I want to start with a paradox. In the secular age, as Charles Taylor has amply illustrated, religious belief no longer structures our social imaginary. Instead, it has become one option, one possibility, among others: one of the ways in which we give meaning to our lives. The secular age, then, is characterized by the fact of pluralism—an irreducible pluralism of beliefs, values, commitments. Yet we secular moderns also give special primacy to freedom of religion. Freedom of religion is standardly presented as the archetypical liberal right. So the paradox is this: how (and why) do we protect freedom of religion in an age where religion is not special?

Here’s a plausible solution to this paradox. We could say, roughly, that freedom of religion is in fact a sub-set of a broader class of freedoms. So instead of seeing religion itself as a special good, we say that religion is one of the ways in which individuals seek the good for themselves. Exercising one’s freedom of religion is one of the ways in which we exercise a more generic freedom—moral freedom. Let us call this an egalitarian solution to the paradox I started with. An egalitarian theory of religious freedom does not deny that religious belief is special and should be respected and protected. What it denies is that religious belief is uniquely special: it can and should be analogized with other beliefs and commitments. Many contemporary liberal philosophers are egalitarians in this sense. John Rawls argues that what the liberal state protects is our ability to form and pursue comprehensive conceptions of the good. Ronald Dworkin sees “ethical independence” as the core value protected by freedom of religion. Martha Nussbaum connects freedom of religion to a conscientious search for “ultimate meaning.”

It is in this context that Charles Taylor and Jocelyn MacLure’s little book Secularism and Freedom of Conscience is of considerable interest. In it, Taylor and MacLure put forward their own egalitarian theory of religious freedom, and a radically inclusive one at that: they argue that all “meaning-giving commitments” should be protected on the same basis as religious commitment. The volume is also fascinating when read as a statement of Taylor’s political theory—a normative companion to the more historical, epistemological, and philosophical diagnoses of our contemporary condition found in Sources of the Self and A Secular Age.

To put my cards on the table: I agree with Taylor and MacLure that normative egalitarianism is the right response, ethically speaking, to the deep moral pluralism of the secular age. What I shall suggest, however, is that they—like other egalitarian philosophers—have underestimated the profound tensions that beset egalitarian theories of religious freedom. And these tensions can be traced back to the difficulties of identifying a liberal theory of the good in the secular age—in a world where conceptions of the good are irreducibly pluralized, individualized, and subjectivized. In brief, the story I want to tell is also a very Taylorian story, for it is one that raises questions about liberal neutrality about the good.

Writing in the context of the Canadian debate about reasonable accommodations, Taylor and MacLure begin by defending the idea that members of religious minorities have a right, on non-discrimination grounds, to enjoy similar opportunities to practice their religion as members of the majority. I have no quarrel with this idea, and have argued along similar lines in relation to the affaire du foulard in France. But I’d like to focus on their second main point, namely that the question of reasonable accommodations raises a more fundamental problem: in virtue of what are religious believers entitled to special consideration in the first place? They answer that religious belief, for purposes of legal exemptions, should only be seen as a subset of a broader category of beliefs that deserve protection: “moral beliefs which structure moral identity”—what they call “meaning-giving beliefs and commitments.” And this also covers a broad spectrum of non-religious beliefs and practices—from secular pacifism to eco-centric vegetarianism, through duties of care to terminally-ill loved ones. The notion of meaning-giving commitments is broader than that used by other egalitarian philosophers: in contrast to Rawls, they do not insist that individual beliefs be ‘comprehensive’ in scope; and they reject Nussbaum’s emphasis on “ultimate existential questions.” It is a feature of the secular age, they point out, that people’s ethical commitments take the form of “fluid, eclectic set(s) of values,” which are not integrated into a comprehensive, integrated whole, and which are not perceived as ‘unconditional rules for action.” At certain times, however—such as the illness of a loved one—the pursuit of certain core values become paramount and gives meaning and shape to one’s life. In sum, we can say that Taylor and MacLure take the ethical pluralism of the secular age far more seriously than other egalitarian philosophers. Rawls and Nussbaum, it seems, still hold a traditionally religious understanding of the scope (comprehensive) and content (“ultimate questions”) of what counts as a morally weighty secular belief.

Drawing on Taylor’s rehabilitation of “ordinary life” in Sources of the Self, Taylor and MacLure detect pockets of moral depth in ordinary life—in the sudden encounter with finitude in the event of the death of a loved one; or in eco-centric vegetarians’ profound convictions about the wrongness of meat consumption—to take their two favourite examples. What makes those commitments particularly weighty is that they allow individuals to act with integrity—where integrity is defined as congruence between one’s perceptions of one’s duties and one’s actual actions. What the end-of-life carer and the eco-centric vegetarian have in common is that they both seek to act in accordance with their conscience. “Here I stand, I can do no other,” as Martin Luther is said to have said. Taylor and MacLure note that forcing someone to act against her deep conscientious convictions constitutes a “moral harm” equivalent to the kind of “physical harm” that justifies the special accommodation of citizens with disabilities. So, they conclude, citizens with intense categorical meaning-giving secular beliefs have a pro tanto claim to be considered for exemptions from burdensome laws.

So have Taylor and MacLure developed a plausibly egalitarian definition of morally weighty beliefs, which is not biased in favor of religious beliefs, yet adequately protects the underlying values expressed by the ideal of freedom of religion? My assessment is in two parts. In the first, I draw attention to one significant virtue of their account, which is that it implicitly relies on a very Taylorian idea of “strong evaluation,” In the second, I cast some doubts about the viability of the individualistic, Protestantized, subjectivist conception of strong evaluation that underpins their account.

First: Taylor and MacLure get to the heart of a key feature of freedom of religion—one that is strangely neglected by contemporary liberals. It is this: what Taylor said about negative freedoms in general—that they are empty without “strong evaluations” of what they allow the pursuit of—applies with particular acuity to freedom of religion. Freedom of religion, by contrast to more generic freedoms of thought, belief, and association, relies on a moralized distinction between valuable and non-valuable activities, and serves to protect a sub-set of the former. It is a freedom to pursue a specific end and activity: it refers to the pursuit of a conception of the good with a specific shape, content, and form, rather than the means through which any conception of the good can be pursued. Furthermore, in the case of exemptions and accommodations, which is our focus here, freedom of religion generates demands of positive assistance in pursuing those activities. This means that, when adjudicating such claims, it must be decided which claim correctly expresses the values underpinning the general principle. Even though they do not explicitly draw on Taylor’s earlier writings, Taylor and MacLure are open about the need to make “strong evaluations” about the values that freedom of religion is supposed to protect. This confirms the long-standing Taylorian view that rights protect substantive values: we care about rights because of the good that they protect, which cannot be reduced to individual freedom of choice. So our authors do not shy away from openly perfectionist evaluations, setting “trivial” against “central” commitments, and “mere preferences” against “core convictions.” Such perfectionist discriminations, it seems to me, are inherent to any serious reflection about the value of freedom of religion. Perhaps this is an obvious point, but it is one that contemporary liberals—punctiliously attached to an ideal of neutrality towards the good—have not fully come to terms with.

Who, then, is to make the strong evaluations required to distinguish between meaning-giving and trivial commitments? Taylor and MacLure’s empathic response to this is: the individual claimant herself. Here they anticipate the charge—often levelled at Taylor’s conception of positive liberty—that the idea of “strong evaluation” could give the state the authority arbitrarily to discriminate between better and worse ways to exercise one’s freedoms. Instead, Taylor and MacLure assert that “the special status of religious beliefs is derived from the role they play in people’s moral lives, rather than from an assessment of their intrinsic validity.” They defend what they call a subjective conception of freedom of religion, according to which only individuals—not the state, nor religious authorities—are in a position to explain which particular beliefs and commitments are key to their sense of moral integrity. Judges only have to assess whether such claims are made with sincerity (so as to rule out, when possible, fraudulent or pretextual claims). Yet ultimately, the subjective conception of freedom points to the sovereignty of private strong evaluations.

There is much to recommend in this account, to which I am very sympathetic. But I’d like to draw attention to two difficulties.

First, Taylor and MacLure effectively collapse religion into conscience, and implicitly assume that the latter category is more inclusive than the former. But we may wonder whether this is the case, or whether anything is lost in the re-description of freedom of religion as freedom of conscience. Assume I am a devout Muslim. I observe Ramadan, say my prayers every day, wear the hijab, give zakat, and send my children to Quranic school. Or assume I am a practicing Catholic. I observe Lent, try not to eat meat on Fridays, celebrate Easter, go to church every Sunday, have my children baptized and confirmed. For many Catholics and Muslims (but also other Christians, Jews, Hindus, and Buddhists) the religious experience is fundamentally about exhibiting the virtues of the good believer, living in community with others, and shaping one’s daily life in accordance with the rituals of the faith. Those rituals are meaning-giving and connected to believers’ sense of their moral integrity. Yet they are not duties of conscience—though they are often re-described as such. The good religious life is a life of constant, difficult, ritual affirmation of the faith against the corrupting influences of the secular world. It is not often one in which one single obligation (say, wearing a particular dress, going to mass) is so stringent as to promise eternal damnation if it is not fulfilled. Taylor and MacLure tend to re-interpret acts of habitual, collective, “embodied practices” of religious devotion as Protestantized duties of conscience. While such a description tallies with the individualization and subjectification of religious experience in contemporary societies, it also has two unanticipated consequences. First, it perversely encourages the most fundamentalist and rigid interpretations of religious dogma. It rewards those Christians who present their objection to homosexuality as a matter of conscience (“here I stand and I can do no other”) over and above those habitual believers who seek to accommodate their religious life to a secularizing world, often with considerable unease and forbearance. So here’s another paradox: in insisting that only beliefs that are intensely held, and experienced as categorical duties, should be candidates for “reasonable” accommodation, Taylor and MacLure accommodate those with the least “reasonable” beliefs.

The second unexpected consequence of the reduction of religion to conscience is that it seems to deny protection to the cultural, habitual, embodied, and collective dimensions of religion. Consider the following practices, which currently generate rights to exemption from general laws on religious freedom grounds in various countries: accommodations of religious dress in the workplace, the consumption of hallucinogenic drugs such as peyote in Native American ceremonies, church autonomy in the appointment of its leaders, tax exemptions for religious charities.

None of these activities are properly described as conscientious activities, and therefore it is unclear whether they would be entitled to accommodations under Taylor and MacLure’s theory. Furthermore, in a Canadian context, where cultural identities often feature as the archetypical meaning-giving, integrity-constituting commitments, Taylor and MacLure’s lack of reference to culture is perhaps surprising. They seem to underestimate the communal, cultural dimensions of religion itself and betray an (unexpected) Protestant bias in the interpretation of where the good of religion is located. Whether such a bias is compatible with the egalitarian impulse of the theory is open to question. Here is a heretical thought. Perhaps the ideal of conscience is not a thin, uncontroversial, neutral, liberal conception of the good. In line with the “social thesis” Taylor himself puts forward as a critique of Rawlsian liberalism, and of which a complex version appears in A Secular Age, perhaps our ideals of individualism and conscience are not what remains (but was always there) once the obscure, mystifying debris of traditional community and religion have crumbled away. Instead, we have become individuals—of a particular kind—through a contingent Christian and post-Christian trajectory. If that is the case, how suitable is this particular conception of individual conscience in the pluralistic secular age?

There is a second, connected difficulty with Taylor and MacLure’s subjective notion of freedom of religion. While they only consider examples of morally admirable commitments (pacifism, caring for the sick, protecting animal rights) it is not difficult to think of a range of conscientious actions that may be morally trivial, morally wrong, or morally bad. In those cases, should individual strong evaluations be supreme, or are different standards called for? One issue is how to distinguish trivial from morally significant beliefs. Taylor and MacLure assume there is a shared understanding of the difference between a morally trivial and a morally significant act. Yet, under conditions of deep moral pluralism, it is precisely those kinds of strong evaluations that are likely to be contested. Consider the standard defense of the smoking of peyote—an otherwise illegal drug—by US courts (post-Smith). While injecting drugs merely to “get high” would count as a trivial, frivolous purpose, injecting drugs for spiritual purposes, as practiced within some Native American groups, rightly fall under the category of a morally significant act deserving of protection. But what if individuals not belonging to a religion sincerely claim that they are also using drugs for spiritual purposes? Does “spiritual purpose” extend to dealing with depression, seeking higher truths through controlled intoxication, or coming to terms with existential pain? How far exactly is moral life in the immanent frame pregnant with spiritual purposes?

The other issue is whether freedom of conscience should permit individuals to do bad or unjust things. Taylor and MacLure avoid the difficult question of whether freedom of conscience positively protects a right to do wrong. One very preliminary hypothesis: In the philosophical tradition of thinking about conscience—whether Greek, Buddhist, Muslim, Jewish, Catholic, or Kantian tradition, to name just a few—conscience is respectable and admirable, not only as a subjective individual faculty to live in conformity with one’s own good; it is, more deeply, respected as the faculty to live in conformity with what one sincerely perceives to be the demands of the good (which is why Antigone’s dilemma is so poignant). In the natural law tradition, conscience is the faculty with which individuals exercise practical judgement about how to apply a general objective moral law to concrete cases. Individuals are fallible, and consciences may err. But conscience is admirable because it is a sincere, though fallible, attempt to find the good. Conscience, therefore, cannot demand us to do evil, inhuman, or outrageous things, even though it can mislead us about the good. But if there is a deep (if complicated) connection between respect for conscience and a non-subjectivist assessment of its content, then individual strong evaluations will likely be an unreliable guide about what conscience really requires of them.

Where does this leave us? To conclude, I see Taylor and MacLure’s succinct but densely argued chapter as the most promising attempt to articulate the morally admirable human faculties traditionally protected by freedom of religion, in ways that respect the deep pluralism of the secular age. I have pointed to some problems, which are not so much fatal flaws as unavoidable tensions within the politico-legal philosophy of religious exemptions.

The suspicion is that liberal neutrality about religion ultimately “piggy-backs” on ideas, conceptions, and values that originally made sense in a world comprehensively structured by a broadly Christian ethics. In that world, where early liberal ideas of toleration and freedom of religion were articulated, Christian ethics provided the moral framework within which “strong evaluations”—between good and evil, significant and trivial, etc—were taken for granted. Then it could be coherently assumed that “religion” was a good thing; that any activity pursued under the aegis of religion was therefore also good, and that churches were alternative, self-standing sources of normativity to that of the state. Religion on that view operated as a normative “black box,” the content of which the state could try to ignore. It is when this box is thrown open by the egalitarian impulse of the secular age that the need for new “strong evaluations” re-appears. Yet those strong evaluations are inherently problematic in a world where there is no publicly validated religious or moral faith, and where the state is expected not to take sides between different ways of conceiving and living the good life.

Egalitarian liberals have struggled to define, in a way that is suitably non-sectarian and evaluatively neutral, the morally admirable faculties that traditional freedom of religion can be said to protect. Taylor and MacLure rightly seek to locate those human faculties in the moral predicaments thrown up by ordinary lives, and in the strong evaluations that individuals make in the process. Yet the emphasis on conscience tends to favor a Protestant understanding of what a religion is, as well as being parasitic on an implicit, unarticulated theory of the good. All of this only illustrates one of Taylor’s most profound contributions to political philosophy, pointing to the complex ambiguities that beset the liberal ideal of neutrality towards the good life. And I have sought to provide the sketch of a Taylorian critique of Taylor—a modest testimony of the astonishing fecundity of Taylor’s thought.

A version of this text was presented at “Charles Taylor at 80: An International Conference,” held in Montreal on March 31, 2012.—Ed.