Religious Exemptions?: What the Free Exercise Clause Means

Click here to download this white paper as a PDF.

Three scholars who work on religious liberty issues share their electronic conversation on the topic.

Michael McConnell, a former judge of the United States Court of Appeals for the Tenth Circuit, has argued for many years and with significant influence that the First Amendment’s original meaning sometimes requires religious exemptions.

Vincent Phillip Muñoz, a non-resident senior fellow at Civitas and professor at the University of Notre Dame, has challenged the claim of religious exemptions on originalist grounds.

‍Andrew Koppelman—professor of law, political science, and philosophy at Northwestern University—has argued for religious exemptions on non-originalist grounds.

Not originally intended for publication, we present the exchange here as we believe it clarifies some of the underlying philosophical and constitutional issues about which we disagree. Our engagement became detailed enough that others may find it of interest. 

‍The exchange was initiated by Koppelman, who emailed McConnell a question about his review of Muñoz’s 2022 book, Religious Liberty and the American Founding. In his response to Koppelman, McConnell cc-ed Muñoz. Muñoz then responded to McConnell and Koppelman point by point, which led to the following discussion. Muñoz’s points appear interstitially within McConnell’s. To clarify to the reader who is speaking, Muñoz’s part of the conversation appears in italics. The exchange has been lightly edited for clarity and typos but remains as drafted in the original email exchange.  

McConnell’s review of Muñoz’s book, Religious Liberty and the American Founding, appeared in First Things (Feb. 2023). Koppelman also published a review of Muñoz’s book, “Madison’s Non Sequitur: A Comment on Vincent Phillip Muñoz, Religious Liberty and the American Founding,” in American Political Thought: A Journal of Ideas, Institutions, and Culture 12, no. 3 (2023). ‍

Dear Michael [McConnell], 

You write in your review of [Vincent Phillip] Muñoz that his conclusion about free exercise exemptions “seems incompatible with his central thesis that religious freedom is a natural right.” I don’t understand the incompatibility. Locke thought religious freedom was a natural right, but he bounded it in the same way Muñoz did. What am I missing? 

Thanks. 

Andy [Koppelman] 

‍

McConnell included Muñoz in his email response to Koppelman. This elicited a long response from Muñoz, who quoted McConnell’s response in its entirety and interspersed reactions to what McConnell had written. Here is McConnell’s email, with Muñoz’s interstitial comments in italics: 

[Here McConnell’s response begins] 

1. The natural right of free exercise of religion is the right to act in accordance with conscience, bounded only by the law of nature. The law of nature’s bounds are not an exception to the right, but define the right’s scope. (Muñoz and I are in agreement about this.) 

2. The law of nature, at its core, is the equal right of others to their own natural rights. (Again, Muñoz and I agree.) 

3. Essentially, early state constitutions’ "peace and safety provisos" were attempted codifications of natural liberty’s scope as so bounded. (Again, a point of agreement.) 

4. Not every state constitution contained such a proviso, though most did. These provisos were not necessary because the the law of nature was already understood to define and limit the natural right of free exercise. The provisos were explanatory and declaratory in nature. (I think Muñoz may disagree with part of this.) 

[Here Muñoz wrote: I agree with this.] 

5. The federal provision basically replicates the state provisions, but as applied to the federal government. It does not contain a proviso, but because of point #4 it did not need to; the natural limits on the natural right were presumed. (I think, but am not positive, we agree on this.) 

[Here Muñoz wrote: I agree with this.] 

6. Thus, the free exercise right’s logical structure is as follows: people have a right to act in accordance with conscience, so long as this does not interfere with others’ rights or the public peace. All this is defined (as a matter of logical structure) without reference to how the state’s positive law is framed. (Here is where I think Muñoz and I begin to disagree.) 

[Here Muñoz wrote: I agree with this. I would only clarify that point #6 articulates the nature of the right of free exercise in the state of nature.]  

7. The Smith rule’s logical structure, which Muñoz supports, is nothing like this. It holds that the state’s positive law prevails over individual conscience whenever that law is framed in neutral and generally applicable terms.ⁱ 

[Here Muñoz wrote: See my comments to #8 and #10 below] 

8. Obviously, the natural right could not have been anything like the Smith rule, because in the state of nature there is no positive law, which means that the law’s legal form cannot possibly determine the right’s scope. Instead, substance determines the scope, and the substance is the effect exercising the right has on others.  

[Here Muñoz wrote: I agree with this but would formulate it slightly differently. Like all natural rights, the law of nature (which is the law of reason) bounds the natural right of religious free exercise. In the state of nature, in other words, reference to one’s individual conscience does not establish subjectively the right of free exercise’s boundaries. Rather, what is objectively reasonable (the law of nature) establishes them. An action’s effect on others in large part determines whether it is objectively reasonable (according to the law of nature), so I think I am in basic agreement with what McConnell writes here.] 

9. Some variation of the Sherbert/Yoder rule is a reasonable translation of the natural right, because the effects on others rather than the law’s legal form determines the limits on the right’s scope.ⁱⁱ 

[Here Muñoz wrote: See my comment to #10 below.] 

10. Muñoz agrees that the Free Exercise Clause carries forth the substance of the natural right of religious free exercise. It follows that we should apply some variant of the Sherbert/Yoder rule rather than the Smith rule to interpret the Free Exercise Clause.  

‍

That concludes McConnell’s email. Muñoz’s response follows: 

‍

I believe where Michael and I disagree is on the transition from the state of nature to political society and what this means for the natural right of religious free exercise. I explain my reasoning below.  

The essence of our disagreement involves two related disagreements about the Founders’ social compact theory, specifically about (1) the nature of “inalienable” rights and (2) what authority we give to the state when we alienate a right. These differences are related to a third (and perhaps the most important) disagreement: what is the right of religious free exercise’s scope and meaning? I hold the right to be “narrow” and “deep”—narrow in that it only covers what I deem “worship as such,” deep in that this right can never be abridged (i.e. is not subject to compelling state interest or least-restrictive means analysis). Michael holds the scope and meaning to be, relatively speaking, broader but also shallower (in the sense that it can be limited by compelling state interests).  

Let me explain my thinking.  

Michael’s translation is elegant and seemingly intuitive. In the state of nature, one has the right to act according to conscience, limited only by others’ freedom to do the same. We retain this right (I think this is Michael’s interpretation of “inalienable” though I am not quite sure) upon entering political society. In practice, we can translate this as a right to act according to conscience except when the state pursues a narrowly tailored compelling state interest. Compelling state interest/least restrictive means analysis (1) functions as a reasonable approximation for measuring what constitutes a harm to others, and at the same time, (2) maximizes every individual’s ability to act according to conscience. This combination of (1) attentiveness to the community’s needs and (2) maximum space for acting according to conscience is why the Sherbert rule is reasonable. 

My understanding of the Founders’ social compact theory is different. In short, I believe:  

(a)  The Founders offer a more limited and precise understanding of what authority individuals retain (inalienable right). 

(b)  Regarding authority that is given to the political community, our Constitution (and Lockean social compact theory), places lawmaking authority in the legislative branch. When we join the social compact, we consent to follow the community’s laws.  

(c)  In practice (a) and (b) mean that the state lacks jurisdiction over that which is not alienated (what I deem the “religious worship as such”) and has jurisdiction (subject only to something like rational basis review) over that which is alienated.  

Let me try to lay out my understanding of the Founders in a few steps.  

In the state of nature: 

• A law of nature (reason) governs what is right and wrong. This is the law of reason. 

• Individuals are allowed to follow their inclination (natural freedom), subject only to the law of nature’s limitations. The Founders (and we too today) would say something like, "you are free to do as you wish as long as you do not trespass others’ rights.”  

• Every individual possesses two authorities:  

1. To make determinations about the the law of nature’s substantive meaning in particular cases. The law of nature says, for example, that you can appropriate from the common lake as many fish as you can catch subject only to the natural law limitations of (a) spoilage and (b) leaving enough for others. In the state of nature, each individual has the authority to make determinations on how many fish is too many to take.   

2. To enforce violations of the law of nature (executive power)   

• Rights in the state of nature tend to be insecure because the state of nature tends to break down into a state of war: 

* When individuals make determinations about the law of nature’s substantive meaning, they fail to make impartial, reasonable judgments. Moreover, what the law of nature dictates in specific circumstances is not always clear. In short, our judgments are partial and our knowledge is imperfect. 

* Individuals tend to be excessive when punishing perceived violations of the law of nature. 

Individuals enter into political society to remedy the state of nature’s defects and thereby to secure their rights. When forming political society, individuals alienate most of their rights. 

When individuals alienate their rights, they turn over to the political community: 

• First, authority to make law for the community. This consists, in part, in making determinations for the community about the law of nature’s substantive meaning, i.e. to apply the law of nature to the political community’s particular needs. This is legislative power’s foundation. Here, I think, is the primary point at which Michael and I disagree. I’ll return to this below.  

• Second, authority to enforce violations of the positive law. This is the foundation of executive and judicial power. Those rights that are not alienated to the political authority are “inalienable” rights.  

Regarding inalienable rights, the constituted polity:  

• Lacks authority to make law for the community.  

• Lacks the corresponding executive and judicial authority to enforce violations of laws presuming to touch on inalienable rights. This is the sense in which “inalienable” rights pertain to jurisdiction. This is another point of difference, I think, between Michael and me.  

Let me focus on the differences (at least in my understanding) between Michael’s and my positions. 

I believe inalienable and alienable natural rights to be mutually exclusive categories in the Founders’ natural rights constitutionalism. An individual cannot alienate and retain authority over a right at the same time. When an individual alienates a right, he no longer possesses authority to determine the law of nature’s substantive content as it pertains to the alienated right; he has relinquished the power to determine the right to the political community (specifically, to the community’s lawmaking authority). 

In the state of nature, I had authority to make the determination that catching one hundred fish per day left enough for others. Once I consented to join the political community, I have consented to follow the community’s determination of the number of fish I can catch per day. The same holds true for any religiously-motivated behavior that is not part of the inalienable right of religious liberty.  

Here we have the exact point of disagreement between Michael and me: I hold that religiously-motivated behavior falls into two distinct categories. For the sake of clarity, I will call them religious “interests” and religious “worship.” First, religious “interests” are religiously-motivated actions that are subject to the state’s legitimate authority. They are innumerable, but examples include objections to taking up arms, the education of one’s children, and the use of hallucinogenic substances.  

Religious “worship” is the set of religiously-motivated behaviors over which the state lacks authority. As you know from my book, I don’t try to provide an exhaustive list of such actions but they would include things like what constitutes blasphemy, how ministers are selected, the process by which one’s sins are forgiven, and how water is made holy.  

The specifics of what belongs to what category are obviously important, but more fundamental to Michael’s and my disagreement is the two categories’ existence. I hold that there is a category of religious-motivated behaviors over which we do not grant the state authority. Such a category’s existence is what “unalienable right to worship according to conscience” meant to the Founders. Upon creating the social compact, we do not grant authority to the state over the category of our religious worship. We cannot grant the state authority over the category of religious worship as it is inalienable. That right’s inalienable character, accordingly, translates into a jurisdictional limitation on state power. I believe an implication of Michael’s position is that this category does not exist.  

In my understanding, many behaviors fall into what I have deemed “religious interests.” We do not retain authority over this category when we leave the state of nature. As such, we give the political community authority to make laws over these subjects (e.g., you must bear arms when drafted, you must send your child to school until the age of 16, you may not use certain hallucinogenic substances). When we consent to join the political community, we consent to following such laws.  

It goes without saying that such laws may not always be sound and that a good constitutional order would establish institutional arrangements (representation in a community with a multiplicity of interests, bicameralism, separation of powers, federalism, etc.) to prevent laws that do not reflect the society’s common good. Nonetheless, when such laws are passed, and assuming they lie within the jurisdiction we have granted to government, we are obliged to obey them. There is no right of “conscience” to not obey a duly-enacted law, in the Founders’ social compact constitutionalism. Again, when we consent to join the social compact, we consent to follow the law when the lawmaking authority acts within its jurisdiction.  

If I understand Michael correctly, he denies that religiously-motivated behavior can be divided into two categories. He posits that there is only one category, and that all religiously-motivated behavior falls into that one category. But he runs into a problem: if government lacks jurisdiction over the religiously-motivated behavior category, then it cannot pass legislation on anything touching someone’s religious behavior. Indeed, given that religiously-motivated behavior can inspire innumerable types of action, the religiously-motivated behavior category would seem to make lawmaking impossible. Alternatively, we would have to conclude that religiously-motivated behavior is not “inalienable,” at least as the Founders used that term.  

What this suggests, I believe, is that Michael has to accept that his category of religion (religiously-motivated behavior) falls within government’s jurisdiction. That would mean we do alienate authority over our religious exercises, but the Founders clearly denied this.  

Nonetheless, let’s play out the argument. If government can extend its jurisdiction over all religiously-motivated behavior, the Founders’ social compact theory would dictate that the legislative authority has power to determine, through its lawmaking power, which religious exercises are legal and which are illegal. This Michael finds problematic. It gives too much authority to legislative majorities and not enough protection for individuals (especially of minority religious faiths) to act according to conscience. So Michael introduces a judicial check on legislative power—strict scrutiny.  

What Michael has done, I think, is adapt the Framers’ constitutionalism to a world in which we:  ‍

a. no longer hold to the conception of inalienable natural rights (and the corresponding jurisdictional limits on government power) 

b. do not distinguish between different types of religiously-motivated behavior  

I believe this to be a different sort of constitutionalism than the Framers’ constitutionalism. I think one can call it an “adaptation” or “translation” of the Framers’ constitutionalism, but it is an adaption that, in at least one fundamental way, radically rejects the Founders. Specifically, it rejects the Founders’ notion of the right to worship according to conscience as an “unalienable” right. I don’t see how Michael’s constitutionalism can accommodate the Founders’ conception of inalienable rights.  

The first key point of difference (at least from my perspective) between Michael and me is that I hold that a right cannot be alienable and inalienable at the same time. If we grant government authority over a right, we have consented to follow the political community’s legislation (and its executive and judicial enforcement) of that right.  

The second key point of difference follows from the first: I hold that religiously-motivated behavior can be divided into the aforementioned categories.  

I grant that if religion cannot be so divided, then the Founders’ conception of religious liberty as an “inalienable” right is incoherent. Folks like Micah Schwartzman (and you too Andy [Koppelman]? I’m not sure) would say that the Founders ontology of nature was deficient—that in reality religion is just a type of deep conviction. Religion as such is not special, but deep convictions are as they are central to our identity as autonomous beings. We ought then to try to arrange a constitution so that individuals can live according to their deep convictions. There really are no “inalienable” rights, but we can devise constitutional arrangements such as the Sherbert rule that facilitate religious and non-religious individuals living according to conscience as much as possible.  

I’m not sure that I have captured Micah’s (and Andy’s?) position perfectly, but I think the position I describe and attribute to them is more than defensible. It’s not the Founders’ position, however. For reasons I set forth in Chapter Three of my book, Religious Liberty and the American Founders (RL&tAF), I think the Founders’ ontology and their conception of “inalienable” rights is rationally defensible. But even if it is rationally defensible, whether it is the best type of constitutionalism for the American people today I think is very much an open question.  

‍

Koppelman’s reply: 

‍

My objection to Michael’s point was actually pretty simple. Natural rights philosophies come in flavors, as with any philosophical framework that persists over generations. Some varieties of natural rights argument entail the kind of accommodation that Michael contemplates. Others—Locke’s, for example—do not. Therefore, to say as Michael does that Phillip’s conclusion about free exercise exemptions “seems incompatible with his central thesis that religious freedom is a natural right” is a non sequitur. It’s like saying that denying the Trinity is incompatible with being a Christian. 

‍

McConnell responded: 

‍

To be precise, I do not claim that natural law in itself entails exemptions. I claim that any exemption the regime makes must rest on considerations of the law of nature, meaning others’ equal rights and public peace and safety. Thus, the FORMAL consideration that the positive law is framed in terms that are neutral and generally applicable has nothing to do with the right’s natural law conception. 

Koppelman responded in turn: 

I am persuaded by Phillip’s account of the original meaning of free exercise. But I also find much of Michael’s political and legal vision attractive, and so support religion’s legal accommodation, though I do not support the extravagant privileging that the Supreme Court now embraces. (See my The Increasingly Dangerous Variants of the “Most-Favored-Nation” Theory of Religious Liberty, 108 Iowa L. Rev. 2237 (2023)).  

Phillip asks me if I think that religion as such is not special. I am agnostic. More importantly, I think that the state had best be agnostic on the question of whether religion is merely a deep, perhaps misguided conviction that people happen to have. The state is a poor source of guidance on that question. The state should be silent about theological questions that divide the citizenry, and religion’s ontological status is today one of those questions. The fact that religion matters so much to so many is sufficient reason for the state to treat it as special. The Supreme Court has recognized (borrowing Phillip’s language, above) “a category of religious-motivated behaviors over which we do not grant the state authority,” which includes “things like what constitutes blasphemy, how ministers are selected, the process by which one’s sins are forgiven, how water is made holy.” At least to that extent, present First Amendment law is consistent with original meaning. I’ve elaborated this description of American law in Defending American Religious Neutrality, Harvard University Press, 2013, and I’ve defended it as a normative matter in How Could Religious Liberty Be a Human Right?, 16 Int. J. Const. Law 985 (2018). On the differences between my view and that of my friend Micah, see “Religion” as a Bundle of Legal Proxies: Reply to Micah Schwartzman, 51 San Diego L. Rev. 1079 (2014). 

Phillip is right that I reject the Framers’ conception of natural rights. That conception, Phillip writes, holds that the fact of free will, the “capacities for reason and freedom,”  

suggests that, if men do have the duty to worship God, that duty can only be discharged through worship animated by conscientious conviction. The only worship fittingly expressed toward a God that created us and authored our nature is a movement of mind and heart that is freely chosen and pursued. (Religious Liberty and the American Founding, 78, 320)  

I respond, in my critique in American Political Thought, cited above, 393: “But the fact that free worship is possible does not mean that it is divinely intended. Coerced worship is also possible. The Creator generated creatures capable of either.” The natural law argument founders on Hume’s old objection that one cannot deduce a conclusion about what ought to be done from premises that only state what is the case. In the familiar phrase, you can’t derive an ought from an is. 

Phillip has responded to me by claiming that I am guilty of the same alleged fallacy. As he puts it, “Koppelman draws an ‘ought’ from an ‘is’—we should not rely on the Founders’ political philosophy (the ‘ought’) because our world is different from theirs (the ‘is’).” The Possibility and Meaning of the Natural Right of Religious Liberty: A Response, 12 American Political Thought: A Journal of Ideas, Institutions, and Culture 417, 417-18 (2023).  

I actually have two objections to the Founders’ philosophy, which Phillip conflates. The first is purely logical: Hume is right, and the argument was fallacious when it was made. The second, which relies on the contingent fact that our world is different from theirs, matters because the law of religious liberty must make sense to Americans here and now, and so can’t rely on the Christian premises that were another, independent basis for the Framers’ views. 

Phillip knows that I’m explicit about my ought, the ought that makes my “is” pertinent, because he quotes me: “when courts ‘interpret the First Amendment’s religion clauses, they must articulate a rationale that will not be unintelligible or repulsive to many citizens.’” (Id., 418) Phillip dismisses the “many citizens” as including only “Rawlsian liberals and other members of the laptop class.” (Id.) In fact, as I note in my earlier critique (394), “Nearly a quarter of Americans now self-identify as having no religion at all.” Ought and is thus combine to make the Framers’ conception untenable. The broad American consensus in favor of religious liberty has now collapsed. Those of us who hope to restore it must think about whom we are trying to persuade.  

Michael nicely states the better argument for religious accommodation. Free exercise’s purpose is “protecting pluralism—the right of individuals and institutions to be different, to teach different doctrines, to dissent from dominant cultural norms and to practice what they preach.” (On Religion, the Supreme Court Protects the Right to Be Different, N.Y. Times, July 9, 2020.) That way of putting it is nicely tailored to the Times’s very secular readership. I made a similar argument in Chapter 7 of Gay Rights vs. Religious Liberty? The Unnecessary Conflict, Oxford University Press, 2020. But none of the Framers thought of accommodation this way. In lawmaking, the protection of pluralism legitimately competes with other considerations. Hence strict scrutiny, a device unknown to the Framers (which is why Justice Alito’s defense of strict scrutiny is phony originalism. See my Justice Alito, Originalism, and the Aztecs, 54 Loyola U. Chicago L.J. 455 (2023), which relies heavily on Phillip’s work. 

McConnell responded to this, and Muñoz concluded the conversation. Again, Muñoz responded to McConnell’s points seriatim. Here is McConnell’s response, with Muñoz’s interstitial comments noted in italics: 

[McConnell begins] 

Once again, I will try to break the line of argument into small pieces and identify where Muñoz and I agree, where we disagree, and why. I continue to believe that something like the Sherbert/Yoder test best captures our shared understanding of the natural right of religious free exercise. It is the consequences they have for other persons and the public peace that bounds free exercise rights and not the form in which the legislature drafts the positive law (which is Smith’s holding). 

1. Muñoz and I agree that some natural rights are alienable and some are inalienable. I think we agree on the following points, though I am not sure about point c).  

a. Alienable rights are those that we may relinquish to civil society at the social compact’s formation. An example is the right to sole use of the product of our labors. We relinquish that right in part to the state for taxation and use in the common interest. There are scores of others. Which alienable rights to relinquish and which to retain is a matter for constitutional choice. A liberal society is one in which most natural rights are retained. A statist society in one in which most natural rights are relinquished. 

b. Inalienable rights are those that either cannot or must not be relinquished. As stated in the Virginia Declaration of Rights: “all men . . . have certain inherent rights, of which, when they enter into a state of society, they cannot, by any compact, deprive or divest their posterity.” 

c. Being inalienable does not mean the right is unbounded or absolute. The law of nature bounds inalienable rights like it bounds all others. That primarily means they must respect other persons’ equal natural rights and the public peace.  

[Muñoz responds: I agree. I make the same point in RL&tAF, pp. 59-67.]  

2. The right to exercise religion in accordance with conscience was the most universally agreed-upon inalienable natural right. The Virginia Declaration of Rights includes that right’s best Founding Era definition: “Religion, or the duty we owe to our Creator, and the manner of discharging it, can be directed only by reason and conviction, not by force or violence: and therefore, all men are equally entitled to the free exercise of religion, according to the dictates of conscience.” I assume Muñoz agrees with this, as both a historical and a theoretical manner. 

[Muñoz responds: Yes, I agree. Chapter 1 of RL&tAF attempts to provide the documentation for Michael’s #2.]  

3. My disagreement with Muñoz begins with his division of religious exercise into what he calls two “distinct categories”—namely, “religious interests” and “religious worship.” 

a. The former category (“religious interests”) includes the vast majority of religious actions that the United States legal system regards as free exercise rights. Muñoz mentions three examples: “objections to taking up arms, the education of one’s children, the use of hallucinogenic substances.” Statute has protected the first since before the Founding, the second is Yoder, and the third is Smith. 

b. The latter category (“religious worship”) is ill-defined, but “narrow,” by Muñoz’s description.  

[Muñoz responds: I agree with McConnell that I have left this “ill-defined.” I’ve been thinking about this point quite a bit since Michael’s initial review of my book (and before, too, I suppose). The category of “religious worship” would include those matters pertaining to religion that are above reason but not contrary to reason. In Aquinas’s schema, my category of “religious worship” is that which pertains to the divine law but is not, properly speaking, part of the natural law. See ST, I-II, Q.91.4. See also Locke’s An Essay Concerning Human Understanding, IV.18 (esp. 5,7). 

[McConnell continues] Muñoz writes: 

“Religious ‘worship’ is the set of religiously-motivated behaviors over which the state lacks authority. As you know from my book, I don’t try to provide an exhaustive list of such actions but they would include things like what constitutes blasphemy, how ministers are selected, the process by which one’s sins are forgiven, how water is made holy.” 

c. According to Muñoz, mere “religious interests” enjoy no constitutional protection. None.   

[Muñoz responds: My response to this point is the same as Hamilton’s response to the Anti-federalist’s complaint that the proposed Constitution lacked a bill of rights. Hamilton said—and I say here in response to McConnell—the Constitution itself is a bill of rights. The Constitution’s structures (representation, bicameral legislature, separation of powers, federalism)—are all meant to protect what I have labelled “religious interests.” To say that religious interests “enjoy no constitutional protection. None.” if the judiciary cannot grant exemptions to otherwise valid laws that burden religious individuals, is to associate the judiciary and judicial review (and not the other constitutional structures) with “constitutional protection.” For what it’s worth, this exact point is the theme of my article, “James Madison’s Political Science of Religious Liberty.”] 

[McConnell continues] He does not even acknowledge Smith-style protection against non-neutral laws. Muñoz writes: 

“We do not retain authority over this category in political society. As such, we give the political community authority to make laws over these subjects (e.g., you must bear arms when drafted, you must send your child to school until the age of 16, you may not use certain hallucinogenic substances). When we consent to being members of the political community, we consent to following such laws. . . . . [W]hen such laws are passed, and assuming they lie within the jurisdiction we have granted to government, we are obliged to obey them. There is no right of ‘conscience’ to not obey a duly-enacted law, in the Founders’ social compact constitutionalism.” 

4. I disagree both with the categories and with Muñoz’s treatment of each. 

a. The First Amendment refers to the “free exercise of religion,” without dividing it into two sub-categories. The legal term, “free exercise of religion” comes from (among other sources) the Virginia Declaration of Rights, and extends to all of the “duties we owe to our Creator.” These go well beyond worship. Most of the Ten Commandments are not about worship. To be sure, some of the state constitutions confined their protections to religious worship, but the First Amendment’s framers did not.  

b. Even religious worship is subject to regulation in conformity to the law of nature. It is not exempt from state authority or outside the state’s jurisdiction. In the seventeenth- and eighteenth-century debates, the most common example of free exercise’s obvious limit was the prohibition on human sacrifice—yet this seems to fall within Muñoz’s category of “worship.”  

[Muñoz responds: This is mistaken. The law of nature bounds the right of religious free exercise likes it bounds all natural rights (alienable or inalienable). So the Founders never understood a part of the right of religious free exercise to include human sacrifice, being that it is against the law of nature (just as libel was never understood to be part of the right of free speech).] 

More to the point, during COVID the state was entitled to restrict worship services insofar as this was necessary to avoid spreading the disease to outsiders. (I am not, of course, endorsing the actual regulations some states imposed.) 

[Muñoz responds: Yes and no. During COVID, state authorities certainly could make safety regulations that would cover religious institutions including (indirectly) how those institutions would conduct worship services. This is no different from how the state can require and maintain that physical churches must comply with building standards. But a state lacks authority to tell a church where it must place the alter or which way the congregation must face during worship. Similarly, the state cannot dictate how communion as such must be distributed during religious worship. The governor of Washington levied such a directive, regulating the administration of communion as such, during COVID. And it was, I think, unconstitutional. See RL&tAF, p. 260, n.10]  

c. The contours of these categories are mystifying.  

[Muñoz responds: The categories, again, follow from Aquinas’s and Locke’s epistemology. The key distinctions are that which is (a) contrary to reason, (b) according to reason, and (c) above reason but not contrary to reason. That which lies within (c) is what is “inalienable.”] 

[McConnell continues] 

Muñoz says that the use of hallucinogenic substances falls in the category of “religious interests,” but for Native American Church members peyote use is an integral part of worship. Why does it not fall within Muñoz’s second category? The first published free exercise case involved whether a Catholic priest could be required to testify in court to information he learned in the confessional. Should that case be understood to be about everyone’s civic obligation to give testimony (much like military service) or about interfering with the sacrament of confession and reconciliation, which is confidential by its nature? Does a Muslim schoolchild have the right to absent himself from class for the five minutes necessary to perform the required mid-day prayer? What side of the worship/interest line do these cases fall on? Why does it matter? 

5. Muñoz describes his view as both “narrow” and “deep,” in contrast to my “broad” and “shallow” conception. His view is “narrow” because he protects only worship, as I have discussed in Point 3. This is a clear point of disagreement between us. His view is “deep” because, he says, this narrow category of inalienable rights is not subject at all to governmental authority. Muñoz writes:  

“I hold that there is a category of religious-motivated behaviors over which we do not grant the state authority. The existence of such a category is what the Founders meant by the “unalienable right to worship according to conscience.” Upon creating the social compact, we do not grant the state authority over the category of our religious worship. We cannot grant the state authority over it as it is inalienable. The inalienable character of the right, accordingly, translates into a jurisdictional limitation on state power.” 

[McConnell continues] 

a. This, too, is a clear point of disagreement between us: I deny that there is any sub-set of the natural right of religious exercise that is unbounded.  

[Muñoz responds: This is mistaken interpretation of my position. The law of nature bounds all natural rights. There is no such thing as an unbounded natural right. The question is one of jurisdiction: does the state have jurisdiction to specify what constitutes true religious worship in matters that are above but not contrary to reason (e.g. God is a trinity, the Eucharist contains the real presence of Christ’s body and blood). I hold that the state lacks jurisdiction over these matters and cannot legislate accordingly.  

‍I also hold, as discussed above, that the lack of authority over religious matters that are above but not contrary to worship entails that the state should not regulate such matters directly (including the distribution of communion) even for otherwise legitimate purposes, such as health and safety.] 

[McConnell continues] 

b. The law of nature bounds the entire right, including the right of worship. This mostly means that the need to respect others’ equal natural rights and the public peace. This limit on the right’s scope does not go away when we enter civil society, even though we cannot and do not alienate the right.  

[Muñoz responds: I agree that the scope of natural rights does not go away when we enter civil society. But entering civil society entails that we grant the legislative branch authority to make determinations about how to best secure this right. See my point above about what alienating a natural right means. This may be one of the key differences between McConnell and me. We both agree that state authorities have jurisdiction over what I call “religious interests.” I hold that we give the legislature that authority, but we do not ignore the ways that the separation of powers operates in our Constitution (e.g. executive veto, the judiciary’s rational-basis review). McConnell, I believe, holds that the judiciary’s role in overseeing legislation is more expansive than I hold it to be.] 

c. When we enter civil society, we give up the right to be the judge of the extent of our rights. To the legislature we give the power to makes laws defining where our rights end and other people’s begin; to the judiciary we give the power to apply those principles in particular cases; and to the executive we give the power to use force to enforce those determinations. This is true of all retained natural rights, both alienable and inalienable.  

[Muñoz responds: I don’t think this is correct when it comes to inalienable natural rights. When it comes to inalienable natural rights, the judiciary’s role is to determine whether the legislature, in fact, possesses the authority to legislate on the matter on which it legislated. It is necessarily a yes or no question. Matthew Alder makes this same point as does Nicholas Rosenkranz in his articles on the Constitution’s subjects and objects.ⁱⁱⁱ] 

d. When our Founders constitutionalized the natural right of religious exercise, they embedded in our highest law the principle that the legislature cannot abridge religious free exercise, as the law of nature defines and bounds it (“Congress shall make no law…”). But it remained true that the judiciary has power to determine in particular cases where one person’s free exercise right ends and other’s equal natural rights and the public peace begin.  

[Muñoz responds: The constitutional remedy in a case where Congress made a law that prohibited the free exercise of religion would be that the law is null and void, because Congress “shall make no law.” The constitutional violation occurs at the point of lawmaking and the remedy would have to undo the lawmaking. Rosenkranz makes this point more clearly than I have.] 

[McConnell continues] 

Moreover, the legislature is entitled to enact laws that define and declare rights’ practical boundaries. In our constitutional system, these laws are subject to judicial review to ensure that they have not gone beyond a reasonable instantiation of the law of nature and abridged the right itself. 

6. Muñoz thus creates a stark dichotomy between his “two distinct categories” of religious exercise. For the “narrow” category of religious “worship,” the right is inalienable and therefore absolute; he posits that the government has no jurisdiction to legislate, adjudicate, or execute law that touches an inalienable right. But for the “innumerable” category of “religiously-motivated behavior,” there is no constitutional protection at all. This natural right was alienable and it was alienated, in toto.  

[Muñoz responds: Again, “no constitutional protection” as used here equates to no judicial review using a tool like strict scrutiny.] 

[McConnell continues] 

This is a point of clear disagreement. I deny that there are two sub-categories. Religious free exercise includes all duties we owe to the Creator, according to the dictates of conscience. Muñoz terms this “broad.” But others’ equal rights and the public peace bounds civil society’s free exercise right, like it does the state of nature’s free exercise right. Civil society defines and enforces these bounds; they are not left to individual’s self-interested judgments.  

[Muñoz responds: I agree, but I hold that the legislature is given the power to make judgments over our natural right to religious liberty’s alienated aspect.] 

[McConnell continues] 

This conception is “shallow” by comparison to Muñoz’s notion of inalienable rights that are outside civil society’s jurisdiction, but it does not seem shallow in any realistic sense. It means that the government does not have the rightful authority to limit our freedom of religious exercise except in service of others’ natural rights and the public peace, and that the judiciary will enforce this limitation.  

7. Muñoz and I seem to agree about our disagreement’s theoretical ground, but in my opinion, he misunderstands the implications. He believes that I am caught in a dilemma—that my position either denies the legislature the power to legislate with respect to religious exercise or I have denied religious liberty’s inalienable status.  

a. This is his description of my supposed dilemma’s first prong:  

He [Muñoz] argues: If I understand Michael correctly, he denies that religiously-motivated behavior can be divided into two categories, and he posits that there is only one category, and that category consists of all religious-motivated behavior. But here, I think, Michael runs into a problem. If government lacks jurisdiction over the category of religiously-motivated behavior, then it cannot pass legislation on anything touching someone’s religious behavior. Indeed, given that innumerable types of action can emanate from religious motivations, the category of religiously-motivated behavior would seem to make lawmaking impossible. 

Not so. Under my view, unlike his, the legislature does not lack “jurisdiction” over the religiously-motivated behavior. (In fact, the first Congress briefly adopted, and then rejected, a version of the First Amendment that would have denied Congress all power over laws “touching religion.”) There is nothing wrong with the legislature passing laws to define contending parties’ relative rights as they affect religion—requiring employers to provide “reasonable accommodations” for their workers’ religious needs, for example, setting up systems for evaluating claims for religious conscientious exemptions from military service, adjudicating divorced families’ religious issues arising from child custody, requiring respect for indigenous sacred sites in federal construction projects, hiring military chaplains, figuring out when a teacher’s right of religious exercise infringes on students’ rights not to be coerced, etc. What the legislature may not do, however, is abridge religious exercise rights in the service of regulatory or religious ends other than the protection of others’ natural rights and the public peace. Fortunately, in this liberal (in a good sense) political culture, most laws are consistent with natural right. 

b. Muñoz describes the second prong of my supposed dilemma as follows: 

[Muñoz] Alternatively, we would have to conclude that religiously-motivated behavior is not “inalienable,” at least as the Founders used that term.  

What this suggests, I believe, is that Michael has to accept that his category of religion (religiously-motivated behavior) falls within government’s jurisdiction. That would mean we do alienate authority over our religious exercises. But the Founders clearly denied this.  

In my opinion, this claim misunderstands inalienability. A right is “inalienable” if it cannot be relinquished. That does not make such rights unbounded or deprive civil society of authority to define and enforce those bounds. The Founders believed that one of the main reasons people enter a social compact is to secure the natural rights that are insecure in the state of nature. They are insecure because they lack definition, an impartial arbiter, and effective enforcement. When we enter civil society, we collectively create a legislature which, among other things, passes laws defining the bounds of people’s rights, a judiciary to make impartial determinations when claims of right conflict, and an executive to use the society’s collective force to enforce the rights against aggressors. Inalienable rights receive the benefit of these civil society features no less than any others. This is a clear point of disagreement: I do not believe that retained natural rights (whether or not inalienable) are outside “government’s jurisdiction.”  Government protects them. 

[Muñoz responds: Michael here identifies perhaps our most fundamental disagreement. What he describes is my understanding of alienable natural rights (and he provides a very nice summary of that position). I hold that “inalienable rights,” are different. An example of an inalienable right would be the right to revolution. Government cannot secure such a right and government has no authority to regulate (i.e. to make regular so we can enjoy) our right to revolution. The nature of the right makes it inalienable.] 

8. What does any of this have to say for the First Amendment’s constitutional law?  

a. Although many details of “strict scrutiny” as modern courts have crafted it are of dubious validity, the original understanding bears a strong resemblance to the Sherbert/Yoder approach to free exercise. In turn, this understanding has been adopted in the Religious Freedom Restoration Act, the Religious Land Use and Institutionalized Persons Act, and by other legal authorities. The right’s effects on other people and the public peace limit it, and not the form in which the legislature has framed its law. The free exercise right’s natural bounds are other’s equal natural rights and the public peace.  

b. A more accurate translation of the original meaning would abandon the too-vague vocabulary of “compelling governmental interests,” and replace it with the language of respecting other’s natural rights and the public peace. The precise formulation of the law of nature limits on the free exercise right’s scope is less important than the right’s structure. 

c. The Smith approach to free exercise, by contrast, is entirely oriented to the form of the positive law that is said to conflict with religious exercise. If the law is framed in neutral and generally applicable terms, it passes constitutional muster no matter whether it protects other’s natural rights or the public peace or serves some other governmental end. But in the state of nature, the form the positive law takes could not possibly have bounded the right, because there is no positive law in the state of nature. The Sherbert/Yoder approach, or something like it, carries forth the content of the natural right, including the limitations the law of nature imposes. The Smith approach has nothing to do with the natural right but seems to be based on separation of powers concerns and a jurisprudential theory of how laws of all sorts should be framed.  

d. In his book, Muñoz argues (and I agree) that the First Amendment right of religious free exercise is the natural right of freedom of religion’s constitutional embodiment. If that is so, he ought to agree with me that Sherbert/Yoder is more faithful to the original meaning than Smith.  

9. In any event, Muñoz supplies no justification for the Smith rule.  

[Muñoz responds: Michael is correct here. My approach and Smith both deny that the Free Exercise Clause provides a right of exemption, but I don’t champion Smith as such.]  

According to Muñoz, rights of worship are absolute—which Smith denies—while mere “religious interests” (i.e., most free exercise claims) are unprotected. Even Smith is a form of protection. Indeed, the aggressive interpretations of neutrality and general applicability in the Court’s recent cases are extraordinarily protective. Muñoz does not explain why, if mere “religious interests” are consigned to legislative judgment at the time of the social compact, the legislature is required to treat religious exercise no less favorably than comparable secular conduct, which is the Smith rule.  

10. In short, Muñoz gives an elegant and accurate account of the natural rights theory behind the First Amendment, but stumbles when he attempts to translate that into constitutional doctrine.  

Endnotes

(1) The Smith rule, established in the 1990 Supreme Court case Employment Division v. Smith, holds that generally applicable laws can limit religious activity if they are neutral and not based in religious animosity.

(2) The Sherbert test is a tool the Supreme Court uses to determine if government actions violate the Free Exercise Clause. It looks first at whether government policy has imposed a substantial burden on a person’s right to practice religion and second, if a burden exists, whether a compelling state interest justifies it. In Wisconsin v. Yoder, the Supreme Court prioritized free exercise of religion over the state’s interest in an educated populace. The Court held that individuals’ religious liberty to educate their children as they see fit outweighed the states’ interest in forcing children to attend school.

(3) See https://scholarship.law.duke.edu/faculty_scholarship/2600/, https://www.stanfordlawreview.org/print/article/the-subjects-of-the-constitution/, https://www.stanfordlawreview.org/print/article/the-objects-of-the-constitution/

‍

This concluded our conversation. It is our hope that publishing this exchange will help scholars of religious liberty understand, and perhaps even adjudicate, the differences among the three of us on the Free Exercise Clause’s meaning. 

‍

‍Andrew Koppelman is the John Paul Stevens Professor of Law at Northwestern University. 

Michael McConnell is the Richard and Frances Mallery Professor and director of the Constitutional Law Center at Stanford Law School, and a senior fellow at the Hoover Institution.  

Vincent Phillip Muñoz is the Tocqueville Professor of Political Science and a concurrent professor of law at the University of Notre Dame. He is also a Distinguished Visiting Professor at the School of Civic Leadership at the University of Texas at Austin.