INTRODUCTION
The rich discussion concerning religious education usually focuses on the content and scope of the parental right to direct the upbringing of their children and the duty the state has regarding the education of religious children. This lively debate is driven primarily by the concern that schools that serve religious communities, and especially anti-liberal religious minorities, will fail to supply children with the education they need in order to become autonomous, self-sufficient, and tolerant members of society in a modern liberal state. Framed this way, religious schools that provide adequate secular education and endorse liberal values seem to raise no concerns, and should be allowed to operate uninterrupted, and possibly even be supported by the liberal state.
This article disputes this stance and argues that religious schools that provide satisfactory secular education, prepare their students for higher education and instill democratic civic values may create a separate category of worries â related to equality â that justify state intervention.
Empirical evidence shows that in recent years, religious schools in many democratic countries are gradually losing their religious distinction and are increasingly becoming a means of gaining educational advantage. As a result, religious schools are attracting high-achieving students, some of whom are not members of the religious community that established the schools. These advantaged students leave public schools, in a process called âcreaming.â Creaming harms public schools and the students that remain in them because studentsâ achievements depend, among other things, on the educational abilities of their peers. Creaming also results in the student bodies of public schools being composed of a higher proportion of students from disadvantaged backgrounds, students who are not native language speakers and disabled students, all of whom are more expensive to educate.
Not all religious schools cause this effect. There is, I argue, an inverse relationship between the strength of the religious character of the school and the extent to which it creates these processes. The more robust the religious education, pervading all classes, activities, and the schoolâs atmosphere, the less likely the school is to appeal to a broader population of students and thus to infringe upon equality in education. In order to demonstrate this claim, I distinguish between two kinds of religious schools that are located at two ends of a continuum of religious schools.
The first kind of school, which I call non-creaming religious schools, consists of schools that provide robust religious education that differs substantially from that of public schools, in content, values, and atmosphere and often does not fully prepare its students for successful participation in modern society. These religious schools are the archetypical kind of religious schools that philosophers and legal scholars have in mind when engaging in the debates mentioned above concerning civic education. They are highly valued by members of their community, as they offer education that is completely different from that provided by public schools and as a result are sometimes perceived as a threat to the liberal state. Yet while they raise concerns pertaining to their ability to prepare students for participation in the workforce and civic life, they do not create the distributive justice concerns caused by creaming.
What I call creaming religious schools, in contrast, are religious schools that offer quality secular education in addition to their religious education. They grant their students all the credentials and skills that traditional public schools do, while their religious characteristics are relatively mild. These schools, when successful, appeal not only to members of the specific religious community but also to high-achieving students from outside this religious community, who would otherwise attend traditional public schools. Losing advantaged students weakens traditional public schools and causes a decline in their studentâs educational achievements. Because they do not raise the aforementioned liberal concerns, creaming religious schools have, for the most part, evaded philosophical and legal criticism.1 In this article, I argue that when religious schools cream, state intervention is justified in order to prevent the negative effects they have on educational equality.
Creaming religious schools need to be distinguished not only from non-creaming religious schools. Creaming religious schools also differ, in the normative considerations that apply to them, from creaming secular schools (most typically prestigious private schools). Creaming secular schools have negative effects on educational equality, and therefore, some scholars argue that they should be restricted (Swift, 2003). The argument I present concerning creaming religious schools endorses this argument and aims to apply it to the case of creaming religious schools. This application, however, is not straightforward and requires a separate justification. The parental right to religious education justifies distinguishing between creaming religious and secular schools and arguably offers religious schools protection from state intervention that secular schools do not have.
In response, I show why the religious justification for creaming religious schools is especially weak and cannot override the egalitarian challenges. First, only members of the religious community can rightfully espouse the religious justification, and creaming religious schools often have large shares of non-member students. Second, their relative similarity to public schools and the nature of their remaining uniqueness suggest that they are not crucial for realizing the right to religious education. As a result, I conclude that when these schools negatively affect other schools, the religious argument in their favor cannot redeem the inequality they create.
Law has a central and unexpected role in the development and thriving of creaming religious schools. Various legal systems have instituted antidiscrimination rules that do not allow religious schools to consider the applicantâs religion in their admission policies. In order to accommodate students who do not belong to the relevant religious denomination, antidiscrimination rules are usually accompanied by rules making all religious activity at school elective. While this seems, at first glance, an effective way of contending with inequality caused by creaming religious schools, antidiscrimination rules, in fact, exacerbate the problem. They facilitate the entry of non-member students into creaming religious schools, making the choice of religious schools a means to gain educational advantage rather than a way to realize religious convictions. Therefore, antidiscrimination rules should be replaced by other means of regulation that are better suited to contend with the special challenge of creaming.
The chapter proceeds as follows. First, I offer a short description of peer effects and creaming, the educational processes that underlie the arguments in the chapter. Then I present the category of non-creaming religious schools and demonstrate it using the examples of Amish one-room schools in the US and Israeli ultra-orthodox schools. I move on to describe the phenomenon of creaming religious schools and how they become a vehicle for social exclusion. The examples in this section include Catholic schools in France, religious charter schools in the US, and Torani schools in Israel. After completing the descriptive analysis, the chapter moves on to the normative argument according to which creaming religious schools should be restricted. Finally, the chapter discusses lawâs role in aggravating inequality in the case of creaming religious schools and suggests possible legal measures that may be more effective in contending with the challenges that creaming religious schools pose to educational justice.
PEER EFFECTS AND CREAMING
Student composition is an important factor in determining the quality of education offered in any given school. Education is a unique good, in which the outcome depends, among other things, on the identity of the consumers. Studying in a class with bright and motivated peers, who cultivate academic aspirations and contribute their cultural capital, improves student achievement (Orfield & Lee, 2005). Some researchers argue that taken alone the peer group is the most influential factor in educational outcome (Johnson, 2000; Zimmer & Toma, 2000). Peer effects can explain, in part, why studentsâ achievements are higher in private schools that group together high-achieving students than in schools that do not (Lubienski & Thuele Lubienski, 2014). Correspondingly, concentrating children with low abilities in one school causes negative peer effects that decrease their educational achievement.
There is also substantial evidence according to which the academic attainment of students from affluent families is higher, on average, than that of students from low-income families (Kahlenberg, 2001; Rothstein, 2004; Rumberger & Palardy, 2005) as a result of various environmental factors such as adequate nutrition and health care, and the amount and quality of interactions with parents (Ermisch, Jantti, & Smeeding, 2012; Lareau, 2003; Ross & Kena, 2012).
What follows from the correlation between socioeconomic status and educational ability is that grouping together students according to their social class (e.g., in neighborhood schools when neighborhoods are socioeconomically segregated, or by charging high tuition) also entails grouping together students according to their academic ability. Similarly, schools that select their students according to seemingly neutral academic qualifications (or practices such as ability grouping) are likely to result in homogeneous student bodies (Cipriano-Walter, 2016; Greene, 2014; Losen, 1999; Oakes, 1995). This separation further compromises the educational opportunities of children of lower socioeconomic classes. Negative peer effects combine with their family background in curtailing the development of their abilities.
The disadvantages caused by social segregation are even more pervasive. Schools that serve advantaged students attract better teachers, whereas the schools that serve disadvantaged children are usually staffed by less professional teachers (Frankenberg, 2009; Peske & Haycock, 2006); and schools that serve advantaged children also benefit from the involvement of parents, who tend to contribute more resources and time the more advantaged they are (Hickman Wehlburg, Greenwood, & Miller, 1995). As a result, social segregation shortchanges children from disadvantaged background and restricts their access to resources crucial to their success.
Creaming leads to segregation, as it results in concentrating high-achieving and motivated students from advantaged families in prestigious (and often selective) schools. When these children are withdrawn from public schools, the latter are depleted of advantaged children and of the resources they bring with them. Disadvantaged children, who cannot access prestigious schools, remain in these public schools that now offer education of poorer quality.2 The deterioration of the public school induces even more students to leave until eventually only the most disadvantaged students who are unable to leave remain.
Various factors enable schools to cream. First, in order to cream, schools must be attractive to parents, offering advantages such as better test scores; a better track record in graduatesâ placements in college; better facilities; more options for advanced placement (AP) courses or college prep programs; better qualified teachers and lower teacherâstudent ratios; ample educational resources for supplies, field trips, labs, and technology; a safe and pleasant environment; and, importantly, high-ability peers.
Most of the factors that account for schoolsâ positive reputations can be reduced to two main advantages: financial (which can ensure a rich curriculum, qualified teachers, better facilities) and student composition. When a schoolâs prestige is related to test scores and Ivy League placement, the schools are, in part, enjoying the benefits of having advantaged children among their students. Student composition also affects violence rates and the quality of teachers (Peske & Haycock, 2006).
While private schools are the paradigmatic examples of creaming schools, enjoying both financial advantages and privileged student bodies, creaming can also occur in public schools. Public schools that charge fees or have enrollment policies that include exams, essays, or interviews also engender socioeconomic segregation (Finn & Hocket, 2012). Even when formal barriers such as fees and selective enrollment policy do not exist, several mechanisms work together to make prestigious public schools more accessible to children from privileged backgrounds. They may offer curricula that are more attractive to advantaged parents; they may be located in affluent neighborhoods; or the admission process may be complicated. Finally, disadvantaged families might forego the best educational possibilities because they lack the sense of entitlement require...