As Ronald Dworkin was writing his Einstein lectures “Religion without God,” at New York University (NYU) in the fall of 2011, I was also working in Washington Square, as a fellow of the NYU Straus Institute for the Advanced Study of Law and Justice. On December 8, just a few days before Dworkin delivered the lectures at the University of Bern, I had the opportunity to attend the last session of his famous Colloquium in Legal, Political, and Social Philosophy at Furman Hall. In that session, co-led by his colleague and friend Thomas Nagel, Dworkin presented the manuscript of his Swiss lectures. After the seminar, we had an anticipatory celebration of Dworkin's eightieth birthday, which would take place three days later. And that was the last time I would see Ronald Dworkin—which might explain why I remember in such detail that seminar, in which he talked about religion without God with more spontaneity and improvisation, I imagine, than he would in the Einstein lectures days later.

Following Dworkin's 2011 publication of Justice for Hedgehogs,Footnote ¹ many of his readers, including me, expected that he would write something deep about religion. His characteristic intellectual honesty and coherence demanded it. For in Justice for Hedgehogs, probably his most comprehensive book, he defended the philosophical thesis of the unity of value and its independence from scientific explanation. He affirmed that the ethical life, morality, and beauty are coherent and mutually supporting. He dealt extensively with ethics, morality, politics, and even with law as a part of political morality. Yet religion was largely absent. In more than 500 pages, there are just a few pages on religious conviction, religious temperament, and religious freedom (see Justice for Hedgehogs [JfH], 502 [index]). Reading these pages, however, one realizes that Dworkin was thinking seriously about religion without finding a definitive treatment for such a large topic. Religion without God was his final response. Unfortunately, his death cut short his plan to extend his treatment of religion. Still, both Justice for Hedgehogs and Religion without God should be read and understood together. They make up an indissoluble unity, as Dworkin himself affirms (see Religion without God [RwG], 166n15).

In Justice for Hedgehogs, Dworkin had deliberately excluded religion from the unity of value for epistemological reasons: “We must not ask our epistemology to make way for whatever it would be nice to believe” (JfH, 83). Religion should remain outside an integrated epistemology, he thought, because “a moral judgment never requires appeal to extraordinary modes of causation. Morality needs no miracles” (JfH, 85). Soon after finishing that book, Dworkin must have started to think that not only how we relate to ourselves (ethics) and to others (morality) but also how we relate to the cosmos (religion) was a source of value independent of matters of scientific fact. So in Religion without God, he tried to recover the concept of religion as a value to integrate it with his theory of the unity of value. But for Dworkin, doing that required banishing God from the religious realm, just as God had been expelled from ethics and morality: “No exercise of creative power, however great, can shift fundamental moral truth” (JfH, 342). Dworkin's theory demanded a sort of “religious atheism” as a unique solution to the problem of fitting religion within his comprehensive theory of value. That is why the first of the lectures in Bern, “Einstein's Worship,” was ultimately retitled “Religious Atheism?” for the book.Footnote ² In this lecture he tried to explain how religious atheism “is not an oxymoron” (RwG, 5) and how religion was not restricted to theism. If a religion without God is possible, as atheistic morality and ethics are, then religion could be part of his integrated epistemology as a matter of objective value.

According to Dworkin, religion is an “interpretative concept” (RwG, 7), in the sense that people use it without any agreement about its meaning. Therefore, religion belongs to the domain of values and arguments, not to the domain of facts and explanations. God, by contrast, is a matter of fact, which requires scientific explanation. As Dworkin puts it, somewhat ironically, the existence of a personal God “is a very exotic kind of scientific fact. But it is still a scientific fact” (RwG, 27). Thus, religion and God could and should be isolated, given the impossibility of “support[ing] a value judgment—an ethical or moral or aesthetic claim—just by establishing some scientific fact” (RwG, 26–27). This is what Dworkin calls “Hume's principle” (RwG, 26; JfH, 17, 222), which supports the independence of morality, with its own standards of justification, as a domain of knowledge. The principle, Dworkin thinks, is based on “the conceptual truth that nothing but another value judgment can support a judgment of value” (RwG, 90). For Dworkin, this statement is like a mathematical truth. Both “mathematics and value are immune from questions about their birth or causal provenance” (RwG, 90).

Dworkin begins Religion without God with a core thesis of his argument: “The theme of this book is that religion is deeper than God” (RwG, 1). For Dworkin, because the existence of God is an (alleged) scientific fact and, therefore, not a source of value, believing in God demands a background value judgment, which highlights the relevance and consequences of the untested (alleged) fact of his existence. So belief in God is independent of the existence of God. Believing in God is just a potential consequence of a deeper worldview which is religion. And religion, precisely as an interpretative concept, can provide value by itself. Therefore, the reality of religion as a value is independent of the existence of God as a fact: “the value part [of religion] does not depend—cannot depend—on any god's existence or history” (RwG, 9).

Inspired by Einstein's religious ideals (RwG, 34–37, 48–59) and by Nagel's approach to religious temperament (RwG, 127–28), Dworkin argued for the existence of a religious attitude that holds that human life and wonder at the beauty of the universe have objective intrinsic value. The religious attitude or impulse is the human being's capacity to discover value in everything that exists. This religion of value is completely independent of scientific explanation, that is, of facts, and therefore also of the existence of any god or gods. For Dworkin, the possibility of the existence of God does not make any difference to the objective truth of atheistic religious values: “the distinction between theism and atheism is therefore itself indistinct” (RwG, 31). According to Dworkin, this religious attitude could be shared by both theists and atheists, and thus could ground a broader right to religious freedom, protecting both theistic and atheistic convictions: “What divides godly and godless religion—the science of godly religion—is not as important as the faith that united them,” concludes Dworkin (RwG, 29).

I agree with Dworkin that a religious attitude binds both theists and atheists and that atheism is not a good argument for non-religiosity. I believe that a religious instinct is a human trait, given voice in questions about the meaning of one's life. But I do not agree that the divorce between God and religion is an epistemic norm. Rather, an epistemology for both theists and atheists is possible if God is not only a matter of fact but also a matter of value and, therefore, relevant for both morality and religion. Moreover, religions, or at least the Abrahamic religions, are not just a matter of value, but also a matter of fact, since they hold that human beings are created by God (fact) and for God (value). So the distinction between fact and value breaks down when we are dealing with a personal living Supreme Being who is Love. A loving God surpasses this distinction of rational human knowledge between explanation and argumentation.

According to the Abrahamic religions, the revelation of God is primarily an event, a fact. The divine plan of revelation is simultaneously realized by deeds and words which are intrinsically bound up which each other. They cannot be separated as Dworkin would have it. No, the conventional theistic religions do not consist of two separable and independent parts, “a scientific part and a value part” (RwG, 23). They form an indivisible unity that cannot be analyzed separately without falling into oversimplification. That is exactly what happens when Dworkin reduces the existence of a personal God to the scientific fact of there being an entity that “cannot of his own will create right answers to moral questions or instill the universe with a glory it would not otherwise have” (RwG, 26). Religious events are value-facts. For Christians, for instance, the resurrection of Jesus is an historical event that nevertheless transcends history, opening the door to a new dimension of human existence, to a new space of life with value of its own. Moreover, for Christians, Jesus himself is Truth: at once a fact and an objective value. Thus, to separate fact and value in relation to Christ is to kill the essence of Christianity.

Continuing with Dworkinian terminology, we can also affirm that God is both the supreme fact (Supreme Being in the Western tradition) and the supreme value (Supreme Good and Supreme Beauty in the Western tradition), because, as creator, God gives the real meaning, the ultimate value, to the goodness and beauty of human life and the universe. Likewise, revelation is a religious fact and also a religious value. In some ways, revelation provides us automatically with the “background value judgment” that allows us to make a correct, although partial, judgment about God. This statement is consistent with the idea that God does not need to figure in an acceptable explanation of morality and religion, and that revelation is not the only source of knowledge, even of the knowledge of God.

Some explanation of objective value can be provided without adverting to the existence of God, but that does not necessarily mean that no further explanation is required or available, or that the existence of God cannot be the ultimate explanation of morality, religion, and objective value. Something similar can be said of divine revelation as a constitutive part of the Abrahamic religions. A reasonable explanation of objective value can be provided without adverting to divine revelation, but that does not necessarily mean that no further explanation is required or available, or that divine revelation cannot be the ultimate explanation of morality, religion, and objective value.

According to the Abrahamic religions, only God—God's existence and action (God-living Fact)—can fully answer questions about the good and the beautiful (God-Value), because God, as supreme Good and ultimate beauty, is the foundational, the ultimate, and even the exclusive source and condition of morality and religion. The existence of such a living and loving God, however, does not affect believers alone. It radically affects all human beings, believers and nonbelievers. Exclusivity does not mean, therefore, that nonbelievers remain outside morality or religion, because morality and religion are reasonable, and the human being is rational. Religion and morality presuppose reason. So reason as such—not in a narrow, positivistic sense open only to strictly scientific explanation, but in the broadest sense—can be considered a source of morality and religion to the extent that it is a creation of God. Reason is the meeting point between believers and nonbelievers. For believers, human reason is a gleam of the divine reason; for believers and nonbelievers, it is the most powerful human capacity to find the good and the beautiful, that is, objective value. If the starting point is reason and not objective value, we do not need to separate scientific reason from religious reason. They can be united.

The existence of a creative reason is not a miraculous idea, but a meta-scientific one, which recognizes the validity and benefits of the scientific method, the autonomy (but not total independence) of morality, and the intrinsic limitations of both. Just as scientific reason is not the whole of reason, neither is creative reason. The latter is a foundational form of reasoning, but not the only possible form. Through reason, one can find ways to approach God by contemplating the order and beauty of the cosmos, and the reality of the human person as a free creature open to objective truth and beauty, and longing for happiness. These converging arguments are not based on a scientific explanation but on human experience and on causality, though they are rejected by Dworkinian epistemology. Dworkin, following Nagel, believed that divine creation does not provide a better explanation of human life than random mutation (RwG, 127); but scientific explanation is not the only source of knowledge. Thus, an expanded reason that integrates scientific explanation, sapiential understanding, and human experience (for example, of love, of faith and of hope), is, in my opinion, more favorable to the idea of divine creation than to the idea of simple random mutation.

Dworkin excludes from his epistemology any transcendent source of divine reality, however vague: “religious atheists do not believe in a god and so reject the science of conventional religions and the godly commitments” (RwG, 24). The Dworkinian approach to religion closes the door to any consideration of revealed belief or transcendent knowledge as an epistemological axiom, reducing potential religious knowledge to a matter of personal conviction with limited epistemological relevance. He argues that “if we declare our faith in some special religious form of perception, however, we have no way of integrating our belief in the faculty that delivers that perception with any more general account of how that faculty functions” (JfH, 84). However, Dworkin's epistemological paradigm impedes the development of such an account by closing the door to any transcendent religious epistemology. That is why he argued that “the science of a godless religion may provide, in a different way, all that the science of godly religions can actually propose” (RwG, 152).

The god or the gods that Dworkin discussed are caricatures of the Abrahamic God, a truly unbelievable god for adherents of the Abrahamic religions. That is why he lowers religion to the status of a religious attitude toward human life, forgetting the superabundant light that revelation provides to human beings. The Abrahamic God, in turn, is unthinkable according to Dworkin's epistemological framework. The reason is that Dworkin rejects any external and meta-ethical inspection of morality, religion, and truth. Taking religion seriously, and not as a mere attitude, demands a truly integrated epistemology—not in the Dworkinian sense, but in the sense of integrating faith and reason, religion and knowledge, metaphysics and epistemology, theology and history, and data and interpretation: namely, all modes of human understanding (including imperfect ones, like analogical reasoning). In sum, epistemology should be open to the possibility of a transcendent reality, without any limitation besides reasonableness. Otherwise, it cannot support a holistic normative paradigm for both religion and religious freedom.

Dworkin affirms that “the religious attitude rests finally on faith” (RwG, 18); however, the faith he is demanding for religion is merely the conviction that mathematics demands: that all the systems work, that the system of value judgments generates objective value. Faith is the “irresistible conviction in our experience of value” (RwG, 21). Faith for Dworkin is just a matter of conviction, not what it is in the Abrahamic religions: an adequate response to the invitation of God to live in his own company. Abraham was a man of faith because he submitted freely to the word of God. A response to God is both a fact and a value. It is a response which transcends history and generates value by itself.

For that reason, believing in God (conventional religion) can contribute to living well (ethics) and to treating others well (morality). For believers, believing in God is a necessary but not sufficient condition for living well. Believing in God and living well have different meanings, but they are related. The more a person's belief in or obedience to God contributes to living well and to treating others well, the better that person's belief is in political terms. It therefore makes sense to speak of believing well. The human person as a source of value can integrate ethical, political, moral, and religious values to generate transversal values. In this sense, the Dworkinian idea of the “unity of the value” is very helpful. What is needed, though, is a holistic unity of value that takes into account transcendent values, as opposed to the more reductive Dworkinian conception.

Human cognition forms an organic whole, but this organic whole is not complete without incorporating the human person's transcendent religious experience. That requires recovering the “sapiential dimension” of human knowledge, which is critical to the search for ultimate meaning. Dworkinian epistemology must acknowledge the human being's capacity to transcend factual and empirical data so as to contemplate foundational reality in the search for truth through analogical reasoning. In order to avoid the fragmentation of knowledge, an adequately integrated moral epistemology must be based not only on the unity of value but mainly on the unity of the human person as such, in his or her physical and transcendent dimensions.

Dworkin's approach to religion shapes his approach to religious freedom. The right to religious freedom in Western culture was established as a right supported by a Protestant value (which Catholics, Jews, and, increasingly, Muslims have come to share) in which each person may choose his or her religion, that is, how to worship God. At the core of the right was the duty to worship God, and religion was an instrument. Thus, the right also protected the option of believing in God apart from religion. An example of this was deism. The possibility of atheism or of religion without God was not especially protected by the right to religious freedom. The original approach to religious freedom was chiefly theistic.

Today, a wider and more multicultural concept of religion is making headway in the international realm—religion as a coherent system of beliefs, convictions and worldviews, regardless of content. The terms belief and religion have been broadly construed in order to protect theistic, non-theistic, and atheistic beliefs, as well as the right not to profess any religion or belief. This new and wider pluralistic approach to religion does not necessarily imply either the existence of God or the transcendent nature of the human being, as the original concept did, so agnosticism and positive non-religious beliefs, for instance, have been recognized as standards of religious beliefs. In the modern secular age, paradoxically, there is a sense in which there are no longer nonreligious human beings, because all human beings (whether theistic, nontheistic, agnostic, or skeptical) are (legally speaking) religious (that is, protected by freedom of religion) even if they are fervently opposed to religion.

I personally agree with this approach as an inexorable effect of multiculturalism and globalization. And I agree with Dworkin when he affirms that there is no justification for a special right to protect exclusively theistic religions (RwG, 117). Both types of “religion” should be protected. However, this does not mean that we must base the right to freedom of religion on ethical independence (RwG, 130), as if God (as distinct from the Dworkinian god or gods) no longer played any role. Otherwise, an agnostic if not atheistic position would become normative.

Indeed, this recent and welcome international legal trend, which tries to avoid intolerance and discrimination against any kind of belief in religious matters, can never justify excluding the Abrahamic God as such from the legal meaning of religion, as Dworkin has sought. It is one thing to deny that the idea of God is constitutive of the legal concept of religion, in order to protect under the umbrella of the right to religious freedom all kinds of convictions and beliefs; it is a very different thing positively to exclude God, in the Abrahamic sense of the term, from the idea of religion, as Dworkin has tried to do in his theory.

A legal system open to transcendence, and therefore, open to God, is not necessarily closed to immanence; much less must it protect only transcendent religions and beliefs. As an essential expression of pluralism and a precious asset for believers and nonbelievers, a democratic constitutional model must protect both transcendent and immanent religions and beliefs, with the only limit being imposed by public order. To regard religion only as an immanent phenomenon, as a religious attitude or temperament, is a negative way to protect religious freedom; it protects only a part of religion, an emotional part of religion, not religion in its fullness. This unjustified reductionism, which could be called the tyranny of legal atheism, would impose a uniquely secular religion without God.

Dworkin's approach to religion demands only a very narrow protection of religious freedom. If religion is just a matter of convictions, without divine commitments and a duty of ritual worship (see RwG, 24), the right to religious freedom should be protected only as a matter of what Dworkin calls “ethical independence” (RwG, 130). In other words, there is no reason to treat religious freedom as a special right. As Dworkin affirms:

The concept of ethical independence (RwG, 130–32) is the key to understanding Dworkin's arguments. According to the Dworkinian principle of ethical independence, government must not rely on any justification that directly, indirectly, or covertly supposes any ethical conviction about what is better or worse for a good life. So, ethical independence, “stops government from restricting freedom only for certain reasons and not for others” (RwG, 131). Ethical independence means an absolute personal sovereignty in foundational matters, such that a person should never accept any judgment in place of his or her own. Insofar as religion is a foundational matter, no government may constrain religion for any reason except when necessary to protect the life, security, or liberty of others (especially to enforce nondiscrimination). Moreover, in no case may a political community dictate ethical convictions to its citizens. A decision of the political community may force us to live in accordance with collective decisions only if it does so without imposing ethical convictions. Therefore, there is no place for any religious “ethical paternalism” that favors one faith over others (see RwG, 134). Thus, according to Dworkin, ethical independence requires individuals to refrain from imposing ethical convictions by treating a particular religion as sui generis or treating religion as such as special. The consequence is that a “priority of nondiscriminatory collective government over private religious exercise seems inevitable and right” (RwG, 137). So, according to the norm of ethical independence, religious exemptions from neutral laws have no justification. In connection with the famous Peyote caseFootnote ³ Dworkin affirms that “if the Native American Church is entitled to an exemption from drug-control laws, then Huxley followers would also be entitled to an exemption, and skeptical hippies would be entitled to denounce the entire drug-control regime as a religious establishment” (RwG, 135).

By applying the principle of ethical independence, Dworkin reduces freedom of religion to freedom of conscience, and freedom of conscience to free morality. Because freedom of religion is for him just a matter of convictions, Dworkin would provide the same treatment to both moral and religious convictions. It is therefore understandable that he would reject religious exemptions in an effort to provide equal protection to citizens. My argument, however, is different: the application of religious exemptions is a consequence of equal protection, and a demand of the right to religious freedom that is irreducible to freedom of conscience. Freedom of religion protects religion; freedom of conscience, by contrast, protects individual ethical autonomy. Without freedom of conscience there is no freedom of religion, but freedom of religion embraces some areas (education, public manifestation of faith, worship) that do not fall directly into the realm of freedom of conscience.

Religious freedom is based on the idea that political communities are by definition non-religious communities. This non-religious character justifies the existence of a political right to religious freedom which protects citizens against any kind of political intrusion in religious matters, even based on potentially neutral legislation, as in the Peyote case. The specific and unique legal tool for protecting religious freedom as an affirmative defense is precisely the religious exemption, which should be recognized by the judiciary. The religious exemption makes possible acts that are in accord with the overall legal system but at odds with a general statute. The legal justification of the religious exemption is that secular legal systems should prefer to abstain from the application of a law, even a neutral law, in cases of conflict with religion. In doing so, political communities protect their own political nature, which again is non-religious by definition. By offering religious exemptions, the judiciary identifies specific violations of the right to religious freedom, thus opening the door, legally, to action in the public sphere at odds with what a statute requires. It does not thereby authorize action against the legal system as such, for it is precisely the legal system that supports the right to religious freedom as fundamental. The judiciary is just resolving a conflict between a basic human right with a special protection (religious freedom) and a specific statutory provision.

Freedom of conscience, however, protects a minimum area of moral freedom which under no circumstances may be violated. Where the frontier between private life and political authority should be drawn is “a matter of argument, indeed of haggling,” we can say with Isaiah Berlin.Footnote ⁴ The scope of freedom of conscience should be narrower than that of religious freedom, since political communities retain the freedom to develop their own public morality but not their own public religion. And because of our interdependence, the exercise of freedom by some citizens always comes at the expense of restricting the freedom of others.

Freedom of conscience should be supported, not by a norm of ethical independence, but by ethical autonomy, a substantially different concept. Ethical autonomy emphasizes the indispensable relationship between human freedom and moral order in all dimensions of the human person: individual, social, and transcendent. According to ethical autonomy, it is possible to harmonize legality, morality, religion, and freedom. For ethical autonomy, the essence (and the end) of moral freedom is the pursuit of the good, in accordance with which people can shape their own choices. As Joseph Raz rightly emphasizes, “Autonomy is valuable only if exercised in pursuit of the good.”Footnote ⁵ From this point of view, moral autonomy is analogous to freedom of scientific inquiry. The purpose of freedom of inquiry is the research as such, not autonomy or self-determination, although researchers need autonomy in their fields of research. Freedom of inquiry serves science, and not the freedom of the scientist. Freedom of inquiry is valuable only as a means to high-quality inquiry. On the moral independence approach, however, moral freedom will be closer to freedom of thought (if it is not a species of it), in which the essence is the freedom, the independence, regardless of the quality of the resultant thought.

The right of religious freedom operates along the three dimensions of the human person. It operates in the individual dimension (along with freedom of conscience) by protecting the human person in his or her personal search for the truth against any constraint by political authorities. Because freedom of religion is the inclusive patrimony of believers and nonbelievers, it fully protects from religion those who choose to have nothing to do with religion. In the social dimension, the right to religious freedom allows individual persons to live their faith in community and protects political communities against religious and secularist fundamentalisms as rejections of legitimate pluralism. Finally, in the transcendent dimension, the right to religious freedom operates both by protecting religion as such as a basic good and by opening the door to transcendent truths. The Dworkinian framework of religious freedom based on ethical independence protects the individual dimension, since this is the place in which the religious attitude is developing, but it banishes the other two dimensions—the social and the transcendent—thereby banishing completely the public dimension of religion.

In sum, although brilliant and well articulated, Dworkin's approach to religion and religious freedom belittles both the idea of God and the idea of religion. It undervalues God, by considering him just an untested scientific hypothesis (an alleged fact) without value of his own. Dworkin reduces God to a metaphor. For this reason, for Dworkin the existence of God, god, or gods is epistemologically irrelevant. The conviction of God's existence and profound ethical and moral convictions are not as independent as Dworkin argues (RwG, 146). Dworkin's approach also underestimates religion, by considering it an interpretative concept, which is the result of a human attitude submitted to a rule of objective necessity without any causal provenance. The argument proposed by Dworkin in order to separate God and religion does not work in practice since it is possible to find value in God and facts in religion and it is not possible, at least in the Abrahamic religions, to separate divine events from divine value. Transcendent facts overtake the scientific dimension. So, a scientific-value division cannot be formulated in Abrahamic religions without distorting them. Facts and values are completely bound. They transcend each other. The “Hume principle” cannot be applied to God, since God is more than a mere event or scientific fact; God is a personal loving being who can make judgments of value true. God is the very Truth. By reducing religion to a matter of value, Dworkin is diminishing Abrahamic revealed religions, specifically the relevance of divine commitments and the duty of worship. In doing so, Dworkin reduces religion to private convictions. Pace Dworkin, religion is something more than a matter of value and conviction, and God is more than an exotic entity that cannot create right answers to moral questions. A loving God is deeper than the path to the divine love (religion).

This religious misunderstanding notably affects the Dworkinian idea of freedom of religion. Dworkin distorts the right of religious freedom, denying its special character and broad protections. This human treasure of religion should be protected by legal systems not just as an attitude or human instinct, but as such, as a gift: one of the most precious gifts, which dignifies humankind. This demands a special protection for religion and not just the general protection provided by the principle of ethical independence. Religion is more than a matter of conscience. Religion is also something more than a personal decision about ultimate concerns and questions. So religion cannot be reduced to a matter of ethical independence in foundational matters. Dworkin established a framework to protect religious freedom within his theory of religious atheism and the unity of value. It serves to protect religious attitudes, convictions, temperaments, sentiments, feelings, and faiths, but not religion as such, and is, therefore, not a truly religious freedom. Our globalized and pluralistic society demands a more comprehensive approach in which both religion with and religion without God will be fully protected. The framework created by Dworkin to protect religion without God does not sufficiently protect conventional religions with God. The protection of Dworkinian religious attitudes only demands ethical independence. However, the true protection of religion demands a specific right of religious freedom even beyond freedom of conscience.