A. Religious exemptions and equality in education

In the early 1970’s Wisconsin law required that all residents receive an education up to the age of sixteen — or tenth grade in the American education system. In the American system, non-Amish students are required by law to complete high school. This is typically accomplished in one of two ways. Either students graduate from the twelfth grade or they “test out” earlier by taking a graduation equivalency examination, as early as tenth grade. However, common practice for Old Order Amish is for children to finish education at the age of fourteen, or the eighth grade. ^{Footnote 15} The Amish filed a suit against Wisconsin and won in the Supreme Court in a six to one decision. ^{Footnote 16} The majority argued on “free-exercise” grounds that the Amish have a constitutionally protected right to refrain from providing their children more than an eighth grade education.

In his majority opinion Justice Burger defends three claims. First, precedent favors the Amish. Among other instances of accommodation, granting conscientious objector status to Quakers and Mennonites has for many years been regarded as sound constitutional law. Second, the burden that education law imposes on Old Order Amish is too severe. Compliance with Wisconsin education law is a threat to the Amish way of life. In Justice Burger’s words, “the Old Order Amish daily life and religious practice stem from the fact that it is in response to their literal interpretation of the Biblical injunction from the Epistle of Paul to the Romans, ‘be not conformed to this world’.” ^{Footnote 17} A religious practice based on convictions such as this one cannot flourish if the practice must conform to state education policy. Third, Amish educational practices are consistent with the state’s interest in securing access to education for each citizen. This claim is crucial to the argument, because if true, it would establish that the state is able to accommodate Old Order Amish educational practices without compromising on requirements of justice in education, such as fostering skills essential to democratic citizenship.

Burger’s third claim is especially relevant here because he does offer a nuanced judgment about the legitimate aims of state education policy. ^{Footnote 18} For example, Burger points out that Wisconsin law requires all children to receive an education through the age of sixteen. Had Wisconsin law set the age at eighteen, the exemption argument in favor of the Amish would be much weaker. The contrast between two versus four years of additional education is significant, especially when we consider the impact of this difference on a student’s ability to enter higher education. Burger is right to stress this point. However, as I argue shortly, our position on what counts as an adequate education should be sensitive to a number of other factors and when we consider Yoder in light of these, the argument in favor of an exemption for Old Order Amish is considerably weaker.

There are familiar and compelling objections to Yoder. Joel Feinberg raises serious concerns about the impact of this decision on Old Order Amish children’s right to an open future. ^{Footnote 19} An education that ends at age fourteen, in the modern world, is an education with far fewer open doors than a high school or equivalent education. Susan Moller Okin’s primary objection to Yoder is that the Court was insensitive to the effects of religious and cultural pressure on children and thus did not give appropriate weight to the liberal conviction that a member of any group, religious or otherwise, should be recognized as having a right to exit and that this right should be protected in ways that provide the material resources to exercise it. ^{Footnote 20} The approach to religious exemptions that I favor concurs with these objections to Yoder. My position differs, however, in two ways. ^{Footnote 21} First, it is guided by the claim that each citizen is entitled to an adequate education, where adequacy standards are guided by an idea of democratic equality that has implications for both civic equality as well as equality of opportunity. The right to exit and the right to an open future are necessary for democratic equality, yet these other views are less helpful when considering a wider range of conflicts that center on religious freedom and education policy. This will be clear in the discussion on science education, which requires an analysis that looks for ways to reconcile a demand for religious accommodation with adequate science education standards. Second, educational outcomes should be measured in part by a standard of relational equality according to which citizens are able to interact as equals in the political domain and beyond. ^{Footnote 22} For example, educational practices fall short when some lack access to higher education simply because their education fails to provide the opportunity to develop skills essential to succeed beyond the high school level. Though not in principle incompatible with this position on adequacy in education, the ideas of a right to exit and a right to an open future are not well suited for addressing these concerns. This is one reason to favor an approach to education policy that is guided by the idea of democratic equality. ^{Footnote 23}

In a developed affluent state, an adequate education will include basic knowledge of the natural sciences, computer literacy, global affairs and the ability to manage personal finances in an age of credit and personal loans. Anyone deprived of skills that are required to put this knowledge into practice is at a significant disadvantage. ^{Footnote 24} This is a compelling reason to revisit Burger’s claim that an exemption for Old Order Amish can be reconciled to the aim of an adequate education. An eighth grade education might be sufficient if the aim is to enable children to read The Declaration of Independence. Yet if the aim of a state education is to provide children with the skills that will allow them to become software engineers, lawyers, accountants, or viable candidates in a small town mayoral race, then a middle school education falls below the adequacy threshold. By contemporary standards, the educational practices accommodated by Yoder are not acceptable. The motives of those who continue to support Yoder are not always malicious, ^{Footnote 25} yet no citizen should have to confront barriers to future success, whether in higher education or in employment opportunities, that result from not receiving a high school level education.

To defend Yoder in a contemporary context is to put an argument for religious freedom in the service of bad education policy. The effects on an Old Order Amish child whose formal education ends at fourteen are unfair in ways similar to disadvantages that afflict minorities and the poor. The causes are different in each of these cases. Yet the point that matters here is that there is something wrong with our educational practices if by design such practices deprive some students of an adequate education. If we agree that education policies should be designed to equip students with the ability to pursue a higher education if they choose to do so, as well as an ability to enter the labor market with decent prospects for a range of careers, then no one should be deprived of the opportunity to enjoy the benefits of a state-sponsored education.

Justice in education also requires that education policy serve the ends of civic equality. One crucial aim of adequate education standards is maintaining civic equality between citizens. However, we should be just as concerned about the effects of education policy on opportunities to enter the labor market or to pursue a university education. ^{Footnote 26} With respect to Yoder this point is more compelling today than it was in 1972.

B. Religious accommodation and science education ^{Footnote 27}

In 2004, a Pennsylvania school board adopted a policy that required teachers to present the following statement in ninth grade biology classes at Dover High School:

The central legal question here is, does an official state policy that encourages students to consider intelligent design theory (ID) as an alternative theory of origins violate the Establishment Clause of the Constitution? ^{Footnote 29} After reviewing the legal basis for rejecting the Dover policy, I present an approach to religious objections to science education. On this view, liberals should favor accommodation for religious citizens without granting exemptions from respectable standards for science education. The approach I favor shows how to reconcile a reasonable commitment to religious freedom with adequate science education standards.

Judge Jones III, the presiding federal judge in Dover, issued the following ruling: “We hold that the ID policy is unconstitutional pursuant to the Establishment Clause of the First Amendment of the United States Constitution and Art. I, Sect. 3 of the Pennsylvania Constitution.” ^{Footnote 30} Since there is a long history of case law and legal opinions by courts on how to interpret the prohibition on establishing a religion, Jones defends this ruling by invoking some of the most relevant precedents. Two are especially relevant. In Lemon v. Kurtzman, the Court promulgated what has become known as the Lemon Test for religious endorsement. This standard imposes three requirements; namely, state policy must 1) have a “secular purpose,” 2) avoid promoting or disparaging a religious viewpoint, and 3) avoid creating an “excessive entanglement” ^{Footnote 31} between state and religion. In Edwards v. Aguillard, the Supreme Court invalidated a state requirement to teach “creation science” on grounds that teaching creation science violates the Establishment Clause. ^{Footnote 32}

No one should be surprised that an American court would rule that “creation science” should not be taught in public schools. Teaching “creation science,” which purports to offer a theory of origins based in scripture from the Old Testament, amounts to endorsing a religious viewpoint. Defenders of ID claim that ID is not creationism. However, Jones was persuaded that ID is a label invented by creationists after Edwards and other legal cases had affirmed the unconstitutionality of policies requiring that “creation science” be presented as an alternative to evolutionary theory (ET). Jones in effect claimed that proponents of ID were engaged in a strategy of promoting religious faith through deception. After reviewing evidence from experts ^{Footnote 33} that exposed the strategic motives behind the Dover policy, Jones concluded that it was clear beyond any reasonable doubt that “the Board knew that ID is considered a form of creationism.” ^{Footnote 34}

The legal argument is made even stronger in light of the scientific and philosophical considerations, many of which were presented in Kitzmiller. First, methodological naturalism provides sufficient and compelling philosophical reason to oppose accommodating ID in science curricula. Not to be conflated with metaphysical naturalism, methodological naturalism holds that scientific inquiry is limited to the domain of the natural, which is understood to include natural causation and hypotheses that can be tested against empirical evidence. ^{Footnote 35} Methodological naturalism can be endorsed in science education policy both because this is good science and also because state endorsement does not violate the Lemon Test or any other legal principle that requires government to remain neutral toward the religious convictions of its citizens. Second, the scientific evidence for ET is compelling. ^{Footnote 36} There are no alternative scientific research programs that challenge ET in a way that would satisfy proponents of ID.

An optimist might predict that these legal, scientific, and philosophical considerations will eventually have a positive effect on the public’s beliefs about ID and ET. This optimism is unfounded. Kitzmiller comes on the heels of many other cases, none of which had much impact on the wider culture. In addition, by the numbers, and this is as important as any other factor in American politics, the percentage of citizens who oppose ET for religious reasons has not changed over the last hundred years. ^{Footnote 37} This is why anyone who is serious about adequate science education standards must focus on the political debates in addition to the legal, philosophical, and scientific arguments.

There are numerous ways we might attempt to address the political dimensions of the debate over teaching ET. One approach is to insist that science education policy should be determined entirely by the consensus among scientists without any effort to acknowledge critics’ points of view, religious or otherwise. Were the participants who shape science education policy limited to experts, this would be the right strategy. In fact, the verdict in Kitzmiller provides a compelling legal argument for this. However, since education policy is shaped in part by local factors, such as an elected school board composed of members who are generally not experts, pressure against promoting ET as good science in public high schools will remain for as long as large numbers of the American public are skeptical of ET. As Kenneth Miller writes, “[m]ost Americans seem to think that evolution is nonsense. . . . In a June 2005 Harris poll, only 38 percent of Americans thought that human beings had developed from an earlier species, while fully 54 percent thought that we had not.” ^{Footnote 38} For this reason, public intellectuals and experts need to present arguments in favor of good science education in ways that address ordinary citizens and not just experts. This practical consideration is lost on some of the more bombastic critics of ID. ^{Footnote 39} It is a serious political mistake to combat efforts by proponents of ID to change science curricula for the worse with heavy-handed rhetoric and ridicule. That strategy does not work.

Questions about strategy are essential to the politics behind the ID-ET debate. As a rule, philosophers are less interested in these issues than they are with methodological naturalism, political morality, and the aims of education in a democratic society. This is a mistake. To see why, consider the familiar contours of public discourse in this context. Policy debates over acid rain, smoking and cancer, climate change and many others are settled not by scientific experts, but rather through protracted conflicts among parties whose values often conflict with science-driven public policy. ^{Footnote 40} Political philosophers cannot ignore this if they want their contributions to be relevant to real politics.

There are a number of philosophical issues concerning reasons and values often overlooked by critics of the ID movement. Philosophers of science are in the best position to address this issue, ^{Footnote 41} yet political philosophers cannot ignore them if they hope to make proposals that are instructive for the policy debates on science education. Outspoken critics of ID sometimes claim that modest convictions about rationality and evidence are sufficient to dissuade anyone from harboring skepticism about ET. This claim is too simplistic. Instead, those who support ET should focus on how to bring about reasonable political solutions when a population is divided, as is the case with ET in the American context.

The conflict over teaching ET in public schools is not simply a matter of rational people arguing with irrational people. Some proponents of ET respond to critics of ET in ways that belittle and browbeat them. This strategy rests on a philosophically dubious view about rationality: it takes for granted a controversial view about the relationship between values and rationality. Many citizens affirm values, religious or otherwise, that are not going to be revised when scientific data is perceived as conflicting with these values. To reject the results of modern science is not by itself evidence of irrationality. Some philosophers of science who accept the authority of modern science acknowledge that conflicts between values and scientific evidence are not equivalent to conflicts between people who are rational and people who are not. ^{Footnote 42} Citizens of many persuasions have convictions that guide their judgments about what matters, including the authority of science. Moreover, we can defend liberal values without having to settle philosophical questions about the relationship between the authority of science and rationality.

It is also a political mistake to admonish critics of ET by calling them irrational. This strategy will not change anyone’s mind about teaching ET in public schools. If the end is good science education policy, we need to promote strategies that are effective as means to that end. Some will object to this approach to the ID-ET debate on grounds that many of the strategists who advised the Dover school board were unscrupulous in their effort to promote a religious viewpoint in a public school. Jones rightly rebuked these supporters of the Dover policy on this ground. ^{Footnote 43} Yet many citizens affirm ID or its equivalent for nonstrategic reasons, and these citizens are the source of much of the political support for policies like the Dover ID policy. Liberals should advocate a respectful engagement with such citizens in their effort to promote good science education.

Some recent work in moral psychology looks at framing effects and the ways that values introduce bias in attitudes about complicated scientific questions. Jonathan Haidt claims that we should engage the so-called “culture wars” in American politics by trying to understand how shared values, such as religious freedom, are construed in different ways by liberal and conservative citizens. ^{Footnote 44} Daniel Kahneman’s research shows that quite often reasons are not the triggers of judgment. ^{Footnote 45} Framing strategies informed by values often result in judgments that do not hold up against the best scientific theories. The relevance of this experimental work on rationality for understanding the ID-ET debate is clear. If we present the scientific evidence in favor of ET in ways that do not needlessly antagonize religious skeptics, a consensus on sound science education policy is more likely.

Some science skeptics are easy prey to those who want to enlist their support for a political agenda. This is true across a range of topics including, climate science, diet and diabetes, and until recently, the harmful effects of smoking. ^{Footnote 46} In America, religious and political values often converge to create what we might call a perfect storm for controversies that are unfortunate, when the science is clear. For this reason, we need to view the ET-ID debate as not just about science but also citizens’ beliefs about religious freedom, the role of parental authority in education, and the local rule approach to education policy that has been generally favored by the courts. If we are mindful of these factors then we are in a better position to consider effective and responsible science education policies. Those who claim the debate is between rational and irrational disputants often favor good science education policies, but their tactics are guaranteed to alienate the very audience whose political support is essential to move beyond the century old debate that persists in an ossified state.

Many critics of ET are not advocating a position that is comparable to other kinds of illiberal positions, for instance, official policies that discriminate against minorities. ^{Footnote 47} Secondly, there are a number of practical remedies that could be taken without institutionalizing the worrisome inequalities that are a result of the Yoder decision. One option would be to develop civics or current events courses in which students are encouraged to debate a range of topics. Another option, which I discuss further shortly, is school choice programs in which parents are able to use public monies to send their children to schools that have a religious orientation.

A. Preferential treatment for religious values

Some liberals do not espouse the mainstream view. For instance, Brian Leiter rejects the idea that religious values should be given the kind of preferential legal status they enjoy under American law. ^{Footnote 50} Brian Barry defends a considerably less accommodating approach to demands for religious exemptions compared to the mainstream view. ^{Footnote 51} Leiter’s work is especially relevant to my position on religious freedom and education policy, so that will be the focus in this section.

In Why Tolerate Religion? Leiter presents a convincing argument that states should not play favorites toward religious citizens when it comes to exemptions. Leiter also argues correctly that granting exemptions to religious or nonreligious citizens should be permitted only when burdens are not unreasonably shifted onto other citizens. The argument for these claims rests in part on an interesting philosophical, as opposed to legal, investigation into the nature of religious convictions.

If there is any warrant for giving preferential treatment to religious values when they conflict with state policy, we need a clear answer to the question “What is so special about religious values such that these and only these value merit preferential treatment when citizens demand exemptions or special accommodations?” ^{Footnote 52} Leiter suggests that if religion is special, this is because religious convictions have at least one of the following four features: 1) they make categorical demands, ^{Footnote 53} 2) they express beliefs that are insulated from evidence, ^{Footnote 54} 3) they affirm beliefs about the ultimate nature of reality, ^{Footnote 55} and 4) they make sense of “basic existential facts about human life, such as suffering and death.” ^{Footnote 56} For some of these features, e.g., “make categorical demands,” there are ordinary secular moral doctrines that qualify. Therefore, if exemptions are granted to citizens on this ground, fairness and equal treatment require not giving preferential treatment to religious over nonreligious values. For others features, such as “express beliefs that are insulated from evidence,” any such doctrine, religious or otherwise, does not merit preferential treatment, especially in the science classroom.

Some readers will no doubt construe Leiter’s choice of framing for some of these points as a jab against people who affirm religious values. ^{Footnote 57} For example, regarding “insulation from evidence,” Leiter writes, “[r]eligious beliefs, in virtue of being based on ‘faith’, are insulated from ordinary standards of evidence and rational justification . . . ” ^{Footnote 58} This claim can be presented in a more or less provocative way. My own view is that many values are insulated from ordinary standards of evidence. Moral, religious, and aesthetic values are paradigm cases. The moral requirement to tolerate those who affirm such values is not undermined by this fact. Yet liberal toleration does not require giving preferential treatment to many kinds values in debates about the authority of law and public policy, so in this respect Leiter’s point is cogent to religious freedom and education policy.

Debates over preferential treatment for religious values will undoubtedly continue. Some citizens agree with Augustine who says, “faith is defined as ‘the evidence of things unseen’” ^{Footnote 59} and will claim that this is a feature of religious convictions that convictions about nonreligious values lack. On this view, a citizen who espouses sincere convictions about French existentialist philosophers will not deserve the same accommodations as a Christian or a Muslim because the French existentialist does not have values that are supported by this special kind of faith. However, a liberal state cannot endorse this position without violating a fair commitment to liberal neutrality. Nor should the liberal state as a general matter prohibit endorsement of this conception of faith, except when it comes to official state policies, such as science education standards. Regardless of whether faith exhibits this special property, a liberal state should tolerate ^{Footnote 60} those who claim to have it. Yet those who claim to enjoy knowledge by faith do not have a moral entitlement to special exemptions that nonbelievers do not. In the context of science education, faith in Augustine’s sense has no place when it is offered as a source of scientific evidence that competes with ET. ^{Footnote 61}

Liberals should of course favor freedom of expression and free exercise for any such beliefs, provided the standard liberal provisos on the limits of expression are honored. On this point the law is rightfully on the side of the scientists, political philosophers, and others who insist that science education policy should not be adapted to meet demands, religious or otherwise, when these demands require promoting claims that violate methodological naturalism. To do so is to favor education policies that fall below reasonable standards, while at the same time enlisting government on the side of science skeptics who lack credibility as scientists. Nevertheless, as discussed shortly, there may be feasible non-rotten compromises that both uphold sound science education standards while at the same time accommodating citizens for their values, when these values put them at odds with ET. To accommodate is not the same as to exempt, however, so the policy options that I propose will not give science skeptics everything they want.

B. School choice, religious freedom, and science education

There are at least three reasons liberals should be open to school choice programs. First, a school choice scheme might be better for students. When the alternative is home-schooling, which in America it often is, subsidized charter schools that are required to maintain adequate science standards should lead to better educational outcomes for students who would otherwise be homeschooled. This of course depends on a number of factors, including the teaching competence of parents and the science standards in public schools. Other things equal, we can reasonably expect that a high school teacher with an undergraduate degree in biology will be more informed about science than most citizens. Second, an experiment with school choice aimed at addressing the stalemate between religious science skeptics and their critics does not have to compromise religious freedom or science standards. The prospect of even nonsectarian religion classes could be an attractive alternative to home-schooling for some parents. A third reason is that parents and political groups quite understandably want to create options for children who attend schools whose educational outcomes are not adequate. Each of these is a prima facie reason to favor school choice programs. I provide evidence for these reasons in what follows.

Political factors limit options in education policy. Some of these include: a lack of resources devoted to studying religious traditions at the primary school level, for instance, in comparative religion courses; a public sphere that has been corrupted in part by science skepticism; ^{Footnote 62} a willingness by political leaders to pander to science skeptics; and a cultural bias against a number of non-Protestant religious traditions. Under better circumstances and with a more equitable distribution of resources, school choice will be less appealing. Of course the classical and egalitarian liberals will differ over issues such as how much weight should be given to parental authority in the context of education policy. ^{Footnote 63} Yet the political reality is that we have to contend with a long tradition in America of giving parents more authority over educational decisions than in many other Western democracies.

To fill out this reason to favor school choice, we should also consider the prima facie reasons against home-schooling. ^{Footnote 64} There are number of concerns liberals should have about home-schooling, ranging from inadequate instruction, the promotion of values at odds with civic values essential to a democratic society, including marriage equality, and the endorsement of patriarchal values that impede educational and career options for women. The considerations against Yoder also apply to the case of home-schooling. However, if a school choice option provides an adequate education, then a religious parent might opt for this rather than a private or home-schooling option. This would produce less objectionable results and possibly even very good results: a school with a religious orientation and with adequate science education standards is likely to provide a better product than a home-school option, when the parents are science skeptics.

Second, school choice might strike the right balance between the liberal aims of education, including adequate training for citizenship, the labor market, and respect for the religious values of parents who are wary of typical curricula in public schools. Some egalitarians will insist that access to education is a basic right that ought not to be balanced against demands by religious parents. In an ideal context, that may be the right position. In America, however, the courts have consistently denied claims by plaintiffs that education is a fundamental constitutional right. ^{Footnote 65} In addition, First Amendment jurisprudence aims to accommodate religious citizens. These are considerations in favor of school choice. When the real choice for a parent is either home-schooling or a public school with an official secular education in all subjects, a choice scheme could result in better education outcomes for students. When a child enters a public school with adequate science education rather than being home-schooled by a creationist or educated in a private school that provides poor training in science and other areas, we should count this as an improvement over the status quo.

A third reason to consider school choice is one that has some appeal in the current political debates over education policy. Parents whose children are in Title 1 schools, ^{Footnote 66} schools with high rates of violence and low graduation rates, might for good reason welcome school choice programs. This reason should be treated with considerably more caution than the others, but it is a reason nonetheless. The reason for caution here is: without other changes to education policy, most especially changes that effectively address significant inequalities, a school choice program may result in further stratification by race and by class. This would be an objectionable outcome, one that in practice could be far worse from the standpoint social justice than the sum of benefits to children who avoid home-schooling.

The test for school choice programs is the real outcomes they generate. What one’s preferred political theory predicts does not carry the argument here. The egalitarian liberal position should be, “in principle we are not opposed, yet our support will depend on the results of such programs in practice.” Classical liberals who endorse Milton Friedman’s argument for voucher schemes ^{Footnote 67} will be more favorably disposed to school choice programs than egalitarian liberals. For a classical liberal, vouchers are appealing because they reduce state authority over education, give parents more control over their child’s education, and thus, according to this view, do a better job than traditional public schools at serving the ideal of liberty. An egalitarian will be less willing to endorse this approach to school choice. From her perspective a number of factors are important to consider when examining school choice and justice in education. These include: 1) the impact of inequalities of wealth on children’s education, 2) whether school choice programs favor gifted or special needs children, and 3) parents’ ability to make informed choices. ^{Footnote 68} Each of these factors can impact the educational fate of children in significant ways. In a context such as contemporary America, these are “in practice” reasons to worry about the effects of school choice programs. ^{Footnote 69} These reasons may in fact outweigh those that support school choice. On the other hand, in the absence of a voucher or choice scheme, a parent might have to decide between a typical public school and home-schooling. If these are the options, she may well opt for home-schooling, which in some cases will be far worse for the educational prospects of her children, especially if she is a young earth creationist who favors a patriarchal conception of the family. However, if a voucher or choice program is in place, provided the options include schools that provide an adequate education, this would be preferable to a home-school option in which children are educated by poorly trained parents. Given this choice, liberals should favor a school choice program.