Constituting ‘Problems’ through Policies: A WPR Approach of Policies Governing Teenage Pregnancy in France

Abstract

This article examines the construction of policy problems through the exploration of the theoretical space opened up by the ‘What’s the Problem Represented to be?’ (WPR) approach developed by Bacchi (2009). It proposes a critical analysis of current policies governing teenage pregnancy in France through the deconstruction of the structures that have participated in shaping the ‘problem’ today. Focusing on discursive practices, the analysis unveils the political essence of the knowledge that constitutes the problem of ‘teen pregnancy’, and points to the constant flux characterising it, which is captured and stabilised in policymaking through problematisation. It is argued that these conclusions call for greater self-reflexivity in research and policymaking and prompt critical and feminist researchers to engage with disrupting current policymaking rationality.

Introduction

An in-depth study of policy through the lenses of feminist, critical and postmodern theories can provide a window into diverse facets of the techniques of governance of a nation, both past and present. These theories have proved valuable in challenging policies that have damaging impacts on marginalised identities. They have also proved powerful in destabilising ‘rational’ and ‘scientific’ models for policymaking and analysis, as well as normative practices, assumptions and knowledge upon which policies are developed. Broadly, this field of research has contributed to answering the complex and ambitious question: why do some policies fail to ‘support’ exactly those for whom they seek to advocate? Building on Bletsas’s analysis (2012) of the ‘What’s the Problem Represented to be?’ (WPR) approach developed by Bacchi (2009) and as an attempt to address the question, this article proposes a theoretical examination of discursive practices drawing on the case study of teenage pregnancy in France. There is a dearth of research that applies a constructivist lens to the policy response to teen pregnancy and motherhood in continental Europe, which appears even more true for the case of France (Le Den, 2014).

The interest of the article is to demonstrate how the WPR approach allows one to transcend the debate of the cause of policy problems while shedding light on two major features of ‘problems’ relating to policymaking: 1) knowledge is inherently political, and by extension, all ‘problems’ do not exist independently of outside forces, but only in relation to specific mentalities and models of governance; 2) phenomena underlying ‘problems’ are not fixed per se but rather remain continuously in flux. As such, it is possible to observe how the problem representations of teen pregnancy have changed over time and context based on fluctuations at the level of hegemonic discourses, knowledge and authority over the issue, yet remain focused on the individual level.

My argument comprises two points. First, the article uses the WPR framework to examine the problem representations of teenage pregnancy in relation to knowledge production, which constructs the issue as primarily individual, rather than structural. Through the dismantlement of discursive practices underpinning the representations coupled with a cursory analysis of the genealogy of the ‘problem’, the very validity of teenage pregnancy as an independent and apolitical ‘problem’ is called into question. This part shows that the problem of ‘teen pregnancy’ exists only in the interstice of other problems – namely, unwanted pregnancies, single motherhood, teenage sexuality and welfare dependency, whose representations are contingent and unstable.

Second, the article prompts the consideration of the implication of such findings for policymaking and contemporary scholarship. It is argued that the political essence of knowledge calls for enhanced self-reflexivity in all engagement with policymaking. More specific to feminist and critical scholarship and emphasising the political effects of research, the purpose becomes one of disruption of ‘problems’. Possible paths forward are advanced in an effort to dismantle hegemonic knowledge and rethink the policy ‘problem’ of teenage pregnancy from a transformative perspective, though some inevitable tensions emerge.

Theoretical framework

In the last decades, the emergence of theoretical frameworks critically analysing knowledge production and constitution expanded the tools for policy analysis. They have illuminated previously unquestioned aspects of methods traditionally dominating the field as a means to understand contemporary social issues. Of increasing influence has been the active acknowledgment of the role of policy in constructing knowledge about social realities through language and discourse. Discursive approaches are concerned with exposing the contingency of policies through unravelling their political dimension, and deconstructing notions of neutrality, objectivity and truth (Durnova and Zitoun, 2013). The article focuses on discursive practices, which describe ‘those practices of knowledge formation by focusing on how specific knowledges (“discourses”) operate and the work they do’ (Bacchi and Bonham, 2014: 174).

As part of this theoretical current, Bacchi (2009) proposes the framework ‘What’s the Problem Represented to be?’ (WPR) approach that enables analysis at the level of the construction of policy problems. There is a widespread tendency to assume that problems have an objective and independent existence, while in fact they are no more than a construct resulting from our mode of simultaneously acting and thinking (Bacchi, 2009). The WPR approach tries to understand how we are governed and what the repercussions are for different groups of people. The focus is shifted from policy as a lens of problem resolution to policy as constructive of problems. Problematisation is thereby conceived not only as a key feature of policymaking, but also central to the process of governing. The ultimate goal of the framework is to encourage continuous reflexive questioning and critique of governmental problematisation practices in order to challenge problematisations that have damaging consequences for certain groups (Bacchi, 2009; Bacchi and Goodwin, 2016).

The WPR approach includes seven questions:

1. 1 What’s the problem represented to be in a specific policy or policies?

2. 2 What deep-seated presuppositions or assumptions underlie this representation of the ‘problem’ (problem representation)?

3. 3 How has this representation of the ‘problem’ come about?

4. 4 What is left unproblematic in this problem representation? Where are the silences? Can the ‘problem’ be conceptualised differently?

5. 5 What effects (discursive, subjectification, lived) are produced by this representation of the ‘problem’?

6. 6 How and where has this representation of the ‘problem’ been produced, disseminated and defended? How has it been and/or how can it be disrupted and replaced?

7. 7 Apply this list of questions to one’s own problem representations.

Instead of applying the WPR approach in methodological form, this article uses Bletsas’s suggestion to analyse the space created by the application of the first three questions of the WPR approach in which ‘it becomes possible to think “problems” in a distinct and different way from that ordinarily available in policy studies and political analysis more generally’ (2012: 38). It means acknowledging that the starting point in analysing teen pregnancy is not to ‘solve the problem’, but rather, question discursive constructions that are grounded in a specific political and governance context. This article explores at length the tensions created by the current policy approach to ‘solve problems’ and the space of possibilities for transformation that a WPR approach opens up in an effort to understand better the relevance of the theoretical space underpinning the WPR approach for policymaking and research. Bletsas argues that the WPR approach revolves around the idea that political debates stand in a tension between solving problems either practically or politically (2012: 38). The dynamic of many social policy debates is further divided along the line of structure/agency, seeking to determine whether the ‘problem’ derives from structural causes or is the consequence of individual actions. A WPR approach allows progression beyond this rather sterile debate. Its strength resides in the possibility to explore the relationship between governance and a specific ‘problem’.

Application of the WPR approach to the ‘problem’ of teenage pregnancy

What’s the problem represented to be in policies?

All policies require the work of problematisation. The starting point of analysis is recognising that problems are not independently formed, but actively constituted and shaped. As a result, all policies are made of implicit representations of what the problem is represented to be. The problem representations analysed in this article predominantly come from the representations translated through the main primary laws and policies governing teenage pregnancy. This assemblage best reveals the broader logic behind the treatment of teenage pregnancy as a policy issue that is located at the intersection of several policies.

In the French context, teenage pregnancy ¹ usually refers to women’s pregnancy under the age of twenty and thus does not differentiate between women who have reached the legal age and those who have not ² . The lack of clarity surrounding the term is further exacerbated by the absence of legal clarity regarding the age for sexual consent for minors ³ . Government policies also seem to problematise teenage pregnancy as only unwanted (Le Den, 2012). Le Den (2012) highlights that the most widely used manner to report teenage pregnancy and abortion is the number and not the rate, thus emphasising numbers rather than their proportion related to the total number, which is marginal. Definitions of ‘teenagers’ also vary among studies. Overall, despite statistical indicators reporting the decreasing number of teenage pregnancies, the level of preoccupation by public authorities remains high in France (Le Den, 2012).

On a national scale, the issue has typically attracted neither much research nor intersectional considerations. Some studies have examined the ‘problem’ in relation to the development and legalisation of contraception, abortion, sexual education and changes in family models and relationships (Daguerre, 2010; Le Guen and Bajos, 2014). These studies, however, do not challenge the rationality of the state linking teen pregnancy to individual action (whether focusing on sex education, contraception or preventive actions), nor do they interrogate the governance structures as powerful means to construct the ‘problem’.

Contrary to the US, which favours a restrictive view on sexuality as a way to decrease the number of teen pregnancies, continental Europe and Scandinavia have progressively adopted a more permissive position, accentuating the guidance and support of teenagers with regard to sexuality (Daguerre and Nativel, 2004). Contraception was legalised in 1967 in France through the Neuwirth Law, which repealed a law voted in 1920 that explicitly prohibited the sale, distribution, and advertising of contraceptive devices and abortion. The 1967 law initially contained a section requiring parental consent to use a contraceptive, which was removed in 1974 on the condition that services are obtained at a government-sponsored centre (Le Guen and Bajos, 2014), for parental consent was rapidly perceived as a barrier for teenage women in accessing contraception. Likewise, the Veil law of 1975 legalised abortion and originally set the condition of parental consent. These laws were reformed in 2001 by the 2000 Aubry bill, modifying the parental consent for minors – now requiring any adult’s consent to obtain an abortion – and ensuring free and confidential access to contraception for minors (Allwood and Wadia, 2009). These changes conveyed the idea that the sexuality of youth has an intrinsic social dimension (Le Guen and Bajos, 2014). The government also adopted several measures to improve access to contraception for juveniles, such as the decree n°2012-883, which facilitates the renewal of oral contraception, and the decrees n°2012-910 and n°2016-41, which posit conditions that enable women to access emergency contraception free of charge.

In parallel, sex education policies developed gradually. A 1973 memorandum about sex education and information (No. 73-299) emphasises ‘instilling responsibility’ in teenagers (Ministère de l’Éducation Nationale, 1973). However, the sexuality of young people did not spark public debates in France until 1974, when the legal age of civil majority was lowered from twenty-one to eighteen and sex education and information started to become part of the school curriculum (Le Guen and Bajos, 2014). National trainings for teachers that have been in place since 1995 aim at reducing risky sexual behaviours. Sex education notably aims to prevent and reduce risks related to teen pregnancy (‘grossesses précoces’), sexually transmitted diseases and HIV/AIDS (Jasmin, 2007). The Act of July 4, 2001 greatly consolidated the role of schools in terms of sexual education and guidance about contraception and risks derived from sexual behaviours. It established mandatory sex education that spans elementary and secondary education and pointed to the importance of the state’s action in the ‘transmission of knowledge and in the development of responsible attitudes toward sexuality’ (Le Guen and Bajos, 2014: 299).

Preventive actions regarding unwanted pregnancy and abortion access represent a third important facet of policies governing teenage pregnancy. The first contraception campaign in 1982 targeted women, emphasising their right to contraception. In 1992 and 2000, the campaigns aimed to remove the taboos in discussing adolescent sexuality (Le Guen and Bajos, 2014). Whereas most preventive programs explicitly target girls, one media preventive campaign launched in 2009, ‘Faut-il que les hommes tombent enceintes pour que la contraception nous concerne tous⁴ ?’ (IGAS, 2009), was found to extend the question to boys and men, presenting the question as a ‘couple issue’ (IGAS, 2009: 39). The border between sexuality and relationship remains, however, grounded in heteronormativity and rigid social constructions (Jasmin, 2007).

In short, four problem representations emerge: (1) teenage pregnancy is the consequence of a lack of knowledge about or access to contraception; (2) teenage pregnancy is the consequence of a failure of preventive sexual education or weak implementation; (3) teenage pregnancy is primarily conceived as an individual’s ‘problem’; (4) it can be prevented through the use of contraception and access to information. By emphasising on individual behaviour, most policy accounts of teenage pregnancy fail to acknowledge gender roles and contingent relations underpinning the representations as well as structural power imbalances shaping sexuality and identities.

To put it concisely, teenage pregnancy is represented as being an individual’s ‘problem’ related to ‘inappropriate’ or ‘risky’ sexual behaviour. Policies and laws are conferred a ‘solving’ role in response to the constructed problem. State intervention is thus meant to affect individual behaviour. The next section explores the knowledge and practices (ways of thinking and doing) that shaped the issue as a policy problem.

What discursive practices have shaped the problem representations?

In policy work, the manipulation of problems is closely related to modes of governance. Analysing the knowledge formation underpinning the problem representations exemplifies how the governance of the question is embedded within wider configurations of governing practices. The relationship between teenage pregnancy and the government is never ‘neutral, natural or apolitical’ (Bletsas, 2012: 53). The problematisation of teenage pregnancy is contingent on constructions of the reality at the discursive level (Monk, 1998). Such a perspective draws attention to the dominance of certain knowledge that underpin public policies and the strategic place of ‘experts’ within the debate (Bacchi, 2009). Legitimacy of knowledge derives from the locus of authority at a certain time and place.

Neoconservative discourse is best described by its attachment to morality (Monk, 1998). It is influenced by institutions such as the Christian Church and some right-leaning political parties. This discourse often draws strict contours of what legitimate sexuality is and constructs teenage pregnancy as a symptom of decadent sexual morality (Monk, 1998). Beyond judgments related to individual lifestyle choices, teenage pregnancy is also conceived as a disruption to the institution of marriage and the traditional model of family (Monk, 1998). Although the neoconservative discourse still holds authority among certain segments of the population (such as Catholic associations and right-leaning parliamentarians) as exemplified by the regular polemics regarding sexual education in schools surfacing from public discussions (Gallot and Pasquier, 2018), its overall influence has waned.

In contrast to neoconservatism, neoliberal discourse is much more concerned with individualistic and economic considerations than morality. Based on rational calculations, neoliberal discourse does not strictly condemn sexuality, abortion or contraception as long as it does not increase dependency on the state (Monk, 1998). In France, rationalisation of the phenomenon through numbers has led to denouncing the negative dimensions of teen pregnancy such as educational failure, poverty and financial dependence on welfare (Le Den, 2012).

Studies have documented the authority that the scientific and medical expertise has on health and social policy (Loriol, 2002; IGAS, 2009). Until the mid-1980s, the dominant discourse in France insisted on exaggerating the medical risks associated with early pregnancy, pointing to higher rates of premature births and higher mortality rates. These statistics were soon discredited by comparative research that found no clear correlation between age and pregnancy risks for women of equal socio-economic status (Le Van, 2006). Ironically, these conclusions were not seized upon as incentives to develop further research taking into account the intersections of class, age and gender, but instead shifted the focus onto pregnancy and its medical risks among women aged forty-plus (Le Van, 2006).

More recent insights from social psychology have shifted the grounds to problematise teen pregnancy, though the issue remains framed as a policy ‘problem’. As scientific-oriented research remains the privileged method of ‘knowing’, the chief discourse today is centred on the socioeconomic difficulties arising from early motherhood such as economic instability, precariousness, gaps in educational development, or dramatic repercussions for the child (Le Van, 2006).

As presented above, discourses surrounding teenage pregnancy have evolved and coexist with each other with varying degrees of salience. Of critical importance is the fluidity along which grounds of problematisation have succeeded or superimposed on each other with almost no respite for people to think differently about the issue. It highlights how the concept of ‘teen pregnancy’ has been substantially evolving, thereby wiping away the illusion of ‘problems’ as fixed entities. Most obviously, if the underlying discourse and knowledge have changed and reshaped the representation of the term, the status of teenage pregnancy as a ‘problem’ has remained stable and unquestioned. Essentially, the right of teenage girls and boys to have a sex life is not questioned or re-questioned by the State. Abortion, contraception and youth sexuality are no longer loathed by government policy, denoting a shift away from a morality to concerns over ‘autonomy’ and, implicitly, dependency upon society and the state. As current discourses emphasise the multiplicity of problems generated by early pregnancy and are based on the assumptions that the teen mother is unable to respond to her own economic and social needs as well as those of her child (Le Van, 2006), there are long-term costs for the society and the state (Le Den, 2012). Furthermore, abortion continues to be considered as a deviant behaviour or a ‘trauma’ for young girls, and contraception is presented as a mean to reduce abortion rate (Pavard, 2012). Consequently, the only way to solve the ‘problem’ of teenage pregnancy is to make policies encouraging contraceptive uses among teenage girls.

While the focus has gradually shifted from sexuality to motherhood as a ground for problematisation–and this is not to say that girls’ sexuality has ceased to be of concern–the policy response remains focused on the ‘individual’ level, emphasising individual behaviour, characteristics and responsibility rather than structural elements. In governmentality, individuals are created as ‘the problem’ with emphasis on their private responsibility for health, economic success, educational achievement or contraception (Bacchi, 2015a). Governing practices that produce individuals and shape their lives are completely invisibilised and left out of the political discourses and representations. The gradual deployment of an assembled policy response to teen pregnancy is warranted based on the current prevailing knowledge and discourse, which are contingent on a particular place, time and governance mentality. The analysis demonstrates the links between neoliberal and scientific discourses, governmentality and the representations of teen pregnancy.

This perspective illuminates the production and contingence of the representations of problems and allows us to move beyond the longstanding debate of whether teenage pregnancy has its roots deep in individual failings or, conversely, in structural deficits, as it has no independent existence. The WPR lifts us to another level of debate by shedding light on a space largely made of political questions and power relations. Hegemonic and authoritative knowledge, currently defined by their scientific dimension, are shaping the problem representation of teen pregnancy. The main preoccupation for the analysis shifts from identifying the ‘cause of the problem’, or the means available to ‘solve’ it, to the relationship between governance mechanisms and ‘problems’.

How has this problem come about?

The emergence of teenage pregnancy as a governmental concern is perhaps less the effect of a fundamental change in the substance of the phenomenon than it is a reflection of a shift in governing logic. A genealogy in the Foucauldian sense serves as a method to question and challenge what is often perceived as natural, revealing the power of norms (Foucault, 1971). Governmentality in the legacy of Foucault encompasses two distinct components. First, it refers to rationalities through which governance is made possible. Second, it describes ‘the form of political organisation in contemporary western societies that is both ‘totalising’ and ‘individualising’’ (Bacchi, 2015b: 131). I illustrate my arguments by drawing on two specific segments of history that epitomise the contingency of categories and ‘problems’. The first section explores the historical tensions underlying the concept of legitimate motherhood as shaped by governmental interest, using the lens of single motherhood as a category of analysis. The relative invisibility of teenage mothers as an independent political entity threatens the fixity of a category that appears unwavering today. In the second section, the focus is on the political emanation of ‘teenagers’ as a group and category of analysis and the medicalisation of sexuality, which are closely tied to an alteration of governing logic and concerns, substantially shaping the path along which teenage pregnancy became a policy ‘problem’. These sections expose that ‘teenage pregnancy’ exists in relation to other problems such as teenage motherhood and adolescent sexuality, with emphasis on hegemonic discourses and knowledge.

Motherhood and the boundary of the ‘problem’

Looking back at history, it seems that the notion of teenage pregnancy entered the mainstream at a very specific period of time in industrialised countries. In the first half of the 20th century, non-marital pregnancy was in fact a more visible ‘problem’ (Le Den, 2014). The progressive dissolution of the nuclear family model compounded with political changes led to the emergence and increasing use of the concept of ‘teenage pregnancy’ as a categorising term in the early 1970s in France (Le Den, 2014). In consideration of this conceptual evolution, it is relevant to dissect the structures that facilitated and legitimised the problematisation of teenage pregnancy. Teenage pregnancy is closely linked to motherhood, the latter being the visible and long-lasting part of the ‘problem’. The transformation of the problem representations regarding ‘motherhood’ offers the possibility for a transformation of the problem representations regarding ‘teenage pregnancy’. The following analysis of the evolution of single mothers – and, in particular, the evolution of their legitimacy, role and acceptance within French society – demonstrates that rather than ‘solving’ problems, policy constructs them.

In the late nineteenth century, norms pertaining to sexuality and motherhood were firmly cemented into French society by physicians. Legitimate motherhood was constructed along a biological norm (physical maturity) and a social norm (legitimate union) (Le Den, 2014). Single mothers therefore bore the social stigma attached to the transgression of social norms. At the beginning of the twentieth century, the emergence of the Welfare State and demographic concerns incrementally incited medical authority (physicians) to reconsider the rigid boundaries of legitimate motherhood, leading to a temporary shift in the story told about single mothers (Le Den, 2014).

In the pre-war period, France developed many strategies to ensure that women would be able to fulfil their role both as mothers and employees in the capitalist workforce with the goal of increasing the birth rate and decreasing the infant mortality rate (Offen, 2018). The laws of July 17 and 30, 1913 establishing an eight-week maternity leave for female employees with salaried status (albeit without full compensation) were clearly designed in a natalist rationale (Odul-Asorey, 2013). During this period, the experts in France were the demographers and the obstetricians (Jenson, 1986). Although the focus at the time was on children, this approach indirectly addressed motherhood and the strategical interest to invest in it. As Jenson (1986) notes, maternity leave in France was skilfully framed so that women could stay home for a short time to take care of their newborn in an effort to reduce infant mortality rate, but also ensured their return to the workforce since maximising women’s potential as sites of (re)production was perceived as crucial to building a strong nation. Mechanisms were also put in place for a close surveillance of pregnancies spanning premarital screening, prenatal and postnatal examinations and pregnancy visits (Cahen, 2014). Financial rewards for women declaring their pregnancy before the third month were implemented for sanitary purposes and to limit abortion risks (Cahen, 2014).

After the First World War, demographic concerns intensified and led some individuals, including physicians, to start a movement towards the popular valuation of motherhood, claiming that childbirth is a duty to all women regardless of relationship status (Le Den, 2014; Rivière, 2015). Jane Misme, a moderate feminist leader, writes that during this period, single mothers enjoyed an ‘unprecedented care’ (Rivière, 2015: §27). Fear of population decline led to the implementation of a law in 1920 forbidding contraception and abortion. Their use was subject to fines and imprisonment (Olszynko-Gryn and Rusterholz, 2019: 125).

In the midst of the Second World War, abortion became defined as a social crime by law in 1942, and the maximum penalty increased to capital punishment (in special cases). Abortion was essentially perceived as a crime against the nation. The death penalty for abortion was dropped in 1944 (Olivier, 2002). Maternity leave was also progressively revised. In 1945, it was extended to twelve weeks for women occupying both salaried and non-salaried employments and compensated at 50 per cent (ordinance of 1945). In 1980, maternity leave was extended to sixteen weeks compensated at 100 per cent (law of July 17, 1980), which remains the current maternity leave regulation (Rixain, 2018).

The increasing involvement of the state in private aspects of reproductive life did not shift the negative view of teenage pregnancy/motherhood overnight (Union Nationale des Caisses d’Allocations Familiales, 1968). Although admittedly cursory, the analysis of this historical moment still sheds light on the unsettled boundary of exclusion towards certain mothers and the shifting contours of what acceptable motherhood is. Both can be redefined along changing interests and the maturation of political projects. At the risk of overgeneralising, it is interesting to observe how the boundary of the ‘problem’ (still considered a problem) has been deviated: from a dichotomy based on matrimonial status to one defined by age regardless of relationship status today (Lefaucheur, 2019). Value acceptability, or the compatibility of a certain idea with the national culture or ideology, shifted because a novel problem in the view of policymakers had come to existence (Marshall, 1997).

The emergence of teenagers as subjects

In Western societies, the democratisation of the concept of ‘teenagers’, previously reserved to the sons of the bourgeois elites, materialised in the post-war era as a symptom of modernity to encompass young people generally, beyond class and gender divisions (Le Den, 2014). While no definite age category had been associated with the notion of teenagers, the concept for girls was closely tied to women’s entry into fertile life (Le Den, 2014). The emergence of ‘teenagers’ as a category allowed studies and research on teenagers to flourish, yet the focus remained narrowly framed by suspicion towards their perceived irresponsibility and the belief that teenagers required close guidance on the path to adulthood (Le Den, 2014). Notably, the guidance of girls was based on the premise that adolescence was a period of transition and education towards the roles of mother and wife that they were later expected to fulfil. Disparities in the treatment of teenage boys and girls grew along the logic of transition to adulthood and the construction of distinct gender roles became essentialised in formal and informal practices (Le Den, 2014). Chastity until marriage constituted the basis of sexual education and morality for teenagers, particularly for girls, until the 1960s. Beyond the moral dimension, chastity was further supported by the medical field (becoming a ‘technical issue’), whose goals were twofold. First, chastity was believed to prevent illegitimate teenage pregnancy or pregnancy out of wedlock; second, it served as a means to combat the proliferation of venereal disease, which was a major concern in the post-war era (Le Den, 2014). Sexual norms and the expectation of chastity held a double standard nurtured by the idea that male and female sexuality substantially differed in terms of desire and sexual needs (Le Den, 2014). Female sexuality was reduced to invisibility and chastity; and was believed to be an effortless and natural ideal for girls given the nihility of their desire, presumed at least until marriage (Le Den, 2014).

With respect to legitimate knowledge, the sexual norms generated in the post-war era were supported by scientific and medical concerns despite the scarcity of empirical evidence to back up the claims. Medical authority was consolidated through the salient representation of physicians in the parliamentary arena. Beyond being excessively medicalised, the hegemonic expertise surrounding teenage girls was also almost exclusively masculine (Le Den, 2014: 47–48). Above all, the parallel emergence of the broad categorisation of teenagers compounded with normative research and the obsessive control over girls’ sexuality and motherhood conceived within a patriarchal rationality provides some evidence of the conditions that enabled the phenomenon to grow as a policy problem.

The problem representations identified in the first question find themselves destabilised because the analysis overtly shows that problems are shaped by governmental logic over time, and therefore do not possess an intrinsic rationale that could be dismantled or ‘solved’ by policies. The production of teenage pregnancy as a ‘problem’ is a result of both ways of thinking and governing practices.

Towards transformative perspectives on ‘problems’ and feminist policymaking

Rethinking the political contingency of ‘problems’ through a feminist lens

A WPR approach constructs a space where it becomes evident that realities, knowledge and problems are intrinsically political and remain in eternal flux. These two conclusions have major ramifications for contemporary scholarship in general, and feminist scholarship in particular, captured in this article through the concepts of self-reflexivity and disruption.

The WPR approach reveals a space where knowledge is in essence political and thereby calls for greater self-reflexivity (Bletsas, 2012; Goodwin, 2012); this being true for policymakers, policy analysts and researchers. This acknowledgement compels us to recognise that trying to research or study any issue beyond political or personal bias is a mere illusion, but this idea is hardly novel in critical research. Bletsas cogently articulates it as she writes that ‘in refusing to treat research as simply neutral and descriptive, post-structuralism in general, and the “WPR” approach in particular, offer the possibility of research as a political engagement’ (2012: 48). For feminist research, seeing problems as entities that need to be politically re-appropriated or intervened with is certainly a promising perspective. A reflexive move implies an acknowledgment that the knowledge we produce cannot exist independently from relations of power. All feminist political engagement is based on particular premises, deep-seated assumptions and representations that all need to be interrogated under the same light as the hegemonic discourse underlying policies. Moreover, no feminist political engagement is ever without the influence of competing knowledge, resulting from the increasing fragmentation and diversification of knowledge. Knowledge entering the scope of politics can destabilise prevailing knowledge, while simultaneously fuelling the potential displacement or reconstruction of a ‘problem’. Contrasting political knowledge creates indeterminate content. The relationship between knowledge, politics and the issue at hand is not a linear or deterministic one. It means that knowledge contributions, as radical and critical as they are, will potentially be evolving when crossing the path of other knowledge, producing unintended consequences that can have detrimental effects on groups of people. Spill-over effects are possible, meaning that changing the representations of teen pregnancy can result in a shift in the representations of other problems. It is perhaps a risk to endorse for defying marginalised people’s exclusion from knowledge creation and research focus, as well as challenging dominant androcentric norms.

According to Collins (2000), because elite white men historically controlled the structures of knowledge validation, their interests are reflected in the themes, paradigms and epistemologies of mainstream scholarship. It also means that the experiences of marginalised individuals are misrepresented or excluded from what counts as legitimate knowledge. Institutional actors governing teenage pregnancy policies are numerous in France, including the Ministry of Health, the Ministry of National Education, the High Council of Public Health and the French Agency for Public Health. Additional actors, such as l’École des Hautes Études en Santé Publique or the Société Française Pour la Santé Publique, often offer their expertise on public health issues. French expertise in the field of public health relies primarily on scientific advice. Scientific expertise is assessed based on ‘scientific independence’, ‘impartiality’ and ‘competence’ (Houssin, 2013), giving the illusion of apolitical knowledge. There is a pressing need to incorporate and develop pluralistic knowledge and practices that make visible the multiple contexts of lives and experiences, effects of structural inequalities, and the intersection of gender and other inequalities such as race and class, with continuous self-reflection.

Imagining ‘problems’ as fluid certainly threatens the common acceptance that assumes that the substance of the object of study is stable or has linearity. The WPR approach encourages one to rethink the ontology of policymaking: ‘problems’ can perhaps not be qualified by the state of ‘being’; rather, they are entrenched in the process of ‘becoming’. If phenomena underlying problems are endlessly in flux, certainty and finality of not only feminist but any knowledge remain forever out of reach (Bietsas, 2012: 48). If problems will remain forever unaddressed and continuously displaced by new knowledge and discourses, the traditional way of ‘solving’ problems appears futile. It seems that the critical intervention occurs in the underpinning space of the WPR, at the level of representations of problems and the battle to make marginalised and diverse knowledge valuable and visible to political debates. By extension, ‘certainty and finality are not the only goals towards which our analysis should strive […] perhaps we need to evaluate it in terms of its political effects’ (Bletsas, 2012: 48). The process of identification of political effects obliges feminist researchers to reflect on the various oppressive aspects that all feminist engagement with policy necessarily entails. For example, feminism in France is often conceived based on a republican and universalist perspective, masking the standpoint of white and bourgeois women (Bereni, 2009) and significantly constraining the emergence of marginalised discourses. The need to reflect on the effects of our own theoretical and ontological bases, even when those are explicitly critical and intended to be inclusive, is urgent.

(Un)(re)making ‘problems’: dismantling political rationalities

Critical directions to reconstruct problems from a transformative feminist perspective can be identified. First, all critical research should be treated as a political engagement aimed at the continuous displacement and transformation of ‘problems’. Second, critical research should reflect on its political effects. In this perspective, developing and nurturing alternatives that illuminate the messiness of lived realities are powerful strategies to destabilise predominant political rationalities at multiple levels. In regard to teenage pregnancy in France, intersectionality taken as a disruptive concept can provide a starting point to feminist engagement with policymaking.

Intersectionality, developed based on the US context, sheds light on the multiple systems of oppression that not only intersect, but also interact in the (re)production of inequalities (Crenshaw, 1989; Collins, 2000). Taken as a paradigm, intersectionality departs from a content-based specialisation to focus on normative theories and empirical research, which allows the conceptualisation of the structural dimensions of oppression (Collins, 2000; Hancock, 2007; Bilge, 2009). Proposing intersectionality as a framework for research involves assumptions regarding ‘what we think ought to change and hence what we assume or represent the “problem” to be’ (Bacchi, 2011: 36). It suggests that teen pregnancy is not only a gender issue but also relates to class and/or race as contextually defined. Intersectionality provides a tool to deconstruct and better understand the messiness of lived realities and relations of power. A review of qualitative and quantitative studies in the US demonstrates that preexisting characteristics, which act as predispositions to early motherhood, are significantly responsible for poor socioeconomic outcomes regarding teenage pregnancy (Smithbattle, 2018). The same study found that deferring motherhood does not fundamentally improve the lives of girls who are already disadvantaged and suggests that cumulative social and economic exclusion and marginalisation bear more weight in the production of negative outcomes for young mothers regardless of age. Instead of assuming that age in relation to gender is the ‘problem’, research should start interrogating class and other social and economic factors in relation to gender based on an analysis that questions the category of age as a central axis of causality.

One mechanism of social control is the power to classify the social world and attribute categories with defined assumptions that most serve dominant interests. The construction of homogenous notions and categories obscures the diversity of experiences, causes and identities composing an entity, and produces the idea that identities are stable, ahistorical and devoid from relations of power. Intersectionality offers the possibility to redefine categories and complexify the often simplistic equation posited by the political power in place, drawing on embodied narratives while acknowledging the fluidity of categories. Based on interviews with teenage mothers (below twenty) in France, Le Van (1998) shows the great diversity of situations of these women and finds that in many cases pregnancy was deliberately chosen. This study challenges commonly accepted public discourses and knowledge developed by ‘experts’ conceptualising teen pregnancy as only unwanted and focusing excessively on the related difficulties.

For intersectionality to be empirically translatable, the interaction between different categories must be analysed, which also implies that data disaggregated according to gender, race or class are available. In France, social sciences have historically been uncomfortable with categories pertaining to race and ethnicity, making it difficult to construct a more nuanced understanding of lived realities with regard to these specific identity categories (Chapman and Frader, 2004; Simon, 2010). The collection of information related to ethnicity and race has been severely restricted by law since 1978 (Data Protection Act).

Further, if intersectionality constitutes a promising lens not only to reorient future research but also to enhance the reflexivity of the researcher vis-à-vis the prioritisation of one or two categories over the others, we can wonder to what extent such an angle would retrace a new boundary of what acceptable pregnancy/motherhood is. Categories in social sciences are primarily conventions that designate objects of knowledge but also target groups for action and policy (Stone, 1997). They are central to the process of problematising, the configuration of public policy approaches and, perhaps more dramatically, the process of assigning stigmas. Powerful political actors exploit stereotypes of social groups, in particular powerless groups, based on a selective use of facts (Schneider and Ingram, 1997). These constructions impact target groups. Powerless groups (deviants if constructed negatively, or dependents if described positively) have often poor resources to challenge or exploit the way they are described by policymakers. Even if their construction changes, we could imagine that shifts in the representations of a group can lead to the creation of new stigmas or displace stigmas from one group to another, as the construction of target groups is perhaps an inevitable condition of policymaking. Social policy design is almost always based on a particular understanding of the target group it seeks to affect.

In short, the rationality of current policymaking derives from: i) problems are the raison d’être of policymaking, ii) target groups are narrowly constructed based on stereotypes and depending on their political power, and, more importantly, iii) its inability to take into account the intersections of an individual’s identities. Rather, individuals are conceived as part of a social group whose homogeneity, based on chief characteristics, often erases other identity axes and reifies individual experiences. As such, if intersectionality holds the potential to shift the problem representations of teen pregnancy, efforts to do so perhaps come at the cost of creating new categories of exclusion and marginalisation as the logic of policymaking is not so easily dismantled. Without such interventions, however, we are tacitly accepting hegemonic constructions of realities.

Conclusion

This article may stimulate critical scholarship or discourage it from further engagement with mainstream policymaking because of the difficulty to substantially transform policy rationality. It sought to challenge the idea that problems are to be solved by departing from the structure/agency debate. To achieve this goal, it focused on two elements: revealing the political essence of knowledge constituting ‘problems’ and the flux underlying phenomena captured through problematisation. This was accomplished through the application of questions 1, 2 and 3 of the WPR framework to the case of teenage pregnancy in the French context.

Teenage pregnancy in France as a ‘policy problem’ remains largely shaped by scientific knowledge and discourse, rather than the voices of those embodying the experience. Social outcomes have been exclusively defined along material lines with no qualitative interest regarding the lived experiences of teenage mothers in terms of caring, empowerment or emotions (Bradley, 2018). It is also focused on individual actions, rather than the structural conditions that constructed the problem in the first place. The implicit normative values underlying problem definition and rationality of policymaking is premised on a normalised life course trajectory. In particular, the social organisation of individuals (and specifically women) seems to be modelled on the normalised trajectory of privileged groups of men, typically divided into three steps: young people have dedicated themselves to education (1) in order to have access to a job (2) in a highly competitive labour market, and when their career is sufficiently developed they can think to have a family (3). Structures must be questioned as they often reflect normative values attached to a non-neutral referential, de facto disadvantaging people that do not fit into this particular model or identity.

By interrogating the problem representations and the discursive practices that gave rise to the policy problem of teen pregnancy, rationalities behind policymaking are made visible. From demographic concerns to neoliberal preoccupations, teen pregnancy remains grounded at the interstice between other policy problems. Shifts in a policy issue entangled in the policy problem of teen pregnancy can transform its representation. Conversely, shifts in the representation of teen pregnancy can shift other problem representations, potentially leading to the creation of new ‘problems’. The current rationality of the (French) policymaking system presents many challenges to engaging in critical and transformative approaches to the ‘problem’. The difficulty of collecting data based on race or ethnicity and the centrality of target groups in policymaking considerably constrains and guides possibilities for feminist research.

By pointing to some of the political rationalities and techniques of governance that have shaped teen pregnancy as a ‘policy problem’, this article articulated some of the implications for contemporary scholarship, and particularly feminist scholarship. The first important conclusion is the need for any policy scholarship to engage in self-reflexivity and carefully analyse its own representations as well as its political effects. The second chief conclusion is the need for a focus on movement and change towards which feminist scholarship should strive through the strategy of disruption. In this perspective, it is possible to daringly argue that a WPR approach contributes to redefining the very nature of ‘problems’: they are not to be solved but transformed into plural possibilities reflecting the vast array of everyday lived experiences and realities. Changing a problem’s representations should not be a goal in itself. It is only a means to dismantle inequalities. Through refusing to readily draw conclusions, self-reflexive feminist research engaged in the disruption of ‘problems’ can provide a point of resistance to the hegemonic constructions of realities and positivist epistemology, and can contribute to altering, even slightly, policymaking rationality.