Looking to the future: the development of an artificial womb

In 2017, two research teams reported promising animal trials of an advanced incubation technology. While experimental models vary, each team was able to externally gestate extremely premature lamb fetuses from the equivalent of approximately 22–24 weeks of gestation in a human for several weeks using a partial artificial womb. Variously referred to as “the biobag” (Partridge et al. 2017), “Ex Vivo Uterine Environment Therapy” (or EVE) (Usuda et al. 2017), and “EXTra-uterine Environment for Neonatal Development” (EXTEND) (Hornick et al. 2019), these technologies replicate the conditions of the uterus, acting as a completely closed system in which the fetus is suspended in artificial amniotic fluid. An external pump which replicates the placenta delivers nutrients and flushes toxins. Research is ongoing in the United States, Japan, Australia, and the Netherlands, with hopes that a prototype will be ready for human trials within the next five years (Bonito 2019). The clinical purpose of the technology is to treat neonates born at the current cusp of medical viability (the point at which a fetus has a chance of survival outside the pregnant person’s body). Morbidity and mortality for neonates born before 28 weeks remain extremely high because their organs have not sufficiently developed to function outside the liquid environment of the womb. The artificial womb will depart from previous forms of emergency neonatal care. While existing technologies treat the complications of premature birth, artificial wombs will prevent these complications from arising to begin with by replicating the conditions of the uterus. If the technology is as successful as anticipated, it could drastically improve outcomes for extremely premature babies by effectively extending gestation until the neonate’s organs have fully developed.

A “moral imperative” for artificial wombs?: improving health outcomes for pregnant people and neonates, and increasing reproductive choice

The primary scientific and medical justification for these technologies is to save the lives and health of extremely preterm babies during wanted pregnancies (Partridge et al. 2017; Usuda et al. 2017). Researchers have also indicated its potential benefits in protecting pregnant people in situations where medications and stem cell or gene therapies are being delivered to the fetus. For these research teams, artificial wombs are a potentially groundbreaking innovation in maternal and preterm care. Popular coverage of the technology, and some of the humanities literature, can tend toward speculative accounts that detract from the primary uses of artificial wombs. But scholars writing across the disciplines of bioethics, law, feminist theory, and sociology have also cited the health risks of pregnancy as a positive incentive for developing this technology (Pence 2006; Singer and Wells 1983; Smajdor 2007; Kendal 2015; Coleman 2004). As biologist Adinolfi writes, beyond “free[ing] the mother from the discomfort and potential risks associated with pregnancy” (2004, p. 570), the artificial womb may be useful in numerous situations that could be dangerous for both pregnant person and neonate. These include prenatal surgery and the onset of preeclampsia or blood incompatibilities between pregnant person and fetus. The feminist literature on artificial wombs has often emphasized that the technology should be considered a worthwhile investment not primarily because of its potential as a life-saving intervention for neonates, but because of its capacity to alleviate the impact of pregnancy on women’s health (Firestone 1970; Smajdor 2007; Kendal 2015). Smajdor and Kendal, respectively, observe that the fact that gestation can produce side effects ranging from nausea to death should provide sufficient rationale for creating artificial wombs. Smajdor argues that society in fact has a “moral imperative” to develop this technology to mitigate the risks of pregnancy. Pregnancy, she writes, has potentially negative physical, social, and financial consequences. She proposes that “the desire of women to be able to reproduce as men do, without risking their physical and mental health, economic and social well-being, and crucially—their bodily integrity” (2007, p. 340) should be sufficient for resources to be allocated to ectogenesis. Kendal argues for the benefits of artificial wombs to improve gender equality by assuaging the effects of pregnancy and birth on women and providing an additional reproductive choice. She writes, “pregnancy and childbirth are known to pose numerous health risks, with some ‘normal’ pregnancy-related symptoms including morning sickness, dizziness, headaches, bone and muscle aches, loss of visual acuity, bleeding gums, breathlessness, heartburn, varicose veins and hemorrhoids” (3). She holds that these issues are not taken seriously as health concerns and makes the case for making ectogenesis available as a “technological alternative” (18) to pregnancy. Romanis’s work (2020), in contrast to Kendal’s more speculative approach to artificial wombs, is attentive to the potential initial clinical use of the technology in the later stages of gestation where continuing pregnancy may pose a substantive risk to the health of the pregnant person. Like Kendal and Smajdor, however, she proposes that artificial wombs could be made available in these situations as an elective choice for pregnant people and critiques the way contemporary English law might prohibit this option. I agree with Romanis that paternalistic laws that limit pregnant peoples’ discretion over their reproductive lives are problematic. I also agree with Kendal and Smajdor that the possibility that artificial wombs could alleviate the health risks of pregnancy is a worthy goal. But I wish to examine the way that the literature on artificial wombs has been constructed around an implicitly neutral classless, raceless, subject who is understood to have access to reproductive care and to the conditions under which reproductive choice can be exercised. Artificial wombs are then approached as a potential means of improving upon the status quo by offering an additional reproductive option. In approaching artificial wombs as a prospective universal good (a means of improving care for pregnant people and neonates and expanding the choices available), the literature implies that the beneficiary of this technology is essentially any pregnant person. The tacitly neutral pregnant person who stands to benefit from the artificial womb is a subject whose health and reproductive autonomy can be presumed to improve with its introduction. Choice and autonomy are fundamental to protecting peoples’ control over their reproductive lives. But this framing of artificial wombs as a kind of universal good, and its beneficiary as any pregnant person meets a challenge in the structural contexts that shape the extent to whether autonomy can be acted upon, and to whom reproductive choices are available. Engaging a critical lens informed by a reproductive justice framework, in what follows I demonstrate that the potential of the artificial womb as a universal good is called into question by (1) the persistence of racialized stratification in health outcomes for pregnant people and neonates and (2) historical and contemporary instances of classed and racialized reproductive coercion. Drawing on these examples, I argue that contrary to claims that the artificial womb will benefit women and neonates as a whole; in its current trajectory, the technology is likely to exacerbate existing inequity. In the final part of the paper, I offer an alternative vision of how we might take a justice-informed approach to this technology. Beyond contextualizing considerations of artificial wombs as a means of improving reproductive choice within a broader consideration of reproductive harms, and beyond acting to prevent the technology from increasing stratification, I propose that we might reorient the artificial womb by reframing the questions that are asked at each stage of its design, development, and implementation. In engaging a reproductive justice framework and centralizing those who have been most marginalized within reproductive health and care, I argue we might trace an alternative path for artificial womb technology.

Situating artificial wombs in the context of reproductive justice

Putting an approach to artificial wombs as a universal good for pregnant people and neonates in conversation with a reproductive justice framework helps us to understand how racialized inequity and patterns of reproductive coercion pose a barrier to this technology functioning to improve care for all. A reproductive justice approach allows us to situate claims about healthcare and choice within the structural, institutional, and relational contexts that shape the nature of this care and the extent to which choice can be exercised. Reproductive justice is a grassroots initiative led by Black women in the United States. The SisterSong Women of Color Reproductive Justice Collective defines reproductive justice as “the complete physical, mental, spiritual, political, social, and economic well-being of women and girls, based on the full achievement and protection of women’s human rights” (Ross 2007, p. 4). The founders of the reproductive justice movement identified that while campaigning for reproductive rights focused on securing a limited legal right to end pregnancy, for Black and Indigenous women and women of color who have been subjected to historical and contemporary sterilization abuses and white supremacist violence, “it is important to fight equally for (1) the right to have a child; (2) the right not to have a child; and (3) the right to parent the children we have, as well as to control our birthing options[… (4) and the] necessary enabling conditions to realize these rights” (ibid). Reproductive justice, then, applies an intersectional approach to fighting for reproductive care and arguing that abortion, access to prenatal support, the treatment of women in prisons, environmental harms, the murder of Black children by the police, the separation of immigrant families, and the access of queer and trans people to reproductive healthcare are not isolated issues but exist on a continuum of inextricably linked reproductive concerns. Engaging a reproductive justice lens to consider the artificial womb compels us to look beyond a focus on whether the technology could act as a “new reproductive alternative” (Kendal 2015, p. 8). It requires us to ask who might be excluded from the ‘choices’ the technology is purported to increase, to consider whether the artificial womb could undermine the right to have a child or to control reproductive options, and to assess whether it could result in harm to those who have been historically and continually marginalized in reproductive care. Both the scientists engaged in contemporary research toward artificial womb technologies and the humanities scholarship discussed in the previous section frame this technology in relation to its intended consequences: an intervention to benefit neonates and pregnant people. A reproductive justice-informed analysis can help us understand both existing inequity in care for pregnant people and neonates that problematizes approaches to this technology as a universal good, and the unintended consequences of this technology that might lead to reproductive harms. Elsewhere, with Romanis (Horn and Romanis 2020), I have argued that speculation about the future impact and uses of artificial womb technologies must be grounded in the realities of existing reproductive stratification. Here, I want to put this into practice by considering specific examples (racialized disparities in health outcomes, and racialized and classed instances of reproductive coercion), that suggest that on its current path, the artificial womb is likely to only increase existing inequity. I have chosen to draw on illustrative examples from the three common law jurisdictions in which my research is focused (namely, Canada, the United States, and the United Kingdom). These examples are in no way intended to be comprehensive: on the contrary, they represent only a few of the forms that inequality in reproductive health and rights takes within each of these jurisdictions and globally. They are selected as prescient cases to indicate the breadth of this problem. Protecting choice remains a vital component of reproductive freedom. But it must be pursued alongside creating the conditions that would allow all pregnant people to meaningfully exercise autonomy in practice. Analysis of the artificial womb that is informed by reproductive justice and the history of the unequal impact of reproductive technologies can help us explore in concrete ways not only the relationship between artificial wombs and choice, but also whether ectogenesis could undermine pregnant peoples’ rights to carry a wanted pregnancy.

Artificial wombs, structural racism, and perinatal care

Racialized inequity in care across high-income nations

Once trialed and demonstrably safe for both the pregnant person and the extremely prematurely born baby,Footnote 1 partial ectogenesis could have the greatest impact on improving maternal and neonatal health overall if made available to the women most likely to experience preterm birth or complications in the latter stages of a wanted pregnancy. The United States, where research toward an artificial womb prototype is ongoing, has the highest rate of preventable death during pregnancy and birth of any wealthy nation, and death rates for preterm infants also remain high. As Novoa and Taylor (2018) and Matoba and Collins (2017) have thoroughly tracked, rates of maternal mortality and morbidity in the US remain high among Black and Indigenous women, and highest for Black women. Novoa and Taylor write that as of 2018, “African American mothers [were] dying at three to four times the rate of non-Hispanic white mothers, and infants born to African American mothers are dying at twice the rate as infants born to non-Hispanic white mothers” (2018, np). A long-awaited CDC report released in 2020 indicated that nationally, rates of maternal mortality and morbidity are rising. But as Martin and Montagne write, even in states where the rate of mortality and morbidity is decreasing overall, it has not decreased for Black mothers, meaning that the gap between health outcomes for Black pregnant people and white is growing, not shrinking. As of 2016, “In New York City, [...] [B]lack mothers are 12 times more likely to die than white mothers,” a “widening gap [that] reflects a dramatic improvement for white women but not for Black [women]” (Martin and Montagne 2017).

As Adinolfi and others argue, artificial wombs could have significant clinical use in reducing maternal and infant mortality and morbidity. But they would only improve these outcomes overall if they improved outcomes for the women and infants most at risk of death or health complications during and after premature birth. And this would only occur if firstly, the technology was readily available, and, secondly, the primary cause of high rates of maternal and infant morbidity and mortality for these groups was lack of access to advanced technologies. As to the possibility that once proven safe artificial wombs will be readily available, under current circumstances, the costs of staffing and equipping Neonatal Intensive Care Units (NICUs), are extremely high (Cohen and Sayeed 2011). In the United States, then, the technology is likely to first be available only in specialist care units and at high cost to the patient. Barring a change to the expense of healthcare in the US, a disparity in access to the artificial womb along classed lines is inevitable. A contributing factor in racialized disparities in health outcomes for pregnant people and neonates is “significant underinvestment in family support and health care programs” (Taylor et al. 2019), which disproportionately impacts Black families. In New York, the most underfunded hospitals, which rank worst for care given to pregnant and birthing people, are the most likely to primarily serve Black patients.

But rates of morbidity and mortality among pregnant and birthing Black women remain disproportionate even when studies control for socioeconomic status, education, and access to resources (Matoba and Collins 2017; Novoa and Taylor 2018). When each of these factors are considered, it remains the case that “Black women are 49 percent more likely than white to deliver prematurely, and closely related, black infants are twice as likely as white babies to die before their first birthday” (Martin and Montagne 2017). A key cause of this stark reality is the way in which experiencing racism throughout one’s life and during pregnancy and birth can exacerbate stress. Before, during, and after birth, Black women are more likely to be subjected to discrimination in the hospital setting, and to feel “devalued and disrespected by medical providers” (ibid). The culture of antenatal care, Martin and Montagne argue, has been built around white middle class women, with those who exist outside of this category confronting bias and microaggressions at each turn. As Julian and Robles et al. write, the “persistent Black-white racial disparities in preterm birth among highly educated, privately insured birthing people” (2020, p. 7) demonstrate that these disparities are not reducible to inequitable access to advanced neonatal care and technology or perinatal resources. The authors argue instead that these continued disparities are a problem of “prioritizing individual behavioural interventions [...] rather than structural and social determinants of health” (5).

It is important to acknowledge the particular social and political realities of the United States. As I have argued elsewhere, to build critical, justice-oriented feminist responses to this technology, the unique limitations of each jurisdiction into which artificial wombs could be introduced must be considered (Horn 2020). But it is also useful to look at the ways that inequity emerges in similar patterns across nations that may be distinct in other ways. While the United States is an outlier among wealthy countries in its continued high rates of maternal and infant mortality and morbidity, disparities in these outcomes are also reflected in wealthy nations with publicly funded healthcare systems and better comprehensive maternal and neonatal health outcomes. To understand the persistence of this problem across different contexts, it is illustrative to also consider prescient examples from the United Kingdom and Canada.

In contrast to the United States, access to perinatal care under the National Health Service in the UK is broadly more consistent. But despite very low rates of maternal morbidity and mortality and lower rates of inequity in access to neonatal care, racialized disparities remain. Notably, “compared to white women, Black women are 5 times more likely to die during pregnancy, mixed race women are three times more likely to die, and Asian women are 2 times more likely to die” (Public Health England 2020, p. 17). These inequities are also reflected in rates of preterm births, stillbirths, and neonatal deaths. Further, studies that consider circumstances where mothers come close to death reflect rates “for African and Afro-Caribbean women [that are] double those for White women” (Smalls et al. 2018). As in the United States, these disparities persist even after adjustments for socioeconomic status and health risk factors. Research as to the causes of these discrepancies, much as in the United States, cite “disrespect from healthcare providers” (ibid) as well as ineffectual and insensitive communication from providers as key determinants.

As in the United Kingdom, the public healthcare system in Canada means more positive health outcomes for pregnant and birthing people and neonates in general. However, maternal morbidity and mortality in Canada are higher among Black women, recent immigrant populations, and Indigenous women than for white women (Kolahdooz et al. 2016). In 2016, researchers convened a study based on the hypothesis that universal healthcare in Canada would mean a less substantive racialized disparity in rates of preterm births than in the United States (McKinnon et al. 2016). The study found that despite much lower rates of preterm birth in Canada, inequity between Black and white birthing people and neonates was comparable to the United States. In addition, Kolahdooz writes, “Indigenous women in Canada have a two times higher risk of maternal mortality in comparison to the general Canadian population” (2016, p. 335) and also “experience higher rates of adverse outcomes including stillbirth and perinatal death, and, in some cases, low-birth-weight infants, prematurity and infant death” (ibid). These circumstances, along with “hemorrhage, infections, high blood pressure [and] ectopic pregnancy” (334) are precisely the kinds of situations in which bioethicists and scientific researchers have suggested neonates and pregnant people might most benefit from the use of a partial or full artificial womb (Partridge et al. 2017; Smajdor 2007; Gefland 2006; Coleman 2004).

Artificial wombs cannot redress inequity in maternal and neonatal care

Firestone (1970), often referenced for her feminist analysis of artificial wombs, did not consider racism in the United States when she wrote about ectogenesis. But her broader assertion that the technology could only be emancipatory after a revolution is relevant here. Once demonstrably safe, an argument could be made that artificial womb technologies should be distributed within the nations and communities in which rates of maternal mortality and morbidity, as well as preterm birth, are highest. But so long as the causes of preventable racialized disparities in health outcomes remain unaddressed, these disparities will remain. As Julian and Robles et al. write, in the United States, “coercive practices of sterilization, contraceptive counselling and provision, and drafted legislation requiring contraception for access to social services are just a few examples of injustices violating Black birthing people’s bodily autonomy” (2020, p. 8). These injustices follow a lengthy historical record of white supremacist reproductive violence against Black women that has yet to be sufficiently confronted. The persistence of racialized disparities in health outcomes for neonates and pregnant people despite the availability of publicly funded healthcare in Canada and the UK is a testament to the fact that these disparities are not reducible to a lack of access to clinical expertise or advanced technologies. In the UK, there continues to be no official target set for closing the gap in health outcomes. As Dr. Christine Ekechi notes, within the UK medical education system, there is a significant lack of recognition of and teaching on systemic racism in Britain. Redressing inequity in maternal health, she argues, first requires acknowledgment that racism drives health outcomes (Ekechi 2020). In Canada, race-based health data are not collected, meaning that the true extent of inequity in outcomes for birthing people and babies is not known. Researchers note that in some instances, there is a need to improve geographical access to antenatal care (Kohlahdooz et al. 2016). But obstetric racism experienced by Black and Indigenous women, as opposed to a lack of access to resources, is cited as a primary cause of racialized inequity in outcomes for pregnant people and their babies. The idea that ectogenesis “would not only free women from pregnancy” (Takala 2009, p. 191) and provide an additional choice in care but also potentially “eventually, lead to true equality” (ibid), then, which arises in much of the literature, is challenged by existing stratifications in reproductive care and is contextually specific. What these inequities, produced and maintained by structural and institutional failings, tell us is that the pregnant subject who is imagined as a beneficiary of the artificial womb in accounts of the technology as a means of improving choice is neither neutral nor universal. Examining how health outcomes for pregnant people and neonates have been shaped by microaggressions and discrimination, maternal stress linked to structural racism, and unacceptable practices of care draws to the fore the way discourses of artificial wombs as a universal good centralize a subject who already has access to sufficient care and the conditions to exercise autonomy over their birth circumstances. As Taylor et al. write, “the maternal and infant mortality crisis cannot be adequately addressed without first understanding and then dismantling racism and bias in the healthcare system” (2019). Making an artificial womb available as an elective choice would not in itself address the “structural and social determinants” of health inequity. If the artificial womb was not accompanied or preceded by significant structural and social changes, we would see a “counterpoint to claims that artificial gestation would benefit women as a class” (Jackson 2008, p. 410), as it may be likely that there may be improvements to maternal morbidity and mortality and care for premature babies overall, but no requisite change in these racialized inequities. Contemporary racialized disparities in preterm births and maternal morbidity and mortality lay bare the importance of thinking about the artificial womb as a technology that can only be as innovative as the social context into which it arrives. Hopes for the artificial womb as a means of radically decreasing maternal and neonatal morbidity are presented with the intention of taking these forms of suffering seriously. But analyzing the technology within the context of existing inequity rather than in the context of choice shows a fuller picture of the contemporary barriers to moving a collective project of reproductive freedom forward. None of this is to say that the pursuit of artificial wombs as a means of redressing the very real risks to pregnant people and neonates that can emerge in pregnancy and birth is an unworthy pursuit, or that emerging projects that consider the artificial womb as a way to improve reproductive health choices are faulty. But emphasizing that this technology is likely to enter a context of substantive inequity allows us to reframe questions about how it should be designed, adapted, introduced, and regulated. As I will consider in the final part of this paper, understanding the artificial womb within existing social and structural barriers to equity of care allows us to consider how we might chart an alternative path. In the next part of this paper, I consider another facet of a justice-oriented approach to assessing the impact of the artificial womb, namely, where the technology may be situated in relation to historical and contemporary instances of classed and racialized reproductive coercion.

Artificial wombs and reproductive coercion

Concerns raised by feminist scholars that artificial wombs might be forced on women considered to be ‘unfit’ mothers are often cited in the broader bioethical literature and subsequently dismissed as paranoia (Singer and Wells 1983; Coleman 2004). Yet these concerns find justification in the context of a long lineage of bioethicists who have extolled the possibility that artificial wombs might be safer for fetuses than human pregnancy (Singer and Wells 1983; Reiber 2010; Pence 2006; Coleman 2004; Adinolfi 2004; Welin 2004; Kaczor 2005). Kaczor writes that “partial ectogenesis may someday become less risky than normal gestation, since an artificial womb would not, presumably, get into car crashes, slip and fall, or be assaulted” (2005, p. 298). Similarly, Pence suggests that ectogenesis could be used to “change the gestational course for a poor baby of an alcoholic mother from being born addicted to alcohol and retarded to being born alcohol free and with superior nutrition and oxygenation” (2006, p. 82). Gefland, too, argues that “ectogenesis would in all likelihood protect the fetus from second-hand smoke, alcohol, and an unhealthy diet” (102). Hammond-Browning, while asserting that artificial wombs should not be mandated, argues that if a pregnant person is a regular user of drugs or alcohol, “fetal transfer to an artificial uterus” (2018) may be in the best interests of both parties.

While feminist scholars have been attentive to the way such proposals might be engaged to pressure women into using an artificial womb (see Sander-Staudt 2006), excluding a brief note in Jackson (2008) and Abecassis (2016) there has been little engagement with precedent in existing law and policy that might allow or prevent such intervention from occurring. In what follows, I consider the extent to which existing laws or policies might prevent or enable coercive use of artificial womb technology. The racialized disparities in health outcomes for pregnant people and neonates I discussed in the previous section of this paper constitute one example of the way attending to justice can highlight the contemporary inequity that draws discourses of the technology as a universal good into question. In this section, I am interested in thinking about instances of past and ongoing reproductive coercion to illustrate that an approach informed by reproductive justice rather than choice calls our attention to the harms that might accompany the pursuit of an intended project of this technology to improve decision-making in reproductive care. Considering artificial wombs with reference to their potential impact on “the right to have a child” and to “control one’s birthing options” (Ross et al. 2017) helps us to think through how calls for elective access to this technology might have the unintended consequence of increasing opportunities for coercion. Feminist scholars (Kendal 2015; Romanis 2020) have explored the ways law and policy could inhibit the elective use of an artificial womb while arguing that such inhibitions would undermine choice. To be clear, I agree that pregnant people should be provided with options that allow them to experience pregnancy and birth in ways that are safe, informed, and comfortable for them. Romanis argues against paternalistic laws that limit pregnant peoples’ autonomy when she calls for changes that would allow them to “opt for ex utero gestation” (2020, p. 374). Dismantling paternalism in regulation governing reproduction is vital, and I too believe that the artificial womb might one day be a beneficial form of elective care. I do not wish to undermine the very legitimate claim that if artificial womb technology could improve care for pregnant people, laws that would prevent them from accessing this care would be groundless. The question here is one of how we frame discourse on choice and the future use of artificial wombs. A justice-oriented approach situates choice in the context of the structural conditions that allow or inhibit people from exercising those choices. Examining forms of reproductive coercion that have been permitted to occur demonstrates that if policies are pursued to facilitate elective access to this technology consideration must also be given to monitoring against coercive uses. As I will discuss when I turn to how we might envision an alternative path for artificial wombs, these analyses are also a means of calling the discourse on this technology toward an orientation that centralizes racial justice and reproductive freedom in speculation on how the technology might be developed, designed, and introduced. Where the first part of this paper looks toward expanding a focus on choice to consider the harms that might occur where existing inequity is not included in framing discussions about the technology’s introduction and use, this part of the paper looks at the harms that might emerge from a pursuit of access without mitigating to prevent coercion. A reproductive justice-informed approach here assesses the existing structures (laws, policies, and practices) that might currently make the pursuit of elective use of artificial wombs incompatible with protecting reproductive freedom.

Reproductive coercion authorized by law

The ideas expressed by Pence (2006), Gefland (2006), and others that artificial wombs might benefit or protect fetuses by providing a “safer” gestational environment than a pregnant person’s uterus in instances where the pregnant person has consumed alcohol may read to some as an innocuous thought experiment. But in the contemporary United States, Paltrow and Flavin have tracked “more than four hundred” (2013, p. 300) instances of pregnant women being detained on charges of endangering their fetuses “in forty-four states, the District of Columbia, and federal jurisdictions from 1973 to 2005” (300). In most of these cases, pregnant people were arrested and detained on the grounds that they had exposed their fetuses to harmful substances. As Paltrow and Flavin have traced, numerous states have statutes under which courts have been allowed to take such actions.

A Wisconsin statute, for instance, allows the state to detain people believed to be pregnant who “demonstrate ‘habitual lack of self-control’ in the use of alcoholic beverages or controlled substances” (Wis. Stat. Ann. § 48.193 cited in Paltrow and Flavin 2013). Similar statutes are in place in Minnesota and South Dakota, and while not every state has legal regulations that explicitly apply to actions in pregnancy, Paltrow and Flavin also identify the use of statutes intended to protect older children to criminalize pregnant people for harming their fetuses. While drugs or alcohol are cited as the primary justification for detention in these cases, other activities were also referenced as concerns for medical practitioners and police, such as the presumption that the pregnant person had not actively sought prenatal care or “gave birth at home or in another setting outside a hospital” (ibid). As Ross and Solinger (2017) aptly demonstrate, criminal cases like these actively target poverty: a lack of seeking prenatal care, for example, in the privatized healthcare system of the United States, speaks to a lack of resources to enable someone to seek such care. Were a wealthy person to drink or use drugs and give birth at home, it is unlikely that the police would become involved. Cases involving drug use are disproportionately targeted at low-income pregnant people, and widely disparate according to race:

Overwhelmingly, and regardless of race, women in our study were economically disadvantaged, indicated by the fact that 71 percent qualified for indigent defense. Of the 368 women for whom information on race was available, 59 percent were women of color, including African Americans, Hispanic American/Latinas, Native Americans, and Asian/Pacific Islanders; 52 percent were African American (Paltrow and Flavin 2013, p. 311)

While scholars such as Pence and Gefland speculate about the possible uses of the technology to preserve the health of the fetus from a mother who may expose it to “harm,” they do so without reference to the ways in which existing statutes such as those traced by Paltrow and Flavin already allow the state to intervene in such cases. Returning to the possibility that broad availability of the artificial womb as an elective ‘choice’ without substantive social and structural changes could result in coercive use of the technology, the application of statutes to arrest and detain a pregnant person is of course not interchangeable with enforced extraction of a fetus to an artificial womb for the duration of the fetus’s gestation. But the legally permitted detention of pregnant women for actions against their fetuses in the United States does present a precedent for possible coercive use of artificial wombs.

A particularly prescient example here is the case of Martina Greywood, an Indigenous woman who was arrested and detained in North Dakota on charges of exposing her pre-viable fetus to toxins. While awaiting trial, Greywind sought an abortion. On learning of Greywind’s abortion, the state dropped the case, finding that it was no longer relevant.Footnote 2 To return to the question of ectogenesis being used coercively, if a woman such as Greywind were arrested for exposing a fetus to harm, she could be informed that the charges may be dropped should she allow the fetus to be extracted to an artificial womb. In this instance, the introduction of artificial wombs could raise a very real probability in the US not of fetuses being forcibly extracted from women’s bodies, but as in the Greywind case, of a pregnancy being ended by ‘choice’ by a person who would face substantive criminal charges if they did not comply.

Equally, as Stone notes, few drug rehab centers are willing to treat pregnant people (2015), and, as Paltrow and Flavin have tracked, many such centers are the sites of pregnant people being referred to the police. In contemporary situations in which a drug or alcohol-using pregnant person has sought rehabilitation, she risks being criminalized and detained. In future, the use of an artificial womb may be offered as a means of receiving treatment and avoiding criminalization. Where a pregnant person may need to consent to such an exchange, given that the choices would be between criminalization without treatment and fetal extraction, the situation would remain coercive. Based on existing precedent, under contemporary circumstances, this kind of coercion would not be circumscribed by US law, as it would involve no forced bodily procedure, and would uphold state statutes targeting drug use in pregnancy. In its initial stages, the artificial womb is likely to be costly and not widely available. Feminist scholars who have made a case for the elective or therapeutic use of the technology (Kendal 2015; Smajdor 2007) argue for its wide availability. To ensure that pursuing the availability of the technology to improve choice does not subsequently also lead to its availability for coercive use, then, it is necessary to take consideration of existing policies that could potentially allow or inhibit such practices. In the contemporary context of the United States, artificial wombs could be used in coercive ways with the approval of the law. If we consider the artificial womb with reference to the reproductive justice purview of protecting the right to carry and birth one’s own child on one’s own terms, these are clear limitations that need to be addressed ahead of the technology’s introduction, and before pursuing elective use of the artificial womb should be considered tenable.

Reproductive coercion through social sanctions

Among wealthy Western jurisdictions, the United States is notable for its restrictive and paternalistic practices around reproduction. But, as examples from the United Kingdom and Canada demonstrate, social policies and practices that inhibit access to resources can also act as a form of reproductive coercion. In the UK, there are no statutes that legally sanction the surveillance and punishment of pregnant people for consuming substances that may harm their fetuses. Strong precedent has been established against criminalizing pregnant people for actions against fetuses. In 2014, CP v. First-Tier Tribunal and Criminal Injuries Compensation Authority was heard by the Court of Appeals. This case involved a claim made on behalf of a six-year-old girl who had fetal alcohol spectrum disorder against her mother for injuring her in utero. Beginning with the Criminal Injuries Compensation Authority, “which refused the claim on the grounds that she had not been a victim of violence”Footnote 3 the case continued to the Court of Appeal, where the court affirmed CICA’s stance. The Court of Appeal established that the case was not analogous, as the claimants argued, with Attorney Generals Reference (No.3 of 1994). In this case, a man stabbed his pregnant girlfriend, resulting in significant injuries to the woman and the premature birth of her baby, who died shortly after birth. In Attorney General’s Reference, harm was caused to a person (the pregnant woman) resulting in the death of a person (the baby born prematurely, who later died due to injuries caused in the stabbing). The court in C.P. v. CICA clearly establishes this as distinct from the circumstances of harm caused by drinking in pregnancy, noting that “an essential ingredient of the offence […] is the infliction of grievous bodily harm in a person. Grievous bodily harm to a foetus will not suffice” (19). The claimant’s argument that an offence had been committed under Sect. 58 of the Offences Against the Person Act 1861, referring to the prohibition against a woman administering poison to herself with the intent of causing miscarriage, was also considered. The court found that this did not apply because the defendant had not consumed alcohol with the intention of having a miscarriage. The court took a firm position that “in English law women do not owe a duty of care in tort to their unborn children” (19). As a test case that was an attempt to establish a crime of consuming toxins while pregnant, C.P. v. CICA instead had the effect of establishing a clear precedent against treating such actions as a crime. The possibility that because “artificial wombs [...] wouldn’t be threatened by irresponsible introductions of alcohol or illegal drugs” (Rosen 2003, p. 72), some women might be forced into use, is likely to be strongly censured in England.

This does not mean, however, that coercive use of ectogenesis would not occur. Alghrani (2008) briefly considers the danger of drug or alcohol-using pregnant women being forced to utilize artificial wombs but asserts that barring a significant change to the legal status of a fetus, this would be prohibited under English law. While I agree it is improbable that a pregnant person would be legally ordered to have her fetus extracted into an artificial womb in the UK context, it is necessary to think about how coercion can operate without requiring the force of law. In her short consideration of whether women could be pressured to use artificial wombs, Jackson (2008) makes the point that if the technology were to come to be seen as safer for the fetus than human gestation, people with “a less than ideal lifestyle” (361) may find themselves under social pressure to use ectogenesis. Jackson does not consider specific examples, and in what follows, I explore two sites through which this kind of social pressure might operate.

First, as of 2017, the two-child benefit limit on universal credit in the UK means that recipients cannot receive additional support for a third or additional child born after April 2017 unless they were “born as part of a multiple birth [or] born as a result of a non-consensual conception (including rape) or conceived when [the pregnant person was] in a controlling or coercive relationship” (UK Government Universal Credit 2017). In the second and third instances, a woman must also be able to prove she is not living at an address with the abuser at the time of making the welfare claim, a stipulation that significantly overlooks the realities of abusive relationships. As the British Pregnancy Advisory Service has stated in criticism of the credit limit, in many situations of abuse, women remain with abusive partners for reasons including financial dependency and a fear of loss of access to their children. The exceptions to the two-child limit are in theory meant to acknowledge situations in which the birth of a child is beyond a woman’s control. But they do not account for the breadth of reasons why a person may become pregnant with an additional child, or why they may continue that pregnancy.

As the British Pregnancy Advisory Service has argued, the welfare cap “fails to meet its stated aim of ensuring that individuals in receipt of welfare supports face the same choices about having kids as those in active employment do [and] traps families in poverty.” The policy has significant implications for reproductive rights. It explicitly invites women to consider abortion to provide for their existing children, using the language of encouraging people “to think carefully about whether they can afford to support additional children” (British Pregnancy Advisory Service 2018). The universal credit limit is distinct from American statutes in which pregnant people might feel coerced into using an artificial womb to avoid criminalization. The consequences of continuing a third pregnancy do not extend to detention or incarceration. But feminist scholars have proposed that the technology could be used electively by people who did not wish to continue a pregnancy (Smajdor 2007; Kendal 2015), or by those whose health was endangered by conditions arising during gestation (Romanis 2020). If policies were in place to allow this elective use, people reliant on universal credit might be presented with artificial wombs as a ‘choice.’ Electing ectogenesis would allow them to continue to receive support for their existing children while also not terminating a fetus. In such instances, social inequity would be a factor in generating pressure to use an artificial womb.

The second example I want to consider is the non-governmental but judicially supported PAUSE project. PAUSE is a charity that works with women who have had more than one child removed from their care. The project “offers women an 18-month, individually tailored, intensive package of support,” which they can access if they “agree to use an effective form of reversible contraceptive for the 18-month duration of the programme” (McCracken et al. 2017). PAUSE has already been purchased by “34 local authorities,” with purported plans to expand to coverage to women considered to be “at risk,” meaning those who are judged likely to have children who might be taken in to care. PAUSE again works on a model of voluntary compliance: an agreement to use long-acting birth control in exchange for support and rehabilitation. But given the lack of alternative broad-reaching support programs for women like those enrolled in Pause, who face challenges including domestic violence, homelessness, and drug and alcohol use, the voluntary choice to use long-acting birth control occurs under circumstances where the opportunity for choice is limited. There is a striking parallel here between the United States context, in which the provision of welfare has long been associated with punitive measures and coercive pressure for women (particularly Black women and women of color) to use long-acting contraceptives (Ross and Solinger 2017).

Although Hammond-Browning does not reference Pause when she proposes that it might be in the best interests of both pregnant people struggling with substance use and their fetuses to choose transfer to an artificial womb, her suggestion is informed by the same principles as the program. Women who chose fetal transfer, she notes, would benefit in that they “would have time to seek treatment for their substance abuse problems, thereby creating a more secure environment for their future children after birth from the artificial uterus” (2018, p. 358). It could indeed be the case that individual women welcomed the opportunity to use the technology in this way, just as it is the case that some women involved in the Pause project welcome the resources it offers (McCracken et al. 2017). But if people struggling with substance misuse are only granted conditional access to care in exchange for agreeing to take a specific type of birth control, or, in future, to use an artificial womb, the necessary conditions for reproductive autonomy are not being met. Resituating Hammond-Browning’s proposal by placing choice within the context of historical and contemporary reproductive coercion highlights the way that applying a pregnant person’s alcohol or drug use as a rationale for artificial wombs risks creating the conditions for coercion. For choice to be meaningfully practiced, resources must be provided that allow informed decisions to be enacted in the absence of external pressure. As Roberts writes, we cannot solely be concerned with choice, we “should also be concerned about the quality of options available […] it is possible that all the alternatives decrease [a person’s] control over her reproductive health” (1997, p. 136). In the context of the UK, practices such as the choice between a prison sentence or the use of an artificial womb would likely be roundly condemned, but both the two-child benefit limit and the PAUSE project should be taken as examples of sites wherein the option of ectogenesis could result in reproductive coercion.

Like the UK, Canada has no statutes in place that criminalize women for the consumption of drugs or alcohol in pregnancy. In 1997, the Canadian Supreme Court made a decisive interpretive ruling in Winnipeg Child and Family Services (Northwest Area) v. G (D.F.). Winnipeg Child and Family Services sought permission from the Manitoba Court of Queen’s Bench to hold a woman (G) who was twenty-two weeks pregnant in treatment for glue-sniffing until she had given birth. The order was granted but reversed on appeal. When the case was ultimately appealed to the Canadian Supreme Court, the reversal was upheld. The Supreme Court affirmed that a fetus is not a legal person in Canada, and that to detain the woman would violate all women’s rights under Morgentaler’s protections to bodily autonomy and liberty.Footnote 4 At the highest level of the Canadian justice system, there is strong precedent against criminalizing women for actions against their fetuses. There is also broad precedent toward affirming that a fetus is not a person until birth. Given this, it is likely that Canadian law would prohibit people being coerced to use the technology to avoid criminalization.

But as in England, existing practices that are not mandated by law are also relevant to the question of coercive use of artificial wombs. I want to consider one particularly pressing example. In the final paper of the National Inquiry Into Missing and Murdered Indigenous Women and Girls, the non-consensual sterilization of Indigenous women is noted as a form of genocide, a project beginning in the 1920s by which “sterilization was viewed as a way to eventually eliminate the Indigenous population entirely” (Reclaiming Power and Place 2019, p. 266). As the authors of the report note, while statutory laws permitting forced sterilization in Alberta and British Columbia were repealed in the early 1970s, these abuses have continued to occur throughout the country. Class action lawsuits on behalf of Indigenous women who have experienced tubal ligation (a sterilization procedure) without consent, often while in hospital recovering from the birth of children, are underway across Canada (Kirkup 2018; Kusmer 2018; Boyer and Bartlett 2017). Three of these suits have been filed against the Governments of Alberta, Saskatchewan, and Manitoba, with the experiences reported by claimants spanning the 1980s through to 2018.

To turn back to the question of potentially coercive uses of artificial womb technology, these cases are not indicative of the possibility of a situation in which a woman would face coercion in the form of a ‘choice’ between criminalization and ectogenesis, as I suggested could be possible in the US context. Nor are they suggestive of the ‘choice’ between losing vital resources and ectogenesis that I have indicated in England. However, the cases addressed in these lawsuits involve practices condoned within Canadian hospitals whereby women were given false information about tubal ligation procedures immediately after giving birth. Many of these cases involve situations in which “women were denied access to their newborn babies unless they agreed to the procedure” (Kirkup 2018). A situation in which a woman might be coerced into using an artificial womb is importantly distinct here, in that this would involve coercive action at the beginning of a pregnancy, whereas these cases refer to actions after birth. However, the significant history of coercive and forced sterilization of Indigenous women in Canada, and the absence of specific legislation to ban these procedures, suggests that practices of offering ectogenesis without provision of actual informed consent or under coercive circumstances could occur in Canadian hospitals. If, as in many of the instances of contemporary sterilization abuse, hospitals could demonstrate that women had given consent (despite an absence of evidence that consent was freely given), this would be a practice that may not be circumscribed by law. As I’ve illustrated earlier in this paper, while artificial womb technologies have significant promise as clinical tools to improve care for neonates and pregnant people, the soonest we are likely to see a prototype for human trials is within five years. On one hand, if the cost of artificial wombs is prohibitive and they can only be used in highly equipped neonatal intensive care units, this may worsen reproductive stratification. But on the other, if the technology is widely available, this may increase the possibility that it could be used in coercive ways. The purpose of attending to these possibilities is not to suggest that either is inevitable. Considering this technology’s impact on reproductive justice more broadly is a means of reframing how this technology is assessed, to attend to contemporary policies and practices that shape the extent to which artificial wombs may improve outcomes for pregnant people and neonates or reinforce existing reproductive harms.

Centralizing justice, not choice

One of the most important contributions of the feminist literature on artificial wombs thus far, from Firestone (1970) to Cavaliere (2020) and Romanis (2020) is to emphasize that easing the impact of pregnancy on women is a worthy research goal. Firestone noted of dismissive responses to both the plausibility and importance of artificial womb research that if we could put a man on the moon, we could create an alternative to gestation. As Smajdor writes, “the fact that pregnancy and childbirth [...] are painful, and not merely trivially so” (2007, p. 91), should be rationale enough to justify research toward artificial wombs. We can, and should, take this pain seriously, and acknowledge advancements that may ease the symptoms, side effects, and complications of pregnancy as important and valuable contributions. So too should we acknowledge that work toward improving outcomes for neonates is work that may redress the suffering of parents and families who have experienced the painful loss of a prematurely born baby. But what I have sought to show in this paper is that how social scientists frame these concerns, and how we pursue redressing them, is of vital significance to the impact of this technology. Thinking informed by a reproductive justice framework demands that we acknowledge, process, and engage with both historical and contemporary instances of reproductive inequity, violence, and coercion, to confront the structures that currently make an artificial womb that benefits all pregnant people impossible.

Cavaliere (2020) emphasizes that the idea that artificial wombs will benefit “women” as a group overlooks the fact that each new reproductive technology that has promised to improve ‘choice’ for women has largely fulfilled this promise only for those already privileged with significant resources (as Cavaliere argues, primarily white middle class women living in the Global North). Though the subject who is understood to reap improved care and expanded treatment options from artificial wombs in much of the broader literature is presented as a neutral ‘every woman,’ in the technology’s current form, it is the privileged few Cavaliere describes who stand to benefit. It could be argued that the fact that artificial wombs may only be available to a small subset of the global population, that they may worsen inequity, and that they may be used to justify coercion does not discount the way they could improve reproductive options for some. One could argue that inequity in outcomes for pregnant people and neonates does not outweigh the health benefits of artificial wombs for those few who might access them. One could also argue that potentially coercive uses of this technology could be effectively prevented through regulating against these outcomes. These potential unintended consequences, we could conclude, do little more than show us that this technology is not a universal good. I agree that it is not enough to simply observe that a worsening of stratification and the risk of coercion could emerge in the wake of this technology. Regulatory frameworks can be engaged to facilitate intended consequences (use as healthcare) and limit unintended ones (coercion). And social scientists who have focused on choice-based frameworks and argued that this technology could have significant benefits for pregnant people make the important contribution of emphasizing the importance of an innovation that could ease the very real and often unacknowledged dangers of gestating a baby. But I believe that we can go further than pushing to regulate for ‘positive’ uses of artificial wombs (improving care in reproduction) and against negative ones (coercive uses and/or increasing stratification). The bigger question here is for what purpose, and for whom, reproductive technologies should be created. By recentering the discourse on artificial wombs around the pursuit of justice, not choice, it becomes possible to reconsider what artificial wombs might mean and do.

In Women, Race, and Class, Angela Davis observed that “the failure of the abortion rights campaign to conduct a historical self-evaluation” (1982, p. 354) and understand how Indigenous women and Black women in particular had been subjected to sterilization and reproductive violence, led to the truncation of holistic access to reproductive care. In centering the interests of white, heterosexual, middle class, and abled women, a focus on securing the ‘choice’ to continue or end a pregnancy superseded broader claims to reproductive freedom. What would it mean to resituate discourse on artificial wombs, not in the lineage of contestations over improving choice, but in the lineage of reproductive freedom? How might the future of artificial wombs be animated and informed by the rich body of literature on reproductive justice, and racial justice and technology? A focus on justice, not choice, that centers those who have been marginalized in reproductive care at each stage of the artificial womb’s development, implementation, and design, might open questions and concerns that could lead to an alternative path for this technology. While scientific projects are currently in development with the intention of creating a platform for external gestation to facilitate neonatal care, I use ‘artificial womb’ broadly here to include projects that have yet to be pursued. Considering a technology that is in development but that does not yet exist invites us to ask, ‘how could this technology exist?’ Drawing questions of justice to the fore directs us to reorient how we consider each aspect of the technology creation and use. What materials, for example, could be applied to assemble the artificial womb: where will they be sourced, what will they cost, who will own them, and what infrastructure will be required for their safe mobilization? And who might be considered a stakeholder in the creation of this technology: midwives, doulas, pregnant people themselves? In other words, what happens if instead of asking questions of choice, we ask questions of ownership, development, and use, questions of who this technology will be built for, in what environments (and by whom) it will be intended for use, who will own it, and what forms of care it might support.

Dána-Ain Davis writes of racialized maternal and neonatal health inequity that “it is through the care practices of radical Black birth workers and expanding care and birth options that we find a potent response to medical racism and effective work to attain outcomes that are more in alignment with the goals of reproductive justice” (2019, p. 15). How might an artificial womb be different if the “care practices of radical Black birth workers” were centralized to explore outcomes for the technology “in alignment with the goals of reproductive justice”? What are some of the ways in which embedding artificial womb technology within existing institutions and regulatory frameworks (both commercial and medical) might limit its use or facilitate the (un)intended consequences examined in this paper? An artificial womb designed for the use of Black birth workers, of midwives, doulas, or pregnant people themselves, a technology owned and facilitated by reproductive justice-informed healthcare practitioners, might follow a different path than one engaged for limited use in a neonatal ward and patented by biotechnology companies. These questions, importantly, are not intended to suggest a normative framework, but to act as invitations for considering how reproductive justice might offer a crucial discourse for how we frame discussions about artificial wombs.

Of racialized inequity in perinatal and neonatal outcomes in Canada, Kohlahdooz et al. emphasize that the pursuit of change must include active engagement with the perspectives of Indigenous women, the situating of maternal services within communities, and cultural sensitivity to the importance of place for Indigenous pregnant people. Kolahdooz et al. particularly note the importance of “educating [health care providers] to understand Indigenous history and [to value] Indigenous maternal traditions and prenatal knowledge” (Kolahdooz et al. 2016, p. 344). If an artificial womb is to be developed, it is precisely the voices of those who have been most marginalized in reproductive care, such as Indigenous women in Canada, that should be at the center of exploring the questions of what this technology, could, should, or might do. Discourses of artificial wombs as a universal good, or a means of improving autonomy in reproduction, cannot account for the structural conditions into which this technology is introduced. Reproductive and racial justice literatures and practices, however, could bring different possibilities and problems to the fore and in so doing, might lead to a different conception of this technology. Julian and Robles et al. emphasize that applying a reproductive justice framework to health systems to ensure patient-centered access and redress racialized inequity in care includes ensuring “equitable access and utilization of culturally relevant options and opportunities for pregnancy, labor, birth, and postpartum” (2020, p. 4). Asking the question of whether or not, in whose hands, and in what form artificial wombs could ever be compatible with these principles is a means of both understanding contemporary research in its current context and of opening the possibility of alternative paths.