartmentalize the body and reduce its reality to functioning molecules. We learn to differentiate between the patient’s “subjective complaints” and “objective findings” as markers of disease (see Holmes and Ponte 2011). In contrast, our anthropological training teaches us to dissect this biomedical model into its cultural and political economic specificities. Because we are anthropologists, we get frustrated when other doctors speak of “medication non-compliance” without questioning the political assumptions of a model of care that values compliance. We might even get more frustrated with doctors’ misguided attempts to attribute undesirable patient behaviors to “cultural beliefs,” overlooking structural forces (Benson and Kleinman 2006).

As integrated insiders, we are also privy to moments when colleagues work hard to secure housing for patients, recognizing the structural inequalities which shape their lives; we see fellow health care workers struggle with the limitations and benefits of medicalizing conditions like addiction and mental illness. On the wards, we rattle off the results of the latest randomized clinical trial (RCT) to help our team make evidence-based clinical decisions, citing p values and odds ratios as justifications for a particular therapeutic intervention. We deploy such statistics knowing fully well that all knowledge is produced, that RCTs are themselves discursive practices (Adams 2002). And despite what we know from Foucault about the cultural specificity of the clinical gaze, when we incise into someone’s peritoneal cavity and see organs, nerves, and blood vessels, we believe, “this is how it really works.” (or perhaps I am merely projecting.) This version of observant participation is not simply about privileged access, or insider knowledge, as it appears for Oldani and Wacquant. For physician anthropologists, the dual embededdness constantly brings epistemological tensions to the fore. We must contend with our complicity in the systems we may critique.

In contrast to Wacquant’s version of observant participation, I am not claiming that a physician–anthropologist experiences what a patient–informant experiences. Rather, this observant participation references the relational aspects of both fieldwork and medicine; the dual positionality comes with a built-in relational orientation as a health professional that may preclude certain types of ethnographic knowledge and facilitate others.Footnote 3

Moreover, Wacquant’s participation as a boxing apprentice was for the express purpose of ethnography. His identity as a boxer was not sustained, nor was it his profession with accompanying fiduciary obligations. Physician anthropologists, in contrast, do have sustained participation in the field, and that practice constitutes a professional identity as well as a set of relationships and hierarchies with colleagues and patients which must be tended to.

This is where relational ethics of anthropology and of clinical care converge. Once Kima and Evelyn gave birth to their babies and were no longer pregnant, I reconfigured my identity as their obstetrician, their doctor, to the past—the condition for my knowing them, but a relationship with professional caregiving obligations which had ended. Relational ethics of ethnography requires confronting the ethical responsibilities we have to people as the nature of our relationships with them change over time (Ellis 2007, pp. 4–5). For Kima, although childbirth may have ended my physical care of her pregnant body, my role as a caregiver was still salient: it was entirely in the realm of possibility for Kima that I could have adopted baby Koia, a gesture of care for Koia’s mother as well. But anthropology too, as I have already discussed, requires an ethic of care, of taking care of relations. Continuing to invest in relationships and not abandoning the prior terms of those relationships—recognizing that our informants may hold onto them—even as circumstances evolve, is a necessary part of that ethic of care.

Privileged Access, Trust, and Taking Care

In addition to this deep insider knowledge, being positioned as a medically authorized observant participant also allows us tremendous access which enhances our anthropological endeavors: established relationships within and knowledge of medical systems, and the cultural capital of being a doctor. My work at the jail would have been extremely difficult were I not already working there. Before I started my research, I had official Sheriff Department clearance which got me into any part of the jail at any time I wished to be there. I had earned the trust of the people working and incarcerated in the jail over 4 years as a doctor before I officially began my fieldwork. For an incarcerated person, a doctor in a jail controls access to important resources which become medicalized in the deprived environment of a jail—a bottom bunk, a special diet, extra ice and water, for instance. I had a known role and a purpose in the jail, one that was highly valued. So when I would ask deputies to allow me to speak privately with an incarcerated woman in the housing units, although they knew about my research, they assumed it was to discuss medical issues.

In this configuration of privileged access, the challenge is less about gaining trust than it is about not abusing the trust (Barton 2011, p. 11). For all anthropologists, not abusing the trust of any informant usually involves some calculus of consent and judgment about inclusions and exclusions of information. We ask, what are the consequences of disclosure? Privileged access may be the basis of intimacy, as perhaps it was through my involvement in Kima’s and Evelyn’s pregnancies. But trust and decisions about privacy are embedded in many intimacies which emerge in fieldwork relationships. This trust often becomes apparent when informants perceive it to have been broken, for instance, when they discover that something they presumed would not be shared was in fact circulated.

The choices of what information to include are ethical ones, and part of the relational ethics of anthropology beyond the protected privacy of health information. What the duality of the physician anthropologist introduces into these vexations of trust is the culturally authorized, fiduciary trust of the doctor–patient relationship. Trust still must be earned in the clinical encounter, but there is a baseline of culturally and professionally scripted trust that the patient has in the physician.Footnote 4

With medically contextualized observant participation, then, the calculus of the intimate trust and what counts as data also includes the consequences of revealing medical details and questioning their analytic utility, for anthropological practice may then shape clinical relations. It is managing what to do not with undemarcated relationships (Lederman 2006), but with demarcated clinical relations. For example, among my patients was a pregnant patient, Deshawna, who was released from jail. She struggled with drug addiction and had no place to call home, like so many other women cycling through our criminal justice system; she had slipped through a porous safety net and had nowhere to go when she was released from jail to the streets, pregnant. Six months later, while high on crack, she delivered a premature infant on the sidewalk and handed it, cold and placenta still attached, to a stranger on the street. Deshawna was arrested that night for child endangerment and her case—which I report here because it has been published in newspapers—became high profile.

I knew a twist to the story based on clinical details from earlier in her pregnancy when she was in jail. While this would have added to my dissertation analysis of reproduction and incarceration, I felt compelled to omit these details—in part because I feared they would further vilify her into the stereotype of a “bad mother” (see Tsing 1992). I did not want to risk abusing Deshawna’s trust in me as a doctor by telling her story—even if she had given me permission—out of concern for the relational ethics of the situation; there were very real legal consequences to her reproduction, for it was the basis of her current incarceration. Deshawna might have disclosed information to me which, were it discovered, could have influenced her court proceedings and the conditions of her criminal case. I did not enroll her as a subject in my study, nor do I reveal now anything more than what is in the newspapers.

In contrast, Kima’s and Evelyn’s births were less legally exceptional and not the subject of criminal investigation. I felt comfortable, and that it was not an abuse of their trust in me, once I was no longer officially rendering medical services to them, asking them to participate in my research (and sign a consent form), and asking them if I could include information about and their narratives of their incarcerated pregnancies (I was relieved that they both said yes, because I had been keeping notes all along). But in retrospect, perhaps my protectionist ethical stance to information about Deshawna’s reproduction was too grounded in the individualistic, HIPAA model of medicine. Perhaps the anthropological ethic of taking care of relations would have challenged me to depict Deshawna’s reproduction—clinical twist and all—in a sympathetic light highlighting the structural forces which contributed to her birthing a child while high on a street corner. This is part of anthropology’s ethic of care.

The Heisenberg uncertainty principle of fieldwork requires us to recognize that we can never know how our presence is altering a situation. It also means that there may be an element of ethnographic performativity; that is, like the classic Far Side cartoon in which the grass-skirted natives are hiding their televisions for the approaching anthropologists (Larson 1984), our informants may at times be strategically showing or concealing what they think we want to see. In an observant participation mode where the ethnographer-physician has technical expertise to intervene in certain scenarios,Footnote 5 what are the stakes of patient–informants’ truth-telling? Just as I am carefully calibrating my actions toward them along my ethnographer–physician continuum, might not they be strategically calibrating as well?

As a speculative example, the night Evelyn left the drug treatment program, she had told me she left because the other residents at the program were annoying her; I had witnessed her short temper before, so this seemed like a plausible explanation. Surely she could make amends, so I urged her to return to the program, or at least to take the subway to her Aunt Vera’s for the night. She provided me with a litany of reasons and program rules why she could not do either right then, that she would have to wait until Monday to go back—but she would, she told me. A few days later, I learned from the program administrators that they easily welcome back absconders up to 3 days from their departure. And I also learned that Evelyn had received bad news from CPS the day she left, the more likely reason for her leaving, rather than her annoying co-residents.

Informants conceal parts of their lives for many strategic and incidental reasons. But with an ethnographer–physician as the observant participant, informant–patients may withhold or modify truths if it may strategically give them access to medical care or other services a professional can provide them with. Perhaps Evelyn feared that full disclosure of her desire to be back on the streets would jeopardize my willingness to continue to use my medical status to help her custody case. Perhaps not. There are, of course, many social relations and power dynamics which matter in Evelyn’s everyday reality. To overdetermine the influence of being a doctor–participant in her life would be as dangerous as overlooking it.

The Messiness of Roles

Observant participation is a shift from participant observation: not a quantitative shift in intimacy, but a qualitative shift in the kind of intimacy and perceptions that emerge from it. I do not contend that observant participation is better, richer, or superior. Rather, it comes with a different set of responsibilities and expectations. Thus far in this paper, I have been talking about how clinical experience enriches and challenges anthropological analysis, about how other physician–anthropologists have capitalized on the intimacy of doctor–patient relationships. What we must be more attentive to is how observant participation and an anthropological gaze play a role in clinical practice and in how people perceive us as doctors. Understanding this helps us then recognize that anthropology too involves an ethic of care. The margin between medicine and anthropology, after all, does not have a vector. It is an interstitial space.

When Evelyn left the drug treatment program and relapsed, I continued to search for her on the streets. I did this both to persuade her to return to the program and to thicken my ethnographic research. In my wanderings, when asking others involved in the drug economy where Evelyn was, I let them know “I’m her doctor. I’m trying to help,” so they would not be suspicious that I was an undercover cop. I deployed my medical identity (and its inevitable authority) in hopes that this would emphasize my status as a person who cared, who was not out to punish Evelyn. I did not say “I’m her anthropologist, I’m trying to learn about her experiences.” Though of course I was also doing that. In fact, were I not an anthropologist, would I, as a doctor, be wandering drug corners by myself trying to rescue a patient? Most likely not. It was anthropology’s open inquiry into everyday reality which drove my wanderings, which led me to caregiving gestures, like trying to help reunite Evelyn and her baby, beyond the clinical encounter. Being an anthropologist—spending time in my patients’ worlds, getting to know them and their families, listening—led them to perceive me as an especially attentive and caring doctor. Following the fieldwork trail led me to expanded relationships of care, where my former patients and current informants saw me as someone whose doctor status in society could help their recovery, or help their child custody battles.

When Quianna saw me that night on the streets and saw how involved I was in Evelyn’s life, I worried that the next time she was in jail she would try to get something out of me in exchange for her discretion. She did not make special requests, but she did tell Kima that I had adopted Evelyn’s baby. For Kima, both my keen interest in her life outside of the exam room, as an anthropologist, and my clinical expertise enabled her to then to insist that I should have taken her baby too. “After all, you’re a doctor. You would know how to take care of a baby. And they would trust you! Then I could have gotten her back when I got out of jail.” Kima’s flawed assumptions aside, she was able to create expectations of me as a doctor which she would not have made if I were not also extending my reach further into her life as an anthropologist. Anthropology compelled care.

Anthropology’s ethic of care is precisely about this tending to relationships. The conventional ethics of both medicine and anthropology are oriented around one’s responsibility to others, and they productively challenge each other. If medicine challenges anthropology to be selective not as a representational crisis but as a gesture of concern for repercussions of disclosure, then anthropology challenges medicine to think about the problems of care.

Although all of the pregnant women I followed inside and outside of jail had signed consent forms after my official doctor role in their lives had ended, they often had trouble calling me Carolyn. Dr. Sufrin I remained. There was always a doctor in the house. One night, when visiting Evelyn at her program, one of the other residents slipped and fell on her arm. Everyone turned to me, even though my clinical expertise as an Ob/Gyn is below the waist. I knew enough to know that her extreme pain meant that she needed to go to the emergency room for an evaluation and an X-ray—an assessment many non-medical people would also make.

Even as we gobbled down burritos at hole-in-the-wall tacquerias, talked about the men in our lives, or celebrated Obama’s re-election together, my informants did not compartmentalize my roles. They boasted to their friends on the street, to other women in drug treatment programs, to their CPS worker or attorney that their doctor was here to support them. Despite my initial, HIPAA-fearing intentions to compartmentalize them into former patients and current subjects, their comfort with my overlapping roles unsettled any real dichotomy to which I might have aspired with the consent form. And so they asked me to refill prescriptions, to schedule doctors’ appointments for them. Those requests were easy to handle on a case by case basis.Footnote 6

But finding Evelyn on the streets, 32-weeks pregnant and high on crack was more challenging. I did not lecture her, even gently, like a doctor, telling her to stop smoking crack, even though she was potentially compromising the health of her future child. I feared that explicit invocation of my medical authority would alienate her and significantly limit our interaction. Instead, I sat with her, observing her and the urine-stenched world around us, like a good anthropologist. This departure from what she would expect from a doctor, I believe, led her to trust me more: both as a doctor and a researcher. Yet my silence in not discussing on the streets the medical risks of crack use in pregnancy (which Evelyn had heard from me and others in the clinic) concealed the judgment I harbored amid my openness, that Evelyn should not be smoking crack. My silence also concealed the authority I had in the situation to act. For I, unwittingly, generated my hopes and expectations of her, like doctors sometimes do with patients—the patient who has had three heart attacks whom we hope and expect will quit smoking. So I was disappointed, even angry, when Evelyn absconded from the drug treatment program, relinquishing her chance to be a mother to baby Carolyn. The tensions between coexisting openness and judgment reveal the unsettling thought that there might be limits to our compassion as sympathetic anthropologists and caring physicians.

I saw Kima for a clinic visit a few days after explaining that I had not adopted anyone’s baby. We talked about contraception, a standard topic for a post-partum visit, to optimize birth spacing within a time frame defined as medically safe. “Oh no, I don’t need birth control. I want to get pregnant now.” Kima explained her reasoning, that if she could just stay off drugs for the pregnancy, then she would finally be able to keep a baby. Despite having a 4-month-old baby with whom she could still try to reunite, she thought she could be more successful if she started afresh with a new pregnancy.

The conventional ethics of both anthropology and medicine necessitated that I listen to Kima non-judgmentally. It is the relational practice of cultural relativism. But in medicine, this professional ethic often runs counter to paternalistic tendencies which still pervade practice, that sometimes we think we know what is best for a person. It is couched in the language of “professional expertise.” When patients do not “comply” with our advice, we are taught to non-judgmentally explore why, concealing our frustration with the patient. This learned approach is characteristic of what P.W. Geissler (2013) calls “unknowing.” Geissler describes how public health research collaborations between North American and African researchers involve practices which actively unknow the obvious inequalities between the two groups, and therefore make the research possible.

Similarly, physicians and anthropologists unknow the reality that there may be limits to our compassion. I wanted to shake Kima, to yell at her to focus on her newborn instead of “giving it another try.” But instead I kept my professionally engrained composure and politely probed. This outwardly non-judgmental response helps not to shut down the therapeutic relationship, and it promotes the ongoing ethnographic relationship. But it also can bury our own human responses to challenging situations. In medicine, the professionalized unknowing of our judgments may in the long run prevent optimal healing if it stops us with respecting patient autonomy, and no further pursuit. If we have an ethical responsibility in anthropology to the relationships in which we are engaged (Strathern 2006; Geissler 2013), then part of that care can include—in a manner less aggressive than shaking or yelling at someone—knowing that our “view from somewhere” (Haraway 1988) includes personal judgment of others; that our compassion is not limitless. This, too, is part of taking care of and with relations.

Embracing these entanglements and the possibility of limits to our compassion is uncomfortable. Physician anthropologists could eschew this messiness by never doing fieldwork where they have provided clinical care. But as anthropologists and as doctors, we deal in messiness. It is intrinsic to the human experience we seek to understand and to heal. Dealing with this intersecting messiness is central to the ethical practice of a physician–anthropologist.

Transformation and Ethical Practice

What I learned from my intimate involvement in Evelyn’s, Kima’s and other women’s pregnancies and incarcerations has been imbricated into how I care as a physician for other women. My medicalized recommendations for drug treatment programs as the answer to addiction, instead of mass incarceration, are more tempered—for I have seen how programs and CPS incarcerate mothers just as much as jail. I ask “history of present illness” questions differently, now knowing rather than imagining what many of these women’s lives are like outside of jail. I know that pregnancy means different things for different women enmeshed in poverty, addiction, and incarceration: some see it as disruptive, others as an incentive, and some as incidental as the flu. I know that the state is deeply involved in these women’s reproduction, a manifestation of “stratified reproduction” whereby women’s reproduction is differentially valued and managed (Colen 1986; Ginsburg and Rapp 1995). Together, these realizations have even changed how I counsel women—and not just those in jail—about contraception use. Moreover, in my research, I no longer strive to compartmentalize my identities as a doctor and an ethnographer, nor do I shy away from practicing medicine at my fieldsite. Although I did not adopt Evelyn’s or Kima’s babies, I engaged these women, sometimes uncomfortably, at the margin between the clinic and the drug corner.

The ethical conventions of medicine and anthropology both involve an ethic of care, in how we practically navigate relationships and our affective responses within them. We all carry multiple selves inside us at all times (Walzer 1994), whether medical, anthropological, or other. But the particular obligations of the clinical or ethnographic moment may necessitate temporarily putting some of our selves aside,Footnote 7 only to be reactivated in combination later. And there are also times when the ethical obligations of the moment necessitate a convergence of the clinical and ethnographic into novel forms of caregiving.

In Nancy Scheper-Hughes’ widely read essay “The Primacy of the Ethical,” she challenges the idea that one cannot be an anthropologist and a companheira at the same time (Scheper-Hughes 1995, p. 410). We are social beings, she reminds us; as such, our existence presupposes the presence of an other. This is the fundamental relational nature of ethnographic research, which forms the basis of an ethic of care in anthropology to tend to relationships in practice and honest, affective responses, not unlike medicine. Whether in managing privileged information, not abusing trust, or grappling with the potential limits of our compassion, these are concerns at the forefront of the ethical practice of medicine and anthropology.

We have a responsibility, Scheper-Hughes urges, to these others with whom we are enmeshed to write “personally engaged and politically committed ethnography” (Scheper-Hughes 1995, p. 419). And this, she argues, is the ethical practice of anthropology. Physician–anthropologists must infuse this political, relational ethical recognition into our dual roles. For me, this means not compartmentalizing the two. It also means not using one perspective merely to frame or inform the other—although that is useful. Rather, it means a willingness to change how I do fieldwork and how I care for patients because of what I learn from each. It means a commitment to dealing with the messiness of the different kinds of intimacies cultivated as an anthropologist and as a doctor.