Abstract
Extensively employed in reproductive science, the term fetal–maternal interface describes how maternal and fetal tissues interact in the womb to produce the transient placenta, purporting a theory of pregnancy where ‘mother’, ‘fetus’, and ‘placenta’ are already-separate entities. However, considerable scientific evidence supports a different theory, which is also elaborated in feminist and new materialist literatures. Informed by interviews with placenta scientists as well as secondary sources on placental immunology and the developmental origins of health and disease, I explore evidence not of interfacing during pregnancy, but of intra-action, or the mutual emergence of entities in simultaneous practices of differentiation and connection. I argue that attending to evidence that can be figured as intra-action enables us to recognize, account for, and attend to diffuse responsibilities for fetal–maternal outcomes that extend beyond mothers to the biosocial milieus of pregnancy. In reimaging the intra-action of placentas, a new understanding of what constitutes a ‘healthy pregnancy’ becomes possible.
According to biological sciences, placentas are transient organs necessary for mammalian fetal development that are produced by interaction of maternal and fetal cells, a process called ‘placentation’. Wooding and Burton refer to placentation as ‘an apposition of fetal to maternal tissue which has two main functions: to maximize oxygen and nutrient acquisition from the mother, but to minimize immunological rejection by the maternal immune system’ (2008: 1). The term ‘apposition’ suggests that maternal and fetal tissues are layered side by side, and that they interact with each other to produce a placenta (see Burton and Jauniaux, 2015). Prominent placenta scientist Robert Pijnenborg and coauthors provide another definition: ‘placentation means bringing two vascular systems, maternal and fetal, into close proximity’ (1981: 72). This ‘apposition’ or ‘close proximity’ is called the ‘fetal–maternal interface’, a term that, although rarely explicitly defined, is used extensively in reproductive science. For example, articles by Pijnenborg and Vercruysse (2008), Pujal et al. (2012), and Olivieri and Tadokoro (2013) have ‘fetal–maternal interface’ in the title but do not define the term in the text. A close reading reveals the presumption of a local connection at the placenta where maternal and fetal bodies are touching, fused, or blended. However, even as they are interfacing, mother and fetus are two distinct beings; it is only the placenta, a temporary organ, whose belonging is ambiguous. Indeed, Burton and Fowden (2015: 1) deem the placenta ‘a transient extracorporeal organ that interfaces with the mother, to sustain and protect [the fetus]’ (emphasis added). Enders and Blackenship affirm the real separation of mother and fetus in pregnancy: ‘the placenta facilitates respiratory exchanges and selective transfer of many molecules, while largely maintaining the integrity of maternal and fetal organisms’ (1999: 4).
In other words, central to the concepts of placentation and fetal–maternal interface is the assumption that mother and fetus are what feminist science studies scholar and physicist Karen Barad (2007) calls relata: pre-existing, interacting entities. However, there is considerable contemporary evidence reported in scientific literature that supports a different theory of pregnancy – one which, as I show, is also advanced and expanded by work in feminist science studies and theories of embodiment and pregnancy. In this article, I explore science that is establishing the ways in which mothers and fetuses exist not prior to, but as results of specific processes of placentation, while some processes do not differentiate them as such at all. Through reviews of literature and analysis of accounts from interviews I conducted with placenta scientists, I examine two scientific fields: placental immunology, which describes immune functions of placentas; and the developmental origins of health and disease (DOHaD), a theorization of how placental development and functioning impact lifelong health. I argue that scientific research intended to establish the nature of the fetal–maternal interface is instead upending it. Because science plays a central role in defining, circumscribing, and intervening in fetal–maternal relations, there is much at stake, both in how the tensions will be settled as well as in the meantime.
To reformulate the fetal–maternal interface, I follow paths opened by Karen Barad’s agential realism (Barad, 1998, 2007, 2010, 2012) and make sense of placentation as fetal–maternal intra-action. Intra-action describes the mutual emergence of entities in simultaneous practices of differentiation and connection, ‘recogniz[ing] that distinct agencies do not precede, but rather emerge through, their intra-action’ (Barad, 2007: 33). I am not merely choosing a different terminology; I am identifying tensions regarding fetal–maternal individuation that are already apparent in the scientific literature and among scientists. These tensions are, I argue, wrapped up with politics of reproduction. I consider how evidence for fetal–maternal interaction is implicated in the placing of responsibility for healthy pregnancy squarely upon pregnant women. Contrariwise, I argue that intra-action is a scientifically supported theoretical construction by which these politics of pregnancy can be challenged. In reimaging the intra-action of placentas, a new understanding of the agents of health and illness, and specifically of what constitutes a ‘healthy pregnancy’, becomes possible. Evidence of intra-action in the biosocial milieus of pregnancy demands that we recognize diffuse responsibilities, beyond those held by pregnant women alone, for fetal–maternal health.
Matters of Method and Theory
I conducted an empirical project consisting of interviews with people involved in the production of scientific knowledge about placentas, observation in laboratories and other locales, and examination of secondary scientific and social scientific publications (see Yoshizawa, 2014). 1 My research question was simple: what makes a placenta? Findings directed me to a so-called ‘naturalcultural’ (Barad, 2007: 32; Haraway, 2004, 2008) explanation for the production of placentas, theorizing that nature and culture are not distinct realms occupied by distinct kinds of beings. Rather, such a duality is a performative effect of knowledge-making practices. Out of that broader programme of research, one finding was that while the notion of the fetal–maternal interface is intimately related to scientists’ understanding of placentation, it is also a concept in tension.
In analysing the naturalculturalism of placentas, I am especially informed by Karen Barad’s book Meeting the Universe Halfway (2007), which is part of the ‘new materialist’ (see Hird, 2004) turn in social theory characterized by commitments to various conceptions of realism. Barad’s highly technical theory, which she dubs agential realism, deploys Niels Bohr’s interpretation of quantum physics, wave–particle duality, and the observer paradox to argue that the means and agents of measuring, perceiving, and understanding must be viewed as providing the conditions of possibility of observed phenomena in the first place. 2 With its focus on practices, discourses, and constructions of difference, agential realism is a performative understanding of scientific knowledge and of individuation. Performative approaches permit the validity, reliability, and reproducibility of science, but require accounting for the material involvement of the scientist and his/her apparatuses in the production of knowledge, addressing the ways in which science is a cultural and political enterprise. 3 Informed by agential realism, I consider that scientific experiments designed to characterize mothers, fetuses, and placentas are simultaneously productive of these entities, which are also simultaneously productive of each other in intra-action. I am particularly informed by the very distinctive challenge that agential realism poses to ‘particularism’, a term used by Barad as well as Vicky Kirby to refer to the idea, upended by quantum physics, that objects and entities exist independent of interactions with other objects (Barad, 2007; Kirby, 2011). As it applies to humans, I use the term ‘metaphysical individualism’: the presumption of the ontological, corporeal, legal, and moral integrity of the human individual.
Many other scholars working in a great many areas are similarly concerned with how what I am calling particularism and metaphysical individualism essentialize ontological differences, which they instead argue are material-discursive constructions, or what I dub ‘articulations’. I highlight an issue of this journal (Body & Society 16(3), 2010) that challenges ‘the idea of bodies as discrete entities, clearly bounded and differentiated’ (Blackman, 2010: 1). Authors explore how microchimerism (Martin, 2010), immunology (Shildrick, 2010), and alienable organs and tissues (Sobchack, 2010) puzzle, as Blackman puts it, biomedical norms of bodily integrity. Similarly, Blackman and Venn (2010) consider the ‘affective turn’ in social theory to theorizing bodies as processes materializing entanglement. These theorizations do not rely, as I do, on agential realism under the broader new materialist programme in order to explore naturecultures and intra- and inter-corporeality (see Waldby, 2002). Indeed, the question of whether there is much ‘new’ about new materialism has been highlighted by Ahmed (2008), who argues that a routine acceptance of feminist work as traditionally ‘antibiological’ or unconcerned with ‘matter’ caricatures incredibly important accomplishments in theory and analysis. Ahmed’s position paper sparked a debate with contributions from van der Tuin (2008) and Davis (2009), who carefully consider the newness of new materialism. For example, Davis suggests that it may be Ahmed who caricatures agential realism, since Barad discusses its informed feminist underpinnings at length. Although resolving this debate is beyond the scope of this article, I suggest that the banner ‘new materialism’ flags an orientation toward identifying and mobilizing ‘developments within in the natural sciences’ (Hird, 2004: 224) that support emergent, past, and ongoing feminist attempts to theorize naturecultures, inter- and intra-corporeality, and reproductive politics.
For example, this article draws from and contributes to an existing feminist literature rejecting the assumption that mother, fetus, and placenta are essentially discrete (Colls and Fannin, 2013; Haraway, 2008; Hird, 2007; Irigaray, 1993; Maher, 2001, 2002; Oliver, 1998; Simms, 2009). Simms finds that common representations of the placenta as a barrier (i.e. the ‘placental barrier’) affirm the illusion that it is a thing separate from the mother and fetus. However, she argues that ‘matrotopy’, the consumption of the ‘mother’s substance’ by the fetus via the placenta and breast milk, undermines this representation (Simms, 2009: 264). In matrotopy, toxic substances can be exchanged. For instance, polychlorinated biphenyls (PCBs) from old household appliances can be acquired by the maternal body through inhalation, consumption, or skin contact. These can be ‘gifted’ (Hird, 2007) to the fetus when the placenta cannot differentiate between them and estrogen, a hormone productively shared with the fetus. I am informed by agential realism to argue that in the consumptive intra-actions of matrotopy, the placenta does not articulate PCBs and estrogen as different. The differences made by this nondifference can be immense: PCBs are associated with breast cancer, uterine and cervical cancer, and skeletal, sexual, and mental developmental issues. Alongside this feminist work, I explore what challenges to fetal, maternal, and placental particularism are apparent in intra-actional biologies emerging from placenta science.
Theories of Placental Immunity: Difference Tolerance
Reproductive scientists report that on the fifth day of human embryo development, a blastocyst forms (Huppertz et al., 2006; Pijnenborg et al., 1981). These are 50 to 100 cells that articulate the earliest cellular differentiation of embryonic life: the ‘inner cell mass’ becomes the embryo, while the outer cells become the trophoblasts, which are the cells that make the placenta. The placenta forms during what is called the ‘trophoblast invasion’ (see Carter et al., 2006; Pijnenborg et al., 1981). Trophoblasts multiply into the maternal endometrium, the uterine lining that is grown during the menstrual cycle (Haig, 1993), and ‘remodel’ (see Lyall, 2005) uterine blood vessels by taking the place of the cells (endothelia) that give the vessels structure. This replacement widens the blood vessels to provide steady blood flow. Inside the uterus, trophoblasts form a disk-shaped sac lined in the inner space with tree-like structures called ‘villi’ that contain capillaries and blood vessels carrying fetal blood to and from the umbilical cord. Maternal blood coming from the remodelled vessels circulates around these villi beginning in the second trimester, keeping the embryo in a low oxygen state for the first trimester to protect the embryo, scientists surmise, from mechanical injury and oxidative stress as its cells rapidly divide during early pregnancy (Burton and Jauniaux, 2004). Later in pregnancy, the placenta becomes richly perfused with maternal blood to create the dynamic intradigitation that instantiates the physical exchanges of pregnancy. From mother to fetus, placentas can transport oxygen, glucose, amino acids, proteins, hormones, maternal antibodies, and alcohol; some viruses such as rubella, cytomegalovirus, varicella-zoster, enteroviruses, and HIV; various drugs including opiates, benzodiazepines, antibiotics, antivirals, anaesthetics, analgesics, and cannabinoids (Pacifici and Nottoli, 1995); and many xenobiotics (human-made chemicals ostensibly ‘foreign’ to the organic body; Simms, 2009). From fetus to mother, placentas transport waste products such as carbon monoxide, urea bilirubin, and cellular remnants including free pieces of DNA (Bianchi, 1998, 2004; Martin, 2010; Roberts et al., 1996). Not merely vessels of exchange, placentas produce hormones to alter blood vessels, prepare for birth, and regulate fetal neurodevelopment (Bonnin et al., 2011).
These intimate, complex exchanges are first and foremost understood in placenta science to be immunological. In fact, the role of the immune system is so crucial that an impaired immune relationship is thought to be associated with all the major diseases of pregnancy. One scientist explained: I strongly believe the placenta is…a unique immune organ…as it controls the maternal immunity locally at the maternal–fetal interface to protect the fetus. Now when that immunity is compromised, [when] immune tolerance is compromised, that’s when problems start, such as recurrent spontaneous abortion [i.e. miscarriage], preterm birth, preeclampsia, gestational diabetes – all these problems are related to disregulated crosstalk between the placenta and the maternal immune system.
I want to assess the fetal–maternal relationship in terms of intra-action by looking more closely at this immunological ‘control’. To understand how scientists see the placenta as immunological, as well as why this is important for thinking critically about reproductive politics, it is first necessary to detour into the history of immunology, which shows a surprising confluence of politics with biology (see Cohen, 2009; Esposito, 2011; Haraway, 1991; Howes, 2007; A Martin, 2010; E Martin, 1990; Weasel, 2001). Originally a legal term referring to ‘privileges and entitlements conferred on individuals or collectivities that exempt them from political obligations and responsibilities’ (Cohen, 2009: 40), immunity prescribed conditions under which an individual could stand apart from a community founded on obligations of universal citizenship. By performing the role of exception, immunity reinforced the natural rule of universal obligation articulated in its antonym community, which refers to ‘both a “municipal corporation” and to a sharing “in common”’ (Cohen, 2009: 43; see also Esposito, 2011). In western democracies, legal and political institutions proffer rights to, expect responsibilities of, and offer representation to individuals (Wilkin, 2008). Likewise, the biological understanding of immunity affirms ‘that living entails a ceaseless problem of boundary maintenance’ of an entity known as ‘self’ (Cohen, 2009: 8). According to the self/nonself theory of immunity first articulated by Burnet and Fenner (1949; see also Tauber, 1994; Vaz, 1978) and presently the predominant explanation for organismic individuation in biology, numerous specialized cell types read antigen ‘signs’ against a predetermined rubric of self and nonself; ‘foreign’ antigens set off a cascading immune reaction leading to the rejection of such matter. In relating the biological and political significances of the term, Cohen writes: with immunity as its avatar, modern biomedical dogma holds…that as organisms we vitally depend on a perpetual engagement against the world to maintain our integrity or indeed our selves.… Immunity incarnates ideas about human being culled from modern politics, economics, law, philosophy, and science, which then belatedly achieve scientific status when immunity inoculates them into the living organism and thereby validates them as essentially ‘natural.’ (2009: 8)
The self/nonself theory of immunology is very influential in the natural sciences, even though there is wide acknowledgement that it does not effectively account for pregnancy (see Martin, 2010). This is because, theoretically, foreign fetal antigens should spur maternal immune rejection of the placenta, resulting in complications or the end of the pregnancy (e.g. miscarriage, preterm birth, or stillbirth). If the maternal immune system is truly tasked with this maintenance of maternal self, why, reproductive scientists ask, are fetuses not routinely rejected (see Esposito, 2011)? Indeed, this was a common point of intrigue for the placenta scientists I interviewed. One said: I’m an immunologist. And the placenta is an organ which is, from the immunological point of view, completely different from the rest of the body. And the reactions inside the placenta are completely different from what students learn in immunology.… So, for example, the major differences: you learn that foreign organisms will be rejected, foreign tissues, [but] that does not occur in the placenta. So it [the conceptus] is accepted. And somehow the immune system of the mother not only tolerates it, but it supports the foreign cells and foreign body of the embryo. It does not reject it. And that makes it really very interesting. And the mechanisms, a lot of different mechanisms behind that, are very much surprising…reacting in another way than you learn how things should happen in biomedicine.
A prolific science of reproductive immunology is concerned with reconciling the contradictions between observations and foundational theories. Now is an era of rapidly expanding work, although the project can be traced to Peter Medawar (1953), who first addressed the problem by characterizing the placenta as a ‘semi-allogeneic graft’ (allo meaning ‘other’) that is similar to a successful transplant graft in its immunological compatibility and the willingness of the recipient’s body to accept the tissue. This characterization was prominent among participants: Usually I describe it [the placenta] as a semi-allogeneic graft. Because I think, from my standpoint, I study immune tolerance. And one of the things we are very interested in is how the mother tolerates the fetus. And so looking at it from a graft standpoint is the most convenient way to do that.
Noting the theory is ‘convenient’ suggests it has analogical value in explaining observations of ‘tolerance’. But others did not speak of analogy; rather, they presented placenta-as-graft as a ‘fact’ to be explained according to theories of transplant biology: Since the fetus is in allograft, theoretically if we use simple laws of transplantation, the fetus should be rejected by the mother’s immune system. But it’s not rejected. We are all born, so we survive in the mother’s womb. So there has to be a mechanism of immune tolerance at the placental level for the fetus to survive.
Rather than characterizing it as a process of ‘boundary maintenance’, these scientists argue that tolerance of difference is instead central to successful pregnancy. For example, Hunt and Langat write, ‘adjustments of the normal activities of human maternal and fetal immune systems are required in order for pregnancy to be established and maintained in the face of genetic dissimilarities’ (2009: 462). An article in Nature dubbed this tolerance ‘immunity’s pregnant pause’ (Pearson, 2002), while Waldorf and Nelson point out that there are ‘changes in maternal systemic immune responses [which] are placenta-induced and result in a temporary change in what the mother’s immune system considers “self”’ (2008: 634). Trophoblasts are involved in the creation of ‘an illusion of self’ (Roberts et al., 1996: 298) or ‘temporary self’ (Trowsdale and Betz, 2006: 241) via immunological ‘ignorance’ (Trowsdale and Betz, 2006). In other words, there ostensibly is a reshaping of the bounds and protectionism of the maternal self, but only for a fixed period of time during pregnancy. Others argue that, in evading maternal immune response, the placenta is a successful ‘parasite’. For example, Lowry writes, ‘the one thing I am sure of is that the placenta is a neuroendocrine organ par excellence in the guise of a parasite (or is it the other way round?)’ (2008: 703). Similarly, Clark et al. write that ‘the embryo is most akin to a parasite, and pregnancy is most akin to a host–parasite interaction’ (1999: 5).
The metaphors and analogies differ, but according to all of these constructions, maternal selves are merely paused or perhaps tricked, but will return to integral selves, as they were before pregnancy. In other words, although employing notions such as ‘tolerance’ or ‘parasite’ to accommodate difficult immunological evidence, these theorizations still preserve the self/nonself duality as well as the attendant corollary that mother and fetus are metaphysically distinct individuals, which only mingle temporarily at the fetal–maternal interface.
Nondifference of HLA-G (Human Leukocyte Antigen)
I do not think that scientists are satisfied with these theorizations, but are instead trying to grapple with making evidence commensurate with other biological phenomena that are already explained by widely accepted theories. Participants in my study were keenly aware, and indeed produced, much of the research that indicates that pregnancies bear articulations of difference that are not confined to fetal and maternal selves with distinct self/nonself corporeal boundaries. Scientists admit they do not fully understand why some pregnancies are tolerated and some are rejected – or even if ‘rejection’ and ‘tolerance’ are adequate theories of pregnancy. One scientist I interviewed noted, Medawar was very much, ‘how does the maternal immune system deal with this, the placenta as a graft?’ But I mean, the graft is a 20th-century surgeon’s construct, and it [the placenta] never evolved as a graft.
Indeed, as Christiansen points out, ‘the HLA antigens responsible for strong graft-rejections are not expressed on trophoblast[s]’ anyway (1996: 276). I argue that the findings that pregnancy does not conform to the predominant theory of self, and specifically that certain HLA antigens are not expressed on trophoblasts, contradict the fetal–maternal interface as a theory of pregnancy. Instead, they support fetal–maternal intra-action.
To explain this requires an overview of how antigens work. The major histocompatibility complex (MHC) is a molecule on the surface of cells (Bainbridge, 2000) that presents proteins called antigens (a portmanteau of ‘antibody’ and ‘generator’). These molecules bind to ‘foreign’ or pathogenic fragments, which may result in the production of antibodies and set off a cascading immune response. As such, antigens and antibodies are articulated by recognition and joining, rather than in foreignness and rejection, which is a clear example of intra-action. HLA is the MHC found in humans. Individuals receiving a transplant must share with the donor the same type of HLA if the transplant is to be accepted. However, the genes that encode HLAs are extremely variable from person to person (i.e. they are ‘polymorphic’) leading to many alternative forms. Transplant recipients therefore still require immunosuppressive drugs to help ensure the graft is not rejected, pointing to the tenuous hold on nondifference that transplants articulate (see Shildrick, 2010).
Trophoblasts are conceived in the mixing of two genetic lineages, maternal and paternal. The articulation of conception therefore entails one level of nondifference of maternal and paternal genes that is meaningful for the conceptus-as-conceptus; in other words, it makes genetic conceptus-being possible. However, for the maternal body, a different difference makes the conceptus intelligible. Being of partly paternal origin, trophoblasts articulate antigens that are inherited from the father. Theoretically, this should spur immune rejection. However, trophoblasts do not express the two main human polymorphic antigens, HLA-A or HLA-B. Scientists have found that a different antigen, HLA-G, is expressed by the trophoblasts – in fact, it is unique in that it is found only on trophoblast cells and thymus cells (the thymus is an organ responsible for eduating immune cells to ‘tolerate’ ‘self’; Bainbridge, 2000). Unlike HLA-A and HLA-B, HLA-G’s polymorphism is limited (Bainbridge et al., 2001), and thus genes that encode for it, inherited from either maternal or paternal sources, are likely to be the same. Thus, a scientist suggested to me that the trophoblasts do not participate in the ‘semiotics of self’ – that is, the reading of self and other instantiated by immune cell interactions with antigens – expected of other cells: Even if you put these trophoblast cells from the uterus and you transplant in the male mice liver, no rejection happens. But the trophoblast cells still keep the invasive capability. The receptor of the transplant, it doesn’t make any biological response against that.
HLA-G materializes out of genetic mixings that articulate antigen nondifference in pregnancy, intra-actions that do not necessarily disambiguate mothers and fetuses. Instead, the trophoblasts are recognized by the maternal immune system as ‘self-same’ – even though they are fetal in origin. This is not ‘tolerance’; it is fetal–maternal nondifference. The consequences of this nondifference concern the very possibility of reproducing an infant in and from the maternal body in genetic and immunologic collaboration with the paternal body.
Even maternal ‘selves-to-their-selves’ may not be self-same in the articulation of immune activity in pregnancy. Three out of four women with rheumatoid arthritis, an autoimmune disorder, experience spontaneous improvement of their symptoms during pregnancy, but women who have never had biological children are at an increased risk for developing the disorder (Nelson and Ostensen, 1997). Conversely, bidirectional exchange of cells between mother and conceptus (or between twins) may contribute to the development of autoimmune disease, such as type 1 diabetes and neonatal lupus syndrome (Waldorf and Nelson, 2008). Aryn Martin attends to this bidirectional exchange, writing: in microchimeric bodies, cells that are coded as ‘not-self’ are living and reproducing happily in body-nations that are not their ‘own’. In this way, ontologies are shifting in light of the unexpected, as are appropriate metaphors of what the biological (and indeed social) ‘self’ is. (2010: 25)
Martin shows how such findings require scientific researchers to ‘reorient’ their research objects away from particularism. What are the bounds of interacting selves in autoimmune disorders impacted by pregnancy? The fetal–maternal interface, if we are to maintain this terminology, can be located throughout fetal and maternal bodies and not just at the placenta; it is also located in arthritic joints, infants, and twin siblings, long after the umbilicus is cut. Perhaps, though, it is time to refine the terminology.
Developmental Intra-action
Yet another body of research establishes an even more diffusive spacetime of pregnancy spilling beyond the bounds of maternal and fetal selves. A vast literature, particularly examining longitudinal cohort data, argues that experiences of health and disease throughout life can be traced to the conditions of development in the womb. This is called the developmental origins of health and disease (DOHaD). A participant described both the basic points of DOHaD and the context in which the science is currently developing: You know one of the things that probably drives a lot of reproductive physiologists a little crazy is that…at least in the United States…the study of reproduction is one of the least well funded. And it’s not the worst funded, but cancer and AIDS are some of the best funded. And things like obesity have gotten huge attention as well. But the thing is, that drives me nuts, is that those are, maybe, those are preventable. And how are they preventable? If you go back, you go back in life. And you think about what’s going on in the womb, and you think about early developmental processes, and I think from that standpoint, it’s very, very valuable to make people aware.… You know, more emphasis should be placed on events that occur very early on in life, and that occur in the womb, to give an understanding to how we might live a better quality of life later on.
The two key ideas of DOHaD are developmental plasticity and fetal programming. Developmental plasticity is the capacity of a developing organism to corporeally respond to its environment by modulating the morphology and functioning of its organs (Bateson et al., 2004). Forms attained by early infancy, like the number of structural units of the kidney, heart, and pancreas (McMillen and Robinson, 2005), become durable and then more or less persist through to adult life. This is called fetal programming (Belkacemi et al., 2011; Godfrey, 2002; Godfrey and Barker, 2001). Extra- and epi-genetic generational adaptations outside the slow process of random genetic mutation are therefore possible, which is advantageous if the infant becomes phenotypically ‘matched’ with the environment beyond the womb (Barker, 2004: 588S).
However, scarcity of nutrients to the fetus during development, which can be caused by variations in the ability of the placenta to deliver nutrients to the fetus because of incidental changes, pathology (Godfrey et al., 2007), or generational patterns of biological inheritance, can mean that some fetal organs, such as the brain (Jones, 2005: 22), are favoured over others. Crucially, the ways in which a placenta develops and functions are determined in part by maternal physiology and metabolism, which is also ‘programmed’ during a time in utero. The totality of these dynamic variables means that the environment that the fetus is ‘programmed for’ may not match what the infant finds when born. This may instantiate chronic limitations of the fetus’ organs to meet its needs throughout the lifecourse, resulting in increased risk for disease in later life. For example, scientists have shown that low birth weight, which may result from nutrient restriction to the fetus via the placenta, is associated with later incidence of coronary heart disease (Barker, 2004), hypertension (Eriksson et al., 2000), diabetes mellitus (Phillips, 1998), renal disease (Vikse et al., 2007), and lung cancer (Barker et al., 2010). These diseases are associated with the ‘placental diseases’ of intrauterine growth restriction and pre-eclampsia (Belkacemi et al., 2011), which compromise nutrient availability. In this way, even though adults are no longer ‘umbilically attached’ to their placentas, they are still intra-acting with them. Newer research is continually affirming and expanding the hypothesis: for instance, a mouse model linked maternal diet to early programming of later mental illness in offspring (Bonnin et al., 2011). As such, as Jackson writes, ‘it is very clear that the nutritional state in which women enter pregnancy is of very great importance in terms of the future health of the population, and healthier women will undoubtedly reduce the burden of disease for their children at later ages’ (2000: 19). It is also clear, though, that the nutritional state of previous generations has an impact on experiences of health and disease today.
Although fetal programming research predominantly focuses on the effects of maternal nutrition on the later life of the conceptus (Harding, 2001), diet is not the only factor of interest to DOHaD scientists. The hypothalamic-pituitary-adrenal (HPA) axis, which controls reactions to stress and effects homeostasis in bodily processes, may also be highly programmable in utero. Sullivan et al. explain: when a pregnant animal is exposed to a dangerous environment that requires increased vigilance, a stress signal is transmitted to the fetus. This signal programs the fetal HPA axis and related behaviours to develop increased vigilance, a necessary adaptation for survival in a hostile environment. Accordingly, prenatal stress has long-lasting effects on the HPA axis of adult offspring, programming a persistently hyperactive system. (2008: 159)
This surmises that maternal ‘stress’ – a concept articulating women’s experiences, social positionality, personality, and chronic or incidental sources of stress – may affect fetal physiology.
Although it may at first seem that DOHaD confirms that women are largely accountable for what happens during pregnancy, I argue, crucially, that what DOHaD actually establishes is the horizontal and vertical diffusion of responsibility for healthy pregnancy throughout the places and times that contextualize, for example, what women eat, experience, feel and are exposed to while pregnant, as well as what their mothers, grandmothers, and so on ate, experienced, felt, and were exposed to while they were pregnant (Yoshizawa, 2012). Uncovering this transgenerational epigenetics (Daxinger and Whitelaw, 2012) is an incredible development vis-à-vis critical theories of embodiment since it points to the embeddedness of the body in inherited and lived cultural and geopolitical worlds (see Lock, 2013a, 2013b). Fetal programming is a matrilineal inheritance endowed by the capacity for developmental plasticity, itself an intra-actional openness to a world beyond the womb. A fetus and mother, and their experiences of health and disease are already intra-acting in their ancestors, their food, their social lives, their emotions, and their exposomes. Put simply, fetal programming comprises phenomena that prompt us to wonder where the supposed fetal–maternal interface can really be located – indeed, whether we should maintain this theory of pregnancy at all. Responding to DOHaD requires fundamental changes to the study, treatment, and prevention of disease via the re-theorization and re-adjudication of responsibility for healthy pregnancy.
Intra-action and the Politics of Reproduction
Equivocation as to whether the fetal–maternal interface is a process (placentation), an entity (placenta), or something else is apparent in the scientific literature on pregnancy. There is also equivocation around what the ‘agents’ of this interface/ing/er are, and what their roles are. For example, in one of the few explicit definitions of the fetal–maternal interface I found, Erlebacher writes, ‘the maternal–fetal interface is the interface between the [maternal] uterine mucosa and the extraembryonic tissues of the developing conceptus’ (2013: 389). Here, it is implied that the placenta is produced by two touching tissues that actively engage one another: the uterine mucosa is not a passive player in embryo implantation but rather undergoes a specialized tissue reaction termed decidualization to support the development and function of the placenta. (2013: 389)
This suggests that the fetal–maternal interface is really an interfacing. Yet other articles suggest the placenta is the interfacer: that is, an interfacing thing that does the bringing-together. For example, Rampersand et al. write that ‘the human placenta interfaces the mother and fetus to prevent immune rejection, to transfer nutrients, to dispose fetal wastes, and to secrete hormones that sustain pregnancy’ (2011: 19). Still other scientists view the placental organ as synonymous with the fetal–maternal interface, such that the placenta is the interface. When asked if they are different, a scientist whom I interviewed said: You know, I’d say they’re [the fetal–maternal interface and the placenta] one and the same actually. Well, unless to be a bit more expansive, and [include] say fetal membranes, but that really is placental tissue as well, at the maternal–fetal interface.
The equivocation is revealing. It points, I argue, to ontological tensions in placenta science regarding the metaphysical individualism of mother and fetus. Indeed, existing alternative scientific theorizations reflect attempts to reconcile these kinds of tensions. For example, although predominant in immunology, the self/nonself theory is not the only one available to make sense of relationships between organisms. Weasel contrasts it with a competing theory called the danger model, developed by immunologist Polly Matzinger, which suggests that immune systems are tasked not with the maintenance of self, but rather ‘rely upon a context-dependent, biosemiotically mediated definition’ of ‘danger’ read in signs of cell injury (2001: 39). If the fetus poses no signs of danger to the mother, it is not rejected. With this theory, we can see that otherness and self are not essential to explain immune processes of pregnancy.
There are many real and potential consequences of the particularism advanced in some streams of reproductive science and medicine. For example, knowledge about fetal–maternal interfacing is co-opted to prescribe women’s reproductive rights and responsibilities, particularly regarding ‘fetal subjects’ which are pre-constituted as legal individuals. Wilkin is critical of both ‘pro-lifers’ and feminists for relying upon metaphysical individualism to parse out sides in abortion debates, arguing that this ignores ‘fetal dependency’ on mothers (2008: 98). Contrariwise, van der Ploeg (2004) shows that acknowledgement of dependency means women become responsible to other ‘selves’ by virtue of their role and behaviours during pregnancy: women now are being medically treated for problems that used to belong to others, that is, for problems that used to be their children’s and male partners’.… [T]his development challenges any self-explanatory use of the notion of a patient’s bodily self-determination in medicine, because it renders unclear which selves or whose bodies precisely are involved and to what extent, or even how many selves and bodies exactly are involved. (2004: 156)
According to van der Ploeg, one outcome is premature implementation of immune therapy for infertility, an argument echoed by Howes, who points to ‘tunnel-vision and occasional recklessness of the fertility industry’ (2007: 248) that focuses on implantation and gestation immunology rather than on other aspects of immune functioning during pregnancy. Howes is critical of the: ontological commitment to the idea that mother and fetus are two clearly distinct immune entities and [the] value-laden assessment of how these two entities relate to one another. Neither…is mandated empirically. (Howes, 2007: 179)
It is also apparent, though sometimes subtly, that women are figured as the most significant threat to fetuses in dominant reproductive politics. Pregnant woman are bombarded with a constant stream of advice, warnings, and denunciations regarding their behaviours around eating, smoking, exercising, using drugs, and other activities as a means to protect the fetus from the mother (see Longhurst, 2005; Marshall and Woollett, 2000). A scientist I interviewed imagined clinics that could monitor and educate women pre- and during pregnancy, particularly focusing on women’s ‘modifiable behaviours’: I think we need a complete change. So we need to run pre-pregnancy planning clinics. So you could actually give advice. Now to have some biomarkers as well, whether they’re genetic or protein markers that can predict risk, that would add to the information. And that’s what we’re doing. We’re developing algorithms that predict risk and we’ve got a sort of three-tiered approach. We’ve got some things that we can say, prior to pregnancy would help, and some things that you can measure during pregnancy that if you add, it improves the quality of the algorithm – so the predictive capacity of the algorithm. But a lot of these things are amenable to modification. So you know, don’t get fat before you plan a pregnancy. Or if you are fat, reduce your weight. Don’t smoke. Have a good quality diet. Don’t drink alcohol. There are simple, there are very simple things that people can do for themselves besides their genetics. You know, people say, ‘we don’t want to know about genetics because we’ll be doomed to failure.’ But you’re not, because a lot of these are amenable to modification. So we know there are lots of lifestyle things that confer risk.… But just things like, we could get back to the pre-pregnancy clinic type thing, we could educate women. We could educate girls.
I am not arguing against making a difference with the kind of interventions this scientist proposed. On the contrary, I am arguing that acknowledging evidence of intra-action portends an even greater scope of responsibility and accountability than has been previously recognized in science, medicine, and politics. Recognition of a greater scope of responsibility and accountability would likely prompt a different set of interventional priorities that do not focus solely on behaviours of individual pregnant women. In addition, fetal–maternal intra-action suggests that researchers need to consider what it means not only to have agencies articulated in our bodies that do not defer to our intentionality as ‘subjects’ or ‘selves’, but to reconsider the very categories of human and nonhuman as they apply to pregnancy. For example, in light of what I see as the intra-actions of pregnancy, Simms calls for the development of a ‘placental ethics’ (2009: 274) that acknowledges intercorporeal debts and responsibilities to the environment. She writes that ‘the fetal ecosystem is nested in the ecosystem of the mother’s body, which is nested in the larger ecosystem of the Earth’ (2009: 274). For Simms, ‘the infant is the missing ecological link between human beings and the natural world: the damage to our environment is not just “out there,” but goes as deep as our placentas’ (2009: 271).
The intra-action of pregnancy attests to the inability to locate an ‘enemy’ or even a ‘self’ in any one ‘body’, process, or thing and certainly not solely in the metaphysical individual agent, ‘pregnant woman’. This raises difficult questions of how to be pregnant better and who can and should be taking actions to do so – an area greatly in need of interdisciplinary research and serious society-wide discussion.
Conclusion
When one looks closely at scientific literatures that describe details of how placentas grow and develop, it becomes clear that the fetus does not precede the placenta; the fetus cannot develop and grow without the placenta providing one set of conditions and resources for its becoming. Fetal and placental DNA is created in the same moment of conception, and so neither does the placenta precede the fetus. This DNA has been provided, in part, by the mother, who also establishes conditions for the possibility of fetal and placental becoming in her womb. By this, she also becomes ‘maternal’, a condition which, in the biological events of pregnancy, does not precede placenta or fetus. In providing a channel for the movement of all that is necessary for the fetus to survive and thrive, the placenta is not what brings fetal circulation into contact with the maternal, but what materializes and makes possible fetal circulation in the first place. It also makes maternal circulation – circulation with the quality of maternality – possible in the first place. Likewise, placentation is possible only through the materializations of these circulations. In other words, the instances of supposed meeting or ‘interfacing’ are the same possibilities of becoming for fetus, mother, and placenta.
Scientific evidence establishes this (see Burton and Fowden, 2015), and yet the notion of the fetal–maternal interface is still very important to contemporary placenta science. In response to a question regarding what topics define the ‘cutting edge’ of placenta science, one participant responded with the fetal–maternal interface, relating its importance to pregnancy outcomes, experiences of women, and social effects: Well still the maternal–fetal interface, as it’s called, the whole concept of tolerance by the mother of this fetal allograft, this foreign individual, and how that works. Probably a large portion of the pregnancy losses are due to failure of those mechanisms. So I think if we understand more about that, we can do something for infertility, recurrent pregnancy losses – which is a significant problem, and a very emotive problem in society. You know, it affects a lot of women that’s for sure.
However, as much as cutting-edge research is establishing the nature of the fetal–maternal interface, it is also establishing the nature of fetal–maternal intra-action. Out of the tension, whole new sets of questions about political matters of reproduction are prompted. Who is responsible for what, and to what degree, concerning matters of pregnancy? I do not offer any answers, except to argue that intra-action means that responsibility becomes imminent and immanently relational, and this responsibility falls not just on pregnant women but others, including researchers – scientific or social scientific – by virtue of their participation in the production of knowledge about reproduction. When I consider the intra-actions of pregnancy, I am prompted to reject the problematic assumption that pregnant women alone can and should modify their behaviours in order to affect outcomes. We require more interdisciplinary, intersectoral, international, and intergenerational collaboration in being pregnant better together.
Footnotes
Funding
The author received financial support for the research of this article from the Social Sciences and Humanities Research Council of Canada through a Joseph-Armand Bombardier CGS Doctoral Scholarship, a Queen's University Tri-Council Recipient Recognition Award, a Queen's Department of Sociology Blakely Trust Travel Scholarship, a Queen’s Graduate Award, and a Queen’s Department of Sociology Thesis Completion Grant.
