Uncertain Knowledge: The Medicalisation of Intersex People and the Production of Ignorance

Abstract

Ignorance is produced through mechanisms related to power relations and socio-cultural context. This article examines whether the theoretical conceptualisation of agnotology may be useful when exploring intersex and the way it has been erased socially and physically. Specifically, based on the work of a PhD in Sociology and History of Science, it proposes categorising three types of mechanisms of ignorance production—cultural, epistemological and physical—with the aim of providing a greater understanding of how medicine, science and technology participate in a continuous process of erasing intersex bodies and lived experiences. Using medical literature, interviews and observations, the article focuses on a specific area of biomedical knowledge and intervention: the prenatal ‘treatment’ of Congenital Adrenal Hyperplasia (CAH) with dexamethasone or ‘prenatal DEX’. It shows how this procedure was pioneered by French doctors and how it continues to be practised in France despite numerous uncertainties and controversies inside and outside of the medical sphere.

1. Introduction

‘Knowledge is possible only through the systematic “social construction of ignorance”’.

(Rayner 2012, p. 111)

Ignorance is not only the absence of knowledge but a gendered and racialised (Mills [1997] 2022) active process that has a ‘complex political and sexual geography’ (Proctor and Schiebinger 2008, p. 2). Several scholars have shown how some parts of the body, especially parts perceived as female (e.g., the clitoris) or gender non-conformity, have been the stage of an ‘epistemology of ignorance’ (Tuana 2004), subjected to mechanisms of invisibilisation. Knowledge is very entangled with power relations and is, therefore, dependent on what is considered to be, in different cultures and times, legitimate subjects of scientific production (Foucault [1976] 1988).

The pioneers of agnotology, Robert Proctor and Londa Schiebinger, showed how ignorance is produced through several “mechanisms, such as deliberate or inadvertent neglect, secrecy and suppression, document destruction, unquestioned tradition, and myriad forms of inherent (or avoidable) culturopolitical selectivity” (Proctor and Schiebinger 2008, p. vi). Following a sociological perspective of agnotology (Funkenstein and Steinsaltz 1987; Gross and McGoey 2022), this article focuses on how these mechanisms apply to intersex variations. It analyses ‘the modes of oversight and invisibility’1 (Fillion and Torny 2016, p. 49) of certain knowledge about different aspects concerning intersex people.2

Since the 1950s, most intersex people have been medicalised and undergone hormonal and/or surgical interventions during childhood and/or adolescence (Fausto-Sterling 2000; Karkazis 2008; Kessler 1998; Holmes 2008). Scholars in critical intersex studies (Holmes 2009) showed that intersex people, whose very existence unsettles longstanding convictions around sex binarism and dominant gender norms, have been subjected to forms of erasure (Holmes 2008; Kennedy 2020; Monro et al. 2024; Morrison et al. 2021). Despite the abundance of medical publications on the issue of intersex, this erasure can be understood as a mechanism of ignorance production. This mechanism mainly concerns how scientific knowledge and technologies concerning intersex are related to power and hierarchies between normative and non-normative bodies.

This paper examines the way the theoretical conceptualisation of agnotology may be useful when exploring the medicalisation of intersex people. It aims to explore the mechanisms of ignorance production—categorised into three types—since the development of Johns Hopkins’ biomedical protocol by looking at the French context and its relationship to global debates. In a nutshell, this protocol, developed by a multidisciplinary staff in the 1950s at the Johns Hopkins Hospital in the United States, called for the early detection and medicalisation of intersex children. Doctors and psychologists suggested deciding on the child’s gender assignment as soon as possible and medically intervening (ideally before the child was 3 years old) to make the body conform to the assigned gender, advising parents to keep the medical assignment a secret. I will argue that the epistemological considerations around ignorance production in science are entangled with practices of erasure, which attempt to make intersex lives impossible, echoing Viviane Namaste’s findings on the ‘invisible lives’ of transgender people in our societies, in discourse, and in everyday life institutions (Namaste 2000).

To illustrate how some of these mechanisms apply in practice, as well as their consequences in intersex people’s (and their parents’) lives, I will analyse a specific area of current biomedical knowledge and intervention in detail: prenatal ‘treatment’3 for Congenital Adrenal Hyperplasia (CAH).4 With treatment pioneered in France in 1980, CAH is the only group of genetic conditions and the only intersex variation for which prenatal hormonal ‘treatment’ is prescribed. My analysis will reveal that biomedical knowledge, technologies and practices concerning intersex people all belong to a specific ‘regime of truth’ (régime de verité, Foucault [1976] 1995), whose premises and consequences are unquestioned. Although professionals justify their protocols with science, the data produced resembles what I propose to call uncertain knowledge, even in the eyes of scientific experts themselves. My analysis will also show how ‘uncomfortable knowledge’ (Rayner 2012) is quickly sidelined in order to preserve the apparent consensus around controversial practices of ‘normalisation’ of ‘atypical’ sex, such as the prescription of prenatal hormones to avoid the ‘masculinisation’ of 46,XX chromosome foetuses with CAH (Dreger et al. 2012). By unravelling the mechanisms of ignorance production, a paradox emerges: increasing biomedical knowledge and techniques seems to produce new questions, dilemmas and debates.

2. Materials and Methods

This paper presents one aspect of a larger research project on medical practices on intersex people during the second half of the 20th century and the beginning of the 21st century, mainly in France and Israel. It is primarily a development of a chapter in my PhD dissertation, which is focused on the French context (Raz 2019, available in French in open access), where I used qualitative sociological and historical methods to analyse medical archives, interviews and observations. The historical materials included scientific and institutional literature between 1950 and 2022 and oral history via interviews. The medical literature sampling, via PubMed, consisted of researching articles published by French doctors in French and English using keywords (Congenital Adrenal Hyperplasia; Hermaphrodism; Hermaphroditism; Pseudo-hermaphroditism; Ambiguous Genitalia; Disorder of Sex Development; Micropenis; Hypospadias). Approximately 420 articles and 30 books/dissertations were identified, from which the most relevant were analysed in depth (about 80). Publications that were not focused on intersex aspects (i.e., only concentrated on biochemical aspects of CAH, genetics, tumours, etc.) were not included in the analysis presented here. Publications from nationally recognised French journals were given priority (for example, Pédiatrie, Annales pédiatriques, Annales d’endocrinologie, Archives françaises de pédiatrie, Chirurgie pédiatrique, Gazette médicale de France, Journal de génétique humaine, La revue du Praticien, L’Année endocrinologique, La Presse médicale, Revue française de gynécologie et d’obstétrique) especially for the 1950s to 1990s timeframe. In the section of the present article concerning prenatal ‘treatment’ for CAH, I also use international biomedical and bioethical publications. These were complemented by French institutional documents, such as official protocols and reports. Every (online) document coming from a French hospital or public health institution concerning CAH was collected.

Besides the literature review, I carried out semi-structured interviews with biomedical practitioners (19) and parents of intersex children with CAH (7) between 2012 and 2015. The interviews were recorded with the knowledge and consent of the interviewees, and all data were anonymised. All of the names that appear, other than from published sources, are pseudonyms. Participants were informed of the subject of the interview and had the opportunity to refuse to reply to certain questions, especially parents, for whom there was a possible emotional impact since the interview addressed intimate issues. To avoid this potential negative impact, the interview was conducted following general questions, avoiding intrusion and letting the interviewees narrate the story they wanted to tell in their own words. All participants received an electronic copy of the final result of the research (the PhD dissertation). Two interviews were conducted by phone due to geographical distance. All interviews were analysed using a qualitative thematic method: the content was analysed manually, identifying major repeated themes and then focusing particularly on less documented topics, such as prenatal diagnosis of intersex variation. Indeed, I was struck by the absence of this topic in social science published work. Other data used in the study came from my in situ observations of national and local hospital staff meetings (five specialised multidisciplinary meetings discussing ‘patients’ categorised as having ‘Disorders of Sex Development’ (DSD)) and observations from international and French medical conferences (4) that took place in 2013–2015. French conferences were open to the public (upon application) and were co-organised by the parents’ association, which was informed about my research. I consider official presentations during these conferences to be almost-public data. These observations helped me acquire a firm understanding of both official medical presentations and unofficial conversations and remarks during breaks, as well as non-public discussions between medical colleagues. I also collected data from documents belonging to a CAH parents’ association with the permission of the association’s president. These documents included the association’s bulletin (2004–2015) as well as information on regional-level meetings between parents and medical staff (2008–2012) and on an online discussion forum (2005–2018). As the in-depth research data were mainly collected before 2018, the findings may not be entirely applicable to current days. The debates and medical practices might have changed since then. Nevertheless, I have been paying careful attention to the subject, and no significant official changes involving new recommendations or new practices have been published in the meantime. Also, for this article, more recent data and publications were used, affirming the relevance of the results to 2024.

I started research on this topic in 2009 as a young non-French endosex (i.e., non-intersex) person living in France, and tried to develop an ethically situated position taking into account the difficulties for intersex people to exist as a collective and to make their voice heard in the then-current context in France where research and activism were almost non-existent. As ignorance production mechanisms lie at the heart of my research, they had to be taken into consideration not only as a theoretical tool used for analysing the materials but also as a reflexive reminder of my own position in this field of research. My research approach was to cast a critical feminist gaze on the processes by which biomedical science aims to reverse the traditional subject–object power relations. Intersex people have been an object of scientific interest for centuries. Accordingly, positioning my research by focusing on legitimate, dominant social actors—mainly doctors—was the methodological and ethical position I was most comfortable taking. This decision prioritised the study of the powerful rather than reiterating the objectification of the dominated population.

3. Intersex People and the Mechanisms of Ignorance Production

‘If we are to fully understand the complex practices of knowledge production and the variety of features that account for why something is known, we must also understand the practices that account for not knowing, that is, for our lack of knowledge about a phenomenon’.

(Tuana 2004, pp. 194–95)

The mechanisms of ignorance production about intersex are deployed consciously to a greater or lesser extent. They include, among others, the following: cultural invisibilisation, secrecy and lack of medical information, division among variations of sex characteristics, social biases in scientific studies and insufficient robust data. The mindset of erasure behind processes of ignorance production has had serious repercussions on intersex people. A better understanding of these processes should help to improve knowledge about intersex and the negative consequences of ignorance.

In this context, the first section of this paper proposes a categorisation of how ignorance is produced in order to help analyse how medicine, science and technology participate in a continuous process of erasing intersex bodies and lived experiences (Monro et al. 2021). This categorisation is possible thanks to the large amount of data now available on intersex people and my own research experience over the last 15 years. I suggest distinguishing three types of social erasure as they apply to intersex: cultural erasure, epistemological erasure and physical erasure. We will look at these three elements in detail below.

• Cultural erasure
  First, on a general level, ignorance is produced by removing the existence of intersex variations from the collective awareness and by allowing the biomedical sphere to monopolise it. Many social actors participate in this process, mainly state institutions, cultural content producers, medical staff and education professionals. Modern Western culture has established, deep in our perceptions, a sense of obviousness of sex binarism that intersex or other bodies that do not conform to norms risk destabilising (King 2016; Laqueur 1990).
  Throughout history, ‘hermaphrodites’,5 as a scientific category, have been positioned between visibility and invisibility, depending on the cultural context and power relations in place. In modern history (Daston and Park 1995), considerable scientific resources have been invested in the search for the ‘true sex’ (Foucault 1980).6 For example, at the end of the 19th and beginning of the 20th centuries, intersex people were publicly exposed, exoticised and dehumanised in the West during Circus shows (De Herder 2020). With the arrival of modern science (starting from the 16th century), intersex bodies, especially the genitals, have often been published in medical books and were super-visible in a dehumanising way. As Amato put it: ‘Ironically, it is this heightened visibility of intersex bodies that entails their invisibilisation’ (Amato 2016, p. 49).
  During most of the 20th century, intersex people were progressively invisiblised and confined almost exclusively to the medical sphere as ‘syndromes’ of ‘sexual ambiguity’. Consequently, they were absent from popular culture, arts and school curricula (King 2016; Sterling 2021) and were excluded from collective activities such as Olympic sports (Bohuon 2015). This had severe consequences on intersex lives. The title of an article written by Sarita Vincent Guillot, a French intersex activist, illustrates this reality: ‘Intersex: not having the right to say what no one told us we were’ (Guillot 2008, p. 37, author’s translation). A change occurred at the beginning of the 21st century with the development of ‘intersex intelligibility in the cultural imaginary’ (Amato 2016, p. 22). This led to intersex people appearing increasingly more often in books, movies and television, as well as in some educational programs. This fundamental shift was made possible by the creation of movements who fought for intersex people to be socially recognised, not as rhetorical or symbolic figures, but as embodied human beings fighting for their rights. This increasing visibility has been slower to develop in France than in anglophone countries (Raz 2023).

2.

Epistemological erasure

The second type of social erasure is epistemological, that is to say, the scientific production of ignorance, whereby access to information and data such as personal files (solicited by individuals) or medical archives (solicited by scholars) is restricted or partial. Many physicians interpreted Johns Hopkins’ biomedical protocol as a recommendation to conceal information from concerned people and sometimes also from their parents (Meoded-Danon and Yanay 2016). In practice, this involved advising parents to hide all details about a child’s intersex variation in order to prevent any potential ‘doubt’ children might have about their sex and gender (Karkazis 2008). This mechanism legitimises ignorance by promoting it as something for one’s own good (Funkenstein and Steinsaltz 1987). However, frequent silence and/or secrecy surrounding their bodies and medical history has negative repercussions on intersex people (Monro et al. 2024). Access to personal files is denied or restricted for many intersex people. Other obstacles to acquiring their own personal files include the financial cost and long waiting periods (Holmes 1994). Many are told that archive files have disappeared, been destroyed or lost. Sometimes, retrieved files are partial; for example, information is missing about medical interventions performed on them when they were young7. Some of these old practices, such as diffusing false information or denying access to medical files, were challenged during the 2005 ‘Consensus Conference’ in Chicago, but they have not entirely stopped: individuals are still receiving partial and mainly pathologised information about their variation and past medical interventions. This production of a subjective ignorance of one’s own body and history hinders the capacity of intersex people to produce their own knowledge. The consequence is that certain information is denied, and medical knowledge is positioned as the only legitimate form of knowledge. The subjectivation process of recovering one’s ‘stolen word’ (parole volée, Guillot 2008, p. 47) implicates the need for intersex people to struggle against ‘epistemic injustice’ (Fricker 2007) of knowledge being withheld from them (Bastien-Charlebois 2017). As a result of this ‘wrong’ which is done to them, ‘specifically in their capacity as a knower’, they are prevented from ‘collective interpretive resources’, which would enable them to make sense of their own social experience (Fricker 2007, p. 1).

Terminology issues are indeed very important. While many doctors present the current medical paradigm, which uses the DSD8 nomenclature as neutral, intersex scholars argue that, on the contrary, it contributes to the stigmatisation and the erasure of intersex variations using a strategy of divide and rule (Davis 2015; Lundberg 2017; Aegerter 2022). More specifically, they believe that it concentrates on biomedical issues, such as diagnosis and genetics, so as to exclude an increasingly larger number of variations from the umbrella category of intersex (e.g., hypospadias, Turner or Klinefelter syndromes, and even CAH).

Up to the present day in France, only authorised medical staff have access to data such as the number of intersex (or DSD) ‘patients’, the number and types of surgeries conducted or hormonal ‘treatments’ initiated and the age of the intersex person at the time of these interventions. These are the same people who promote the protocol itself. Indeed, a French Senate report indicated that even the Minister of Health admitted to having no data on the interventions and management of intersex variations (Blondin and Bouchoux 2017). This is still the case, though a report with data on the number of operations is to be published soon as directed by a law passed in 2021. The obstacles that non-medical scholars face in trying to access medical archives and data in France hinder the possibility for anyone other than medical practitioners to conduct intersex research. Since doctors and psycho-medical employees in hospitals are the only persons with access to these data, they are both the judge and jury—evaluating their own practices (Raz 2016). The inaccessibility to medical data, even for ‘patients’ themselves (and therefore persons for whom doctor–patient confidentiality does not apply), not only produces ignorance at the individual and collective levels but also difficulties in developing intersex studies in France.

3.

Physical erasure

The third type of erasure in the context of intersex people is physical obliteration. Embodied and concrete, it uses physical modifications to ‘repair’ or ‘normalise’ intersex bodies. It is related to a more general tendency to necropolitics (Mbembé 2003), whereby intersex lives are considered unworthy, attested to by degraded mental health, suicide attempts (Rosenwohl-Mack et al. 2020) and infanticide (Carpenter 2020; Behrens 2020), practices we may qualify as eugenic. Here, we especially think about voluntary interruptions of pregnancy that take place in different countries. That is attested by several scholars (Hashiloni-Dolev 2006; Jeon et al. 2012), open European data (Eurocat, European surveillance of congenital anomalies) and my own research in Israel (Raz 2015) and France (Raz 2019), where several doctors said they try to avoid these selective abortions but that they do occur in some centres. In the West, physical modifications to intersex persons, often without their consent, are widespread and commonplace. They consist of biomedical interventions which aim to erase any gap between social and medical norms. They include invasive non-reversible surgical operations (e.g., gonadectomies, vaginal creation or extension, cosmetic alteration of penile appearance, clitoris reduction or ablation, mastectomies), post-operation acts such as vaginal dilatations and hormonal prescriptions (in order to enlarge the penis, reduce the clitoris or body hair, replace physiological hormones that were stopped by gonad ablations, etc.) and laser epilation prescriptions. Some of these practices have evolved over time. Gonadectomies, for example, are not systematically conducted during a child’s first years of life, but no proof exists to this day that ‘normalisation’ interventions have stopped or diminished in France. Many of the debates surrounding intersex people concentrate on the physical mutilation of bodies already born; few mention other types of erasure of bodily variations in the prenatal phase, such as pregnancy interruptions of intersex foetuses and hormonal intervention aiming to prevent the ‘virilisation’ of the foetus. I will detail this particular practice in the next part of this paper.

The three categories described above are related and intertwined. Epistemological erasure is always entangled with concrete practices of obliteration or regulation of bodies. The various processes of erasing intersex have enabled the illusion of a natural, exclusively binary sex to be maintained. This illusion serves as the postulate, the justification and the result of intersex ‘normalisation’ practices. In this sense, the production of ignorance about intersex stems from the belief that the existence of intersex people might disturb the cognitive bases of a society (Funkenstein and Steinsaltz 1987). In the following section, I will concentrate on a specific case study of these mechanisms situated in the biomedical sphere: the prenatal regulation of intersex bodies, mainly through the controversy surrounding prenatal hormonal ‘treatment’ for Congenital Adrenal Hyperplasia (CAH). Prenatal medicine is a field inherently filled with uncertainties, giving rise to numerous technical and ethical debates. My example aims to illustrate the way mechanisms of ignorance production apply in real-world settings.

4. Case Study: Uncertain Knowledge and the Controversy Surrounding Prenatal ‘Treatment’

4.1. The Erasure of CAH Girls: A Short History

The history of the medicalisation of CAH illustrates the three types of erasure, particularly the physical and epistemological ones. CAH is related to the production of hormones by the adrenal glands. Already in the 1930s, doctors knew that there was a relationship between the hormonal activity of the adrenal glands and the ‘masculinisation’ of genitals in XX chromosome individuals. Salt-losing CAH is one of the several CAH variations and can be life threatening. It was only in 1949 that Lawson Wilkins, the founder of paediatric endocrinology, discovered that cortisone was an effective substance in reducing androgen production in a baby being treated at the Johns Hopkins Hospital (Eder 2010).

Wilkins claimed that children with XX chromosomes should be assigned as female, irrespective of their phenotype. This recommendation was revolutionary since many individuals with CAH were historically raised as boys before CAH diagnosis and the new protocol were introduced. The medicalisation of birth, new genetic technologies and other social and technological transformations made it appear obvious to physicians like Wilkins that people with XX-CAH are ‘true’ girls. Following the introduction of the Johns Hopkins protocol and the increasing ‘technological imperative’ (Koenig 1988), early interventions (i.e., hormonal therapy and surgery) to ‘feminise’ genitals via clitoridectomy and vaginoplasty were promoted and became routine practice typically between one and four years of life (Karkazis 2008).

Wilkins also promoted the idea that CAH variations must be distinguished from other intersex variations since the children were ‘authentic girls’ (Bernard et al. 1962). As it was related to a chronic endocrine condition, this variation was defined as a distinctive category with, supposedly, no relation to gender issues once the ‘virilised girls’ were ‘normalised’ (i.e., once all signs of what could be considered to be their ‘ambiguous’ sex characteristics were obliterated). This erasure transformed the way intersex people and their parents understood and experienced CAH; specifically, it made it difficult for them to relate their experience to the larger intersex reality of medical invalidation, thus exemplifying cultural erasure as well as epistemological and physical erasure. Due to the lack of cultural representations and social existence of the intersex category, CAH physical erasure could not be apprehended by concerned people as a social rather than a medical process.

Most parents were, and still are, influenced by this paradigm of erasure. The French parents of intersex children association Surrénales (the Adrenals) endorses medical discourse and practice. One of the mothers I interviewed who was a member of this association had a 14-year-old daughter with CAH. She described her understanding of her daughter’s condition—probably because that is how it was explained to her—as follows: “There is a malformation of the genitals because she didn’t have… a vagina. So, they had to build all this and, well, after… we all agree that it’s a girl and that there is absolutely no doubt about it.”9 Thus, girls with CAH undergo physical and epistemological erasure by the medical and discursive removal of anything that might raise ‘doubt’ about their sex.

4.2. The Birth of Prenatal DEX

The physical erasure of intersex traits has taken place mostly after birth but also prenatally. This new temporality—even before a child is born and has a social existence—helped reinforce the three types of erasure: physical, of course, but also epistemological and cultural. More specifically, the erasure of the ‘virilisation’ of CAH girls that has mainly operated through postnatal biomedical interventions starts during pregnancy. From the late 1960s to the end of the 1970s, various research projects sought to find a means to identify intersex people prenatally. Throughout these 10 years or so, several attempts were made to find a marker for a specific genetic diagnosis, but all resulted in failure (Frasier et al. 1975). Elsewhere, in 1965, British researchers (led by gynaecologist Thomas Norman Jeffcoate) suggested the possibility of Prenatal Diagnosis (PND) of congenital adrenal hyperplasia through an analysis of steroids in amniotic fluid. From the beginning, the stated goals of this research were to improve postnatal care and to find a prenatal ‘treatment’ (Merkatz et al. 1969). At that time, it was still impossible to put this into practice. In the following years, efforts were made to elucidate biological mechanisms by genetically locating intersex variations and, at the same time, proposing ways of applying this knowledge to PND.

In 1969, a team of physicians in New York published the results of their study to test the validity of hormone sampling techniques during pregnancy (Merkatz et al. 1969). Between 1974 and 1977, studies identified the genetic link causing this variation, allowing a more precise biological diagnosis. In 1979, the prenatal diagnosis of CAH using amniocentesis became a reality in two French laboratories: one in Paris and the other in Lyon (Floret et al. 1980). This hormonal-based diagnosis technique spread in the 1980s, both in France, as illustrated by a retrospective study listing 274 pregnancies monitored for CAH in Lyon between 1979 and 1993 (Forest et al. 1993), and internationally.

In the early 1980s (i.e., shortly after their discovery of the hormonal-based technique for prenatal diagnosis in 1979), the same scientific team in Lyon made another discovery concerning hormonal intervention during pregnancy to reduce the ‘masculinisation’ of foetuses with CAH. Specifically, paediatrician Michel David and endocrinologist Maguelone Forest experimented with giving hydrocortisone to pregnant persons with a history of CAH to investigate the substance itself and its effects (David and Forest 1984). Despite the high cost of this ‘virilisation’ treatment, administering hormones to pregnant persons progressively became the norm in families considered ‘at risk’. Although the procedure itself is technically quite simple, its timing is important: the aim is to start ‘treatment’ as soon as possible, even before one had diagnosed whether the foetus actually has CAH or not. Results for the first six treated cases were published in 1984. One of the doctors involved recounted to me in an interview:

“We gave a fairly large dose of hydrocortisone, and then we still did an amniocentesis at mid-gestation to measure 17OHP in the amniotic fluid to know if it worked. Well, the hormone was very high, so we said to ourselves that it’s not enough, so we increased the dose, and the child was born almost normal. She just had a clitoris a little bit big, but she’s never had surgery, she’s never had any problems, and things have always gone well in this family”10.

Clearly, the goal of this intervention is the birth of a child considered ‘normal’ and the physical erasure of intersex traits. In the words of one of the specialists involved in the first PND treatment, it aimed to ‘restrain’ the virilisation (Ibid.) of XX foetus’ genitals, mainly the formation of the vagina, and to avoid having a clitoris judged too large. Initially conceived as a tool to fight against ‘serious’ and ‘incurable’ diseases, as French law puts it,11 prenatal diagnosis in intersex persons became a field where sex and gender were regulated. The purpose of this prenatal ‘treatment’ was not to prevent the metabolic risks linked to the CAH condition. From the outset, it was used explicitly as a tool to avoid and erase ‘atypical’ sex. The results presented in the aforementioned study in 1984 suggest that it was thanks to the prenatal intervention that things had ‘always gone well in this family’. Indeed, the medical literature on PND DEX ‘treatment’ often emphasises its alleged benefits: genitals are less virilised, and therefore, surgery is sometimes considered unnecessary (Xu et al. 2020). A French retrospective study of foetuses who received DEX found that the majority of foetuses with CAH who received DEX at an early stage of gestation ‘had normal external genitalia at birth’, thus indicating that ‘early DEX initiation is [essential] to prevent any surgery’ (Tardy-Guidollet et al. 2014).

After these early experiments, this practice of administering hydrocortisone to pregnant persons at risk using a molecule called dexamethasone (DEX) began to extend to other countries. As Dreger states, “‘at-risk’ mothers-to-be throughout the world started being offered prenatal dexamethasone” (Dreger 2015, p. 361). For example, in New York, a clinical team led by Dr. Maria New started to experiment with this procedure in 1986, leading to the centre she led becoming the largest centre of prenatal hormonal intervention worldwide for women at risk for CAH. Interestingly, back in France, Paris’s largest paediatric hospital—Necker-Enfants Malades—did not immediately provide this service. Dr. Bonnet was a surgeon practising there at the time. According to her, some endocrinologists were reticent about the ‘treatment’, fearing ‘complications’. However, she considered that the idea of prenatal ‘treatment’ “is very good. If we can suppress androgens, we will have a much smaller clitoris at birth”.12 The history of the invention and extension of the prenatal DEX show how mechanisms of physical erasure of intersex traits were implemented. This erasure stems from an epistemological and cultural perception expressed by physicians who consider intersex bodies (here, a large clitoris) unacceptable. The epistemological erasure also concerns producing ignorance about this ‘treatment’, a process I will detail in the next section.

4.3. International Controversy: Sidelining Uncomfortable Knowledge

While this ‘treatment’ was generally considered by doctors as something positive, ‘to avoid surgery’,13 they gave no consideration to potential physical side effects to the pregnant person or the foetus or to ethical issues of the procedure. This section aims to illustrate how these mechanisms of ignorance production were activated, paying special attention to epistemological erasure and its consequences.

The fear of harmful effects (on the pregnant person and/or the foetus) was present from the beginning of DEX therapy. However, all criticism was quickly dismissed by the pioneering doctors, who asserted that “there is no substantial evidence of foetal effects of high doses […] or low doses of glucocorticoid therapy in human pregnancy” (David and Forest 1984, p. 799). One of the members of the original French team also stated to me during an interview that psychological and physical studies were conducted on the first cases to verify that there were no adverse effects and that nothing was found to validate the fear of such effects.14

The assertion that no harmful side effects were found in the early years was supposed to reassure critics. Nevertheless, the logic behind the defence of prenatal DEX is based on the idea that we must prove that a ‘treatment’ is harmful after its use instead of proving that it is not harmful before its use. Hence, the absence or quasi-absence of well-founded knowledge about the long-term effects did not hold back the use of the procedure. It seems, therefore, that epistemological erasure, via scientific production of ignorance, is a central part of prenatal DEX.

DEX ‘treatment’ started to become controversial by the end of the 1990s, when it began to be criticised both from inside and outside the medical sphere itself. This controversy, which continues today, does not call into question the effectiveness of DEX, which is measured in terms of the ‘feminisation’ of the genitals. Rather, it is based on two criticisms: the first regards ethical, sociological and political concerns about the legitimacy of this ‘treatment’ (whose main goal was to ‘normalise’ bodies); the second regards its potential harmfulness.

The first criticism came from activist groups and social science researchers who warned about the ethical consequences of the prenatal erasure of intersex variations. From a bioethical point of view (Dreger et al. 2012; Dreger 2015), some persons oppose the very idea of wanting to ‘devirilise’ foetus genitals. This viewpoint affirms that even if the ‘treatment’ is not harmful, it should be questioned since it is an intervention that does not alleviate a vital danger but instead ‘normalises’ bodies by trying to erase not only intersex physical traits but also their social and cultural existence (Holmes 2008). This criticism does not find much support in the medical sphere, which considers that DEX ‘treatment’ helps to avoid painful surgery. This argument ignores the fact that there is also an alternative route with no DEX and no surgery. Thus, prenatal erasure of intersex is justified by medical claims of avoiding physical postnatal erasure that is never questioned.

The second criticism came from the fields of biomedicine and bioethics and specifically concerns the benefit/risk ratio. Stakeholders felt there could be significant adverse effects of the ‘treatment’ on foetuses, most of which are unaffected by the condition.15 Another element of this criticism was the issue of insufficient data or missing data of good scientific quality on the long-term follow-up of children and adults exposed to DEX ‘treatment’.

For more than twenty years now, doctors have been increasingly warning and arguing that the safety of this intervention has not been proven. The controversy began in the late 1990s with the publication of articles by Walter L. Miller, a paediatrician endocrinologist and professor at the University of California (Seckl and Miller 1997). In the conclusion of an article from 1999, he warned that: “the ethics of needlessly subjecting 7 of 8 foetuses to an experimental therapy with unknown long-term consequences remain unresolved because the long-term safety and outcome have not been established” (Miller 1999, p. 538). He also stated that “therefore, prenatal treatment of CAH remains experimental” (Ibid., p. 538). Potential harmful effects could affect all foetuses exposed to DEX: growth retardation, hypertension and effects on ‘emotionality’.16 Miller did not initially recommend prohibiting the ‘treatment’ but suggested framing it with safety measures in mind because “it is important to be cautious now so that we do not have regrets later” (Ibid., p. 538). In his opinion, prenatal DEX should only be used in large specialised centres that collect cases and establish a research protocol. Additionally, he claimed that physicians should obtain written, informed consent from the pregnant women concerned. In 2008, Miller reiterated his criticism and maintained that “this experimental treatment is not warranted and should not be pursued, even in prospective clinical trials” (Miller 2008, p. 17).

Other criticisms of prenatal DEX in the last twenty years highlight the production of ignorance, including the absence of studies, whether animal- or human-based, and determining whether or not foetal dexamethasone is harmful (McCann-Crosby et al. 2018). Indeed, several animal studies on in utero exposure to corticosteroids have raised questions about the safety of this ‘treatment’ (Hirvikoski et al. 2008). As for human studies, besides the Lyon research team’s publication of their safety results in 1993 (Forest et al. 1989, 1993), a Swedish team was the primary force accentuating the debate in the 2000s, with different publications on a follow-up study of their ‘patients’. They argued that their methodology was more scientifically rigorous than previous ones. As a precaution, and because of the possibility of long-term consequences, they only used DEX within the framework of a well-controlled clinical trial that started in 1999 (Hirvikoski et al. 2007). Their studies constitute the first short- and long-term neuropsychological assessments of children and adolescents treated in utero based on children’s self-reports. In 2007, the first results were published and suggested that exposure to DEX in utero would have effects on brain development, with problems linked to verbal and visuospatial working memory, moderate behavioural abnormalities (shyness, strong emotionality, low self-esteem and poor academic performance), as well as defects in growth and psychomotor development.

Armed with these results, the team addressed the regional ethics committee in Stockholm in 2010, saying that it would be more prudent to interrupt recruiting ‘patients’ for their study, pending results from larger and more conclusive studies worldwide. The Swedish team, with its relatively small samples, continues to publish and speak at international conferences, saying that their study results to date are contradictory or unreliable and require further exploration (Hirvikoski et al. 2007; Lajic et al. 2018). In a 2014 interview, Dr. Dupuy asserted that, unlike the Swedish team, the Lyon team conducted large and reliable studies on their ‘patients’:

“The problem is that the Swedes, they had, I don’t know, 9, 10, 12 cases, while I had 50 or 60 antenatal diagnoses of hyperplasia with the treatment. I don’t know how many I have done, more than a hundred. […] And each time I have done a study after 5 years, 10 years, to review our patients, to review everything that we thought we had to follow. I’ve done that several times, eh, and I’ve always published our work on it. No, we didn’t find anything salient.”17

In terms of studies with relatively large samples, two suggested that there may be long-term harmful cognitive effects (New et al. 2001; Meyer-Bahlburg et al. 2012) and that more comprehensive studies need to be implemented (Forest et al. 1998). However, follow-up studies in large medical centres struggled to find and recruit ‘patients’ for the follow-up questionnaire, an aspect which also makes the results unreliable (Dreger 2015). One of these studies—conducted in New York—attempted to evaluate the long-term outcome of women and children exposed in utero. However, only half of the families included in the study actually responded (72 out of 154 questionnaires sent) (Speiser et al. 2010). In short, “the medical-scientific literature was utterly devoid of well-controlled studies of efficacy and long-term safety of prenatal dexamethasone for intersex prevention” (Dreger 2015, pp. 228–29). These various points highlight that knowledge about the adverse effects of prenatal DEX is still insufficient and uncertain.

A lack of scientific knowledge about the safety of prenatal DEX ‘treatment’ must not be perceived as a coincidence: the relative ignorance about this ‘treatment’ is part of an ignorance production mechanism related to the epistemological erasure of intersex. We see that science does not impose its truth in a linear and homogeneous manner and that physicians choose to position themselves differently in relation to the controversy: the way of presenting the risks and evaluating them largely depends on the point of view adopted by the individual physician and the staff concerned. More specifically, some doctors who administered prenatal DEX—notably Maria New—put forward a double discourse. In front of parents, they promoted the ‘treatment’ as effective and safe; in front of the authorities, they sought funding to carry out follow-up studies on their own ‘patients’, claiming that the effects of the drug had not yet been determined (Dreger 2015). In other words, on the one hand, they were promoting the practice of administering DEX by attempting to ignore uncertainties; on the other hand, they were working at promoting research that would resolve these same uncertainties, which were ignored in clinical practice.

Alice Dreger, a researcher and former member of the Intersex Society of North America (ISNA), which was the first intersex activist group in the world, joined forces with health professionals to fight against the use of DEX. She distributed information about the potential harmfulness of prenatal DEX, demanded a stop to its use and called for an investigation by (American) federal authorities, explaining—with the support of others—that it violated the fundamental principles of clinical medical research. In her 2015 book, Dreger recounts this fight, which included standing up to doctors during international congresses, including in January 2010 when she confronted Maria New. In the French context, promoters of prenatal DEX were very aware of this growing public opposition worldwide. Dr. Dupuy recalls: “I was attacked at conferences, and I always said thank you; I never argued”.18

The numerous alerts on prenatal DEX right from the outset gave rise to new international recommendations. As early as 2002, the Paediatric Endocrine Society and the European Society for Paediatric Endocrinology issued a joint consensus statement. In 2010, the Endocrine Society issued a warning following a report affirming that ‘the evidence regarding foetal and maternal sequelae […] is of low or very low quality due to methodological limitations and sample sizes’. That report recommended that prenatal DEX should be used only in clinical trials and that parents had to be informed about the possible risks of the ‘treatment’. Other countries issued the recommendation that ‘the administration of dexamethasone for prenatal treatment of CAH only take place as part of research projects that have ethics approval and patient follow-up protocols’.19

The above paragraphs, focusing on the substantial international controversy surrounding prenatal DEX ‘treatment’ over the last twenty years, highlight that doctors and health institutions have been well aware all along that they have been using a molecule whose safety has not been verified scientifically. Indeed, the poor quality of the available data reproduces mechanisms of ignorance and uncertain knowledge production. It seems like this ignorance production was motivated by a larger aim to physically erase intersex as early as possible, thus socially and culturally erasing the existence of intersex people.

The current use of DEX is controversial and considered experimental by the latest international guidelines (Speiser et al. 2018). Its use is not well documented, but a European medical study provides some data attesting that prenatal DEX continues to be used in France: from 2002 to 2011, about 154 foetuses were subjected to prenatal DEX in the French participating centres (Nowotny et al. 2022).

4.4. France and the Maintenance of a Policy of Ignorance

The epistemological erasure of intersex, as evidenced by the absence of official data about medical practises, is reinforced by the dismissal of existing knowledge. As described above, prenatal DEX was invented in France and then spread internationally during the 1990s. From the outset, ‘France is the country that has used it the most’20 as one of the Lyon team asserted in a 2009 meeting. Despite a great deal of international criticism (see above), French doctors continue to use prenatal DEX, thereby participating in a general logic of erasure of ‘atypical’ sex culturally, epistemologically and physically. This use is reflected in an information note from the Parisian hospital Robert Debré, published in 2013 and still available on the hospital website in 2024 (see Appendix A). The management recommendations on CAH published by the HAS (French High Authority of Health) in 2011, mentioning the possibility of prescribing prenatal DEX, is still in effect in 2024.21 Furthermore, the head of the national expertise centre in Lyon, Dr. Pierre Mouriquand, publicly confirmed this French approach during a French Senate special committee meeting:

“This treatment is very controversial because the side effects can be serious […].” Dr. Pierre Mouriquand

“These are the reasons why certain countries—Sweden or the United States—have abandoned these hormonal treatments.” Maryvonne Blondin

What about in France?

“We continue to prescribe them.”22 Dr. Pierre Mouriquand

French promoters of prenatal DEX consider the controversy to be a foreign one that does not concern them; as one such promoter said: “the American criticisms are not very valid”.23 The first promoters of prenatal DEX from France and the US are now retired, and a new generation of practitioners has emerged, aware of the international controversy surrounding the ‘treatment’. Some continue to advocate the procedure, echoed in the words of a French gynaecologist–endocrinologist: “we see couples, we offer treatment, we are for it. The benefit is greater than the risks”.24

Given the risks involved, some French doctors remain cautious and express more nuanced points of view. During a conference titled ‘Congenital Adrenal Hyperplasia’ organised by Surrénales on 5 April 2013, an important French paediatric endocrinologist spoke publicly and conceded that “there is a debate, not enough data, methods not well-founded enough, without a control group”. However, despite these reservations, the staff at the hospital continues to prescribe prenatal DEX in certain cases.

When talking about DEX ‘treatment’ for a case during a staff meeting, Dr. Schapiro, a geneticist at another Parisian hospital, said: “sometimes it works, sometimes it doesn’t. And when it works, we don’t really know the collateral effects”.25 But this attitude of uncertainty did not lead to a direct condemnation of this ‘treatment’; rather, it led to the proposal of a preventive approach using pre-implantation diagnosis and ‘to only re-implant the boys’ (Ibid.). In other words, in order to protect CAH boys from being exposed to a potentially harmful intervention, these medics suggested pre-selecting embryos by erasing all of the XX ones in advance, hence making the prenatal DEX unnecessary. The pre-implantation procedure is an invasive, exhausting medical procedure for (to be) pregnant persons.

The above examples highlight the fact that while some French doctors question the use of DEX, they do not go so far as to condemn it. The well-being of pregnant persons is rarely mentioned or taken into account during discussions about prenatal DEX. As far as pregnant persons are concerned, doctors generally state that ‘the impact of the treatment on the mother seems minor’26 even though the national protocol mentions that there are side effects, and in particular, ‘excessive weight gain, stretch marks, maternal discomfort, sleep disorders, high blood pressure’.27 Basically, doctors consider the effects of the ’treatment’ to be marginal compared to the ultimate objective, in other words, what they see as the ‘benefits’.

4.5. Parents’ Experience and Medical (Dis)information

The goal of erasing intersex justifies, in the view of some medics, the dismissal of physical risk to the pregnant person. However, the experience of pregnant persons highlights that the impact of prenatal DEX on their health is a major cause for concern that is frequently ignored. This section continues to explain how the three types of erasure are entangled and related to ignorance production mechanisms that do not directly concern intersex but, more broadly, parents’ and children’s bodies.

The potential long-term risks and negative side effects for pregnant persons and their children lead many women to wonder about taking prenatal DEX. For example, research contributor Annie’s experience brings to light a complex reality and highlights the importance of considering not only medical reasoning but also the consequences on pregnant bodies in any evaluation of the ‘treatment’. Coming from the Paris region, her first affected daughter was born in 2004 and diagnosed with CAH at birth. A year later, she became pregnant again and immediately began prenatal DEX. Although she was being followed by a major Parisian paediatric hospital, she said she was not warned of the side effects of taking DEX during pregnancy, effects which she experienced severely. In addition to very significant weight gain, she said:

“I had water retention, I had a beard; most of it went away after giving birth. I wasn’t sleeping. I was a bundle of nerves; I vacuumed all the time. I wasn’t well. […] I asked if I could stop it or reduce [the ‘treatment’] and I was told no, and I had to continue for a month after the birth.”28

When I asked her if she would take the ‘treatment’ if she had to do it again, she said no. The Surrénales Association (see above) supports prenatal hormonal intervention and encourages parents to trust what doctors tell them. In an online forum created by the association, some members express their disagreement and their fear of the procedure. They speak about the dilemma they face and the lack of clear information given to them.

Following international criticism and recommendations, the current French use of prenatal DEX gives key importance to the issue of information (i.e., A Parisian hospital’s note Appendix A). Doctors are now obliged to inform parents of the potential risks of this prenatal hormonal intervention as part of a new ‘therapeutic alliance’, a concept based on the promotion of the informed choice of medical care users who must have all the information necessary for their decision, and the freedom to choose what suits them. However, several studies in social sciences have underlined that this therapeutic alliance is not between equals because, currently, the doctor–user relationship is entangled in the dynamics of power and authority (Alderson 2001; Vassy 2006). This alliance is, therefore, a fiction of autonomy since it is structured by strong pressures from doctors (Press and Browner 1997). As Annie and other parents who contributed to my research affirmed, information about the state of the science of prenatal DEX is not always clear or complete; some indicated that the potential adverse effects on the foetus were not even mentioned to them. Furthermore, unconscious representations influence the way parents imagine having an intersex child, so the verbal and non-verbal language used by doctors affects their perception and acceptance of intersex people and of prenatal DEX (Streuli et al. 2013).

In the context of prenatal DEX, health professionals, for the most part, inform parents of potential side effects for the pregnant persons and the child but use strategies to convince them that despite these risks, ‘treatment’ is almost obligatory. This is highlighted in the testimonies I collected a few years back—this may have changed since—where parents declared they were put under a lot of pressure by doctors to agree to ‘treatment’, the latter insisting on the so-called benefits instead of the disadvantages or risks. This strategy reduced the pregnant person’s agency and engendered a feeling of helplessness in the face of this pressure. In this way, medical professionals placed the responsibility on pregnant persons who, if they refused the ‘treatment’, might feel guilty for not having acted to prevent the ‘virilisation’ of their daughters. Pregnant persons were faced with the dilemma of having to choose between the desire to follow the medical advice of prenatal erasure of intersex traits and that of protecting themselves and their future child from side effects.

All of these testimonies confirm that in France, prenatal DEX is not only strongly recommended by the medical profession but that parents are sometimes pressured into accepting it. Parents’ ignorance is produced by the way information is conveyed or concealed, the uncertain context, and a timeline that puts pressure on them to decide and which does not leave space for them to seek further information. This ignorance is a mechanism that is not only cognitive but also emotional, as it produces negative feelings such as anxiety and guilt. The lack of concealment of information from parents in order to promote prenatal DEX exemplifies the epistemological erasure not only of intersex but of the general health of non-intersex family members. It illustrates the importance given to the physical and cultural erasure of intersex altogether.

5. Discussion

The internal tension contained in the expression ‘uncertain knowledge’ in the title of this article underlines that all scientific knowledge, technologies and medical practices on intersex belong to a specific regime of truth whose premises and consequences are left unquestioned, thereby generating a regime of ignorance. The latter continues to function in France today despite numerous uncertainties and controversies inside and outside the medical sphere.

The context of prenatal DEX allows us to reflect on the way in which certain medical practices are maintained despite criticism and uncertainty. It also reveals how biomedical practices that claim to be based on consensual knowledge are embedded in a local framework. Why do French doctors sideline criticism of this ‘treatment’ or even ignore it? This approach follows similar French public health scandals surrounding the iatrogenic effects of hormones (growth hormones, contraceptive pills, etc.). It is remarkable that prenatal DEX was introduced in the 1980s, a period when it was discovered that administering hormones to pregnant persons could have serious consequences, mainly after the scandal around diethylstilbestrol (DES) in the 1970s (another synthetic oestrogen given to pregnant persons which had serious risks) (Bonah and Gaudillière 2007; Fillion and Torny 2016)29. The denial of potential harm is somewhat incoherent with the highly preventive biomedicine and injunctions to pregnant persons concerning, for example, alcohol or smoking (‘zero risk’).

It is unsettling to observe the similarities between DEX use, the mechanisms of ignorance around DES, and the kind of intentional blindness or historical amnesia that accompanied it. Furthermore, in France, there is a quasi-absence of public controversies about hormonal use in general: the contraceptive pill or replacement menopausal ‘treatment’. This is due in part to the way feminist movements have thought about issues related to the body and medicine and, more generally, to the relatively high trust women have in the French medical system and technological apparatus, which is often seen as emancipatory for women (Löwy and Gaudillière 2004). The reaction time of the French medical profession in the face of alarm signals was very long, and a change only occurred when the women directly concerned managed to make it a public scandal in France.

The maintenance of such practices despite the continued scientific debate regarding the safety of DEX testifies to a denial, not only of history but also of current published data: dismissing ethical and scientific criticism allowed French specialists to reproduce the erasure of intersex. French doctors’ resistance to criticism also indicates the importance given by the erasure of intersex variations, which, in our specific context, concerns the ‘virilisation’ of the genitals of XX-CAH. This goal, taken for granted by doctors, seems so important to them that they are willing to promote a ‘treatment’ which is, at best, under-studied and, at worst, harmful to exposed foetuses, mostly non-CAH babies. Accordingly, preventing ‘atypical’ sex would appear to have a much higher priority than the well-being of pregnant persons and their future children. Knowledge about prenatal DEX, which is uncertain and insufficient, is sidelined so as not to infringe on the assumed benefits of this ‘treatment’: erasing intersex from bodies and minds.

Overall, the case study provided allowed me to examine the mechanisms by which biomedical science not only produces knowledge but ignores or excludes knowledge that contradicts its paradigm. Several dimensions of ‘uncomfortable knowledge’ appear to be deliberately excluded because they threaten a pre-existing system. Hence, the establishment of practices considered acceptable or even taken for granted is inseparable from processes minimising uncertainty.

This ignorance points to a larger effort to erase intersex and demonstrates how the three types of erasure are linked and form a circular system that justifies and maintains itself: physical erasure is enabled by the cultural erasure of intersex and the incapacity to socially accept and represent intersex bodies and experiences. But this physical prenatal or postnatal erasure also reproduces cultural ignorance of intersex lives. This becomes possible because several epistemological mechanisms of erasure are used: producing ignorance impedes intersex people and their parents from conceiving alternatives to the medicalised path of physical obliteration; concealing or dismissing data and information on prenatal DEX also enables the maintenance of the system of erasure by putting ‘normalisation’ as a higher goal than physical integrity, diversity or general health. Thus, ethical issues around it are minimised and considered to be marginal within the larger process of erasing intersex.

6. Conclusions

This paper examined how medicine, science and technology participate in a continuous process of erasing intersex variations by practices of invisibilisation. In turn, this logic of erasure reveals a larger logic of ignorance production. The case study of the use of prenatal DEX illustrates the three types of erasure presented at the beginning of this paper, cultural, epistemological and physical, and the way they are entangled and co-produce each other to maintain the system of medicalising intersex bodies. The social difficulty of conceiving a body with a large clitoris in relation to binary categories underlies the origin of medical justifications for erasure. It is thus related to the cultural erasure regarding any variation of the female/male normative binarism. It also demonstrates epistemological erasure by illustrating how biomedical science sidelines certain data and minimises uncertainty or long-term risks such as side effects. Finally, the use of DEX is directly connected to physical erasure since it aims to obliterate the existence of intersex traits by medically intervening as early as possible on intersex children’s genitals.

By unravelling mechanisms of ignorance production, a new paradox emerges: the increase in biomedical knowledge and techniques also produces many dilemmas and debates. The questions left unanswered are whether the production of ignorance is necessarily conscious and whether it stems from an intention to hide information or whether it is a social process, largely unintentional, which can be observed by examining precisely what knowledge is left unknown, which facts are minimised and which realities do not interest science. It is important to bring to light the epistemological and political consequences of this regime of ignorance and its paradoxes in order to build an alternative field of knowledge about intersex lives and represent other paths than early medicalisation and erasure of intersex.