Text of the Free Exercise clause as a measure of Employment Division v. Smith and the Religious Freedom Restoration Act, The

text of the Free Exercise clause as a measure of Employment Division v. Smith and the Religious Freedom Restoration Act, The

Ides, Allan

Congress adopted the Religious Freedom Restoration Act of 1993(1) (RFRA) to overturn the decision of the United States Supreme Court in Employment Division v. Smith.(2) In Smith, the Court declined to mandate a constitutional exemption from the State of Oregon’s drug laws for two members of the Native American Church who had ingested peyote as part of a religious ceremony.(3) More generally, the Court held that the Free Exercise Clause provided religiously motivated actors no special exemption from the reach of laws of general applicability.(4) Although religious beliefs were absolutely protected by the First Amendment, religiously motivated conduct was not. Laws of general applicability that adversely affected such conduct were subject to neither more nor less judicial scrutiny than laws that adversely affected conduct of a nonreligious nature.

As a consequence of Smith, a law that indirectly, although perhaps substantially, affected a religious practice would be examined under the lowest level of scrutiny, that is to say, virtually none at all. More intense judicial review was reserved for those rare instances in which a law regulated conduct because of, and not merely in spite of, that conduct’s religious nature.(5) Smith did not, however, announce the complete demise of religious exemptions from otherwise neutral laws. The Smith Court made clear that its holding did not preclude a legislature from creating an exemption for a religious practice; the Court did hold, however, that such exemptions were not constitutionally required.(6)

The response to the Smith decision was intense and highly negative.(7) And although some of that intensity may have been attributable to what was perceived as the Court’s insensitivity to a minority religion, the critique of Smith transcended the specific facts of the case. The trouble with Smith, according to its critics, was the doctrine it created, or, perhaps more to the point, the doctrine it destroyed. Prior to Smith, and most notably in Sherbert v. Verner(8) and Wisconsin v. Yoder,(9) the Court had, at least under some circumstances, applied a compelling state interest test to laws of general applicability that either indirectly affected or more directly burdened religiously motivated conduct.

Thus in Sherbert, a Sabbatarian was entitled to receive unemployment benefits even though her unemployment resulted only from her refusal to work on Saturdays as required of all other citizens; the otherwise neutral unemployment law substantially and unlawfully burdened the practice of her chosen religion.(10) Similarly, in Yoder, Amish parents were granted a special privilege to withdraw their children from school after the completion of the eighth grade since, according to the Amish, further education would undermine the religious values and salvation of the Amish community.(11) Although the Smith opinion rather inartfully fudged the point, in essence, the Smith Court jettisoned the reasoning and doctrine of Sherbert and Yoder. No longer would courts use the compelling state interest test to examine the constitutionality of a law of general applicability under the Free Exercise Clause, regardless of that law’s impact on religious practices.(12)

The RFRA was expressly designed to return free exercise claims to their perceived legal status prior to Smith,(13) that is, to affirm the doctrinal legitimacy of Sherbert and Yoder, and, as RFRA’s proponents would have it, to return religious liberty to the United States.(14) To this end, section 3 of the RFRA provides that the government–broadly defined to cover all agencies of federal, state and local government–“shall not substantially burden a person’s exercise of religion even if the burden results from a rule of general applicability.”(15) In short, the RFRA creates a blanket, generic religious exemption. The only exception, which parallels the doctrinal jurisprudence of Sherbert and Yoder, states: “[government may substantially burden a person’s exercise of religion only if it demonstrates that application of the burden to the person–(l) is in furtherance of a compelling governmental interest; and (2) is the least restrictive means of furthering that compelling governmental interest.”(16)

Of course, the two key assumptions of the RFRA are that Smith was wrongly decided as a matter of constitutional law–indeed, obviously and tragically wrong might be a more apropos description–and that a legislative reassertion of past doctrine and the accompanying doctrinal tests will provide a proper and effective cure for the Court’s flawed interpretation of the Constitution. The latter assumption strikes me as founded upon a somewhat naive and unreal vision of the meaning, effectiveness, and function of doctrine, but I will not dwell on that point.(17) I am more curious about the first assumption, namely, that Smith was wrongly decided. I am not sure what it means when one says that Smith was wrongly decided. Wrong under what standard? I am sure that the critics of Smith assume that the Warren and Burger Court interpretations of the Free Exercise Clause were somehow correct or better, though I think “familiar” would be a more fitting term. Indeed, Smith is undoubtedly quite inconsistent with the Warren and Burger Court decisions, Justice Scalia’s protestations to the contrary notwithstanding.(18) But by what measure is the decision wrong as opposed to merely reflecting a different policy judgment about the scope of the protection afforded by the Free Exercise Clause? Similarly, by what measure does the RFRA represent a correct or incorrect interpretation of the Free Exercise Clause?

One possible measure to both questions is the language of the Free Exercise Clause. What does that language reveal to us? Or better yet, what would it have meant to an eighteenth century reader? It would seem–at least if one assumes some relevance to the constitutional text–that a close examination of the text would be appropriate and perhaps even somewhat enlightening. In any event, this Essay will pursue that inquiry as an attempt to see which interpretation of the Free Exercise Clause, the Smith version or the RFRA version, most closely resembles any meanings that can be discovered in the text.

The text of the First Amendment states, “Congress shall make no law respecting an establishment of religion, or prohibiting the free exercise thereof.”(19) In the present context, the focus is on the latter phrase, the so-called Free Exercise Clause. Thus, “Congress shall make no law prohibiting the free exercise of religion.” In briefly pursuing this examination of text, and in particular the last six words of the text, I will proceed from the fictional perspective of a reasonably literate and informed lay reader of the eighteenth century (created, of course, by a reasonably literate and informed professional reader of the twentieth century). My reader is also curious, open-minded and blessed with a perspicacity somewhat more intense than the casual word browser. The basic goal is to consider those possible interpretations our fictional reader might assign to the Free Exercise Clause. Of course, we cannot expect to know with certainty which meanings will be assigned by our reader. With any luck, however, we should achieve a fair approximation of the possibilities. For reasons that should become apparent, I will take the words in reverse order.


We begin with the word “religion.” What is religion? Just to ask the question hints at the complexity and potential futility of the inquiry. This question has and continues to perplex philosophers, theologians, and legal scholars. But it need not perplex us, nor would it perplex our reader. Before succumbing to a potential cascade of philosophic, theologic, and jurisprudential possibilities, let us consider a simpler approach. What, in common parlance, would be understood by the use of the word “religion” in the Bill of Rights? After all, the Bill of Rights was not designed as a preface to philosophic, theologic, or even jurisprudential inquiries. It was the product of pragmatic and political judgment, meant to be read, used, and presumably understood by a broad spectrum of people, including such persons as our reader.

Doctor Samuel Johnson’s dictionary of the English language–a dictionary familiar to readers of the late eighteenth century–provides two definitions of religion: “1. Virtue, as founded upon reverence of God, and expectation of future rewards and punishments…. 2. A system of divine faith and worship as opposite to others.”(20) The examples given under the first definition draw a distinction between duty toward men and duty toward God. Only the latter is properly considered religion. Thus, “[r]eligion or virtue, in a large sense, includes duty to God and our neighbour; but in a proper sense, virtue signifies duty towards men, and religion duty to God.”(21) The second definition refers to specific systems of belief and worship, defining what we would now refer to as religious sects, as in the Catholic religion or the Jewish religion. Taken together, according to Doctor Johnson, religion has a behavioral aspect connoting virtuous duty to God or the divine, and an organizational aspect describing specific systems of divine belief and worship.

The definitions in the Oxford English Dictionary (OED), although somewhat more complex, appear to cover the same basic spectrum of ideas found in Doctor Johnson’s dictionary. Thus, the OED defines religion as “a particular system of faith and worship” and as “recognition on the part of man of some higher unseen power as having control of his destiny, and as being entitled to obedience, reverence, and worship; the general mental and moral attitude resulting from this belief, with reference to its effect upon the individual or the community; personal or general acceptance of this feeling as a standard of spiritual and practical life.”(22) Another definition indicates a somewhat secular usage–“devotion to some principle; strict fidelity or faithfulness; conscientiousness; pious affection or attachment.”(23) This secular usage, however, seems to derive its force and meaning by suggesting an analogy to spiritual practices and beliefs: “An old Word is retained by an Antiquary with as much Religion as a Relick.”(24)

Webster’s Third New International Dictionary–bringing us well into the twentieth century–provides several usages, most of which roughly parallel Doctor Johnson’s definitions:

[T]he personal commitment to and serving of God or a god with worshipful devotion, conduct in accord with divine commands esp. as found in accepted sacred writings or declared by authoritative teachers, a way of life recognized as incumbent on true believers, and typically the relating of oneself to an organized body of believers … the state of a religious … one of the systems of faith and worship … the body of institutionalized expressions of sacred beliefs, observances, and social practices found within a given cultural context … the profession or practice of religious beliefs … religious observances … a personal awareness or conviction of the existence of a supreme being or of supernatural powers or influences controlling one’s own, humanity’s, or all nature’s destiny … the access of such an awareness or conviction accompanied by or arousing reverence, gratitude, humility, the will to obey an serve … religious experience or insight.(25)

Two definitions in Webster’s have a more secular tone: “[S]crupulous conformity [to] … a cause, principle, or system of tenets held with ardor, devotion, conscientiousness, and faith … a value held to be of supreme importance.”(26) But again, used in this context the word “religion” seems not to be used as a separate construct but as a creative analogy to the spiritual order, as in, “Baseball was his religion.”

In sum, in both the eighteenth century and in modern contexts, the word “religion” conveys two possible meanings relating to the divine or spiritual. The first describing a way of life inspired by the divine and the second describing an organized system of beliefs and practices similarly inspired. A third possible usage is the secular meaning suggested by both the OED and Webster’s, namely, religion as any system of deeply held values. This third usage is not irrelevant from a modern perspective. Certainly arguments have been made that secular moral beliefs are entitled to the same constitutional respect granted divinely inspired beliefs. Indeed, this is one of the important questions implicitly posed by the Smith decision as well as the Sherbert and Yoder line of cases. Would our reader have understood the word “religion” to convey this secular meaning in the context of the Bill of Rights? I think not.

In the first place, the secular usage reflects a more creative spin on the idea of religion than it does a separate construct. A person who holds secular ideas firmly may be said to hold them with a religious fervor; indeed, those ideas may even be called that person’s religion; but it does not follow that we would view that person’s ideas as a religion in the commonly used sense of the word. We simply mean to establish a strong analogy. Similarly, a committed Marxist may share certain traits with a religious zealot, but only by way of analogy is the Marxist referred to as religious.

Next, the actual placement of the word “religion” in the First Amendment is in the phrase “law respecting an establishment of religion.” Without fully pursuing a linguistic study of the Establishment Clause, it seems quite likely that the use of the word “religion” within that phrase was not meant in the secular sense, but in the divine sense, in other words, as pertaining to divinely inspired institutions. Most importantly, the word “establishment” was commonly used to refer to state-supported religious institutions of the divinely inspired sort. An eighteenth century reader would have readily picked up on that meaning, and would not have been confused by any potential secular usage, analogous, philosophic, or otherwise.

If this brief assessment of the “establishment of religion” is correct, and my reader and I are both confident it is, one would think that the subsequent phrase “the free exercise thereof” would refer to the same type of usage, namely, religion as divinely inspired. At least I expect that is what a reader of the phrase would likely infer from the juxtaposition of the establishment and free exercise clauses, if not simply from the common usage of the. word religion as relating to divinely inspired beliefs, practices, and organizations. In short, our reader would most likely understand the word religion to encompass the two definitions provided by Doctor Johnson.


How then would our reader interpret the phrase “exercise of religion?” Doctor Johnson provides nine definitions for the noun exercise:

1. Labour of the body; labour considered as conducive to the cure or prevention of diseases…. 2. Something done for amusement…. 3. Habitual action by which the body is formed to gracefulness, air and agility…. 4. Preparatory practice in order to skill: as, the exercise of soldiers…. 5. Use; actual application of any thing…. 6. Practice; outward performance…. 7. Employment. . . . 8. Talk; that which one is appointed to perform. . . .

9. Act of divine worship whether public or private.(27)

Conjoined with religion, definitions one through four, as well as definition eight, do not seem to work in any sensible fashion within the phrase “exercise of religion.” Definition five is a possibility, though the phrase “use of religion” seems somewhat peculiar and more than slightly ambiguous. Definition six, which renders “the practice or outward performance of religion,” seems closer to the mark. If this is the meaning our reader assigns, then the exercise of religion could be seen as the outward performance of duties derived from one’s religion. A Quaker’s refusal to fight might, therefore, be an exercise of religion within this definition. The ninth definition directly refers to divine worship and would seem particularly apt. If this is our reader’s definition of exercise, then clearly the exercise of religion refers to acts of divine worship inspired by one’s religion. One immediately thinks of church services, hymns, prayers, and, of course, the ingestion of peyote.

The OED definitions include the general connotations–“the habitual carrying out (of any particular kind of conduct)”(28)–as well as the explicitly religious :

The practice or performance of rites and ceremonies, worship, etc.; the right or permission to celebrate the observances (of a religion). . . . A religious observance. . . . An act of public worship. An act of preaching or prophesying; a discourse.’. . . The discussion of a passage of Scripture; a meeting of the Presbytery for holding such a discussion.(29)

Webster’s is in full accord.(30)

Given the foregoing, it would seem that the exercise of religion surely includes ritual acts of divine worship such as praying or the performance of religious ceremonies. This is true regardless of whether our reader adopts the general definition of exercise (practice, performance, conduct, and so forth) or the specifically religious usage (performance of religious ceremonies or rites, acts of worship, acts of observance). The more general definition embraces the specific religious usage when used in conjunction with the prepositional phrase “of religion.” An act of worship, for example, is a practice (or exercise) of religion. And, of course, the specifically religious definitions of exercise expressly refer to acts of worship, the performance of religious ceremonies, and so forth.

That is not to say that a choice between the general and specific definitions of exercise makes no difference. One could argue that the specifically religious connotations of the word exercise, referring to acts of worship or the performance of religious ceremonies, ought to trump the more general usage of the word. The assumption is that there is another category of religiously motivated conduct that we would not characterize as either worship or ritual. Thus recital of the Lord’s Prayer would be an exercise of religion, but an act of benevolence motivated by one’s belief in the divinely imposed obligations of Christian charity might not. Similarly, the ritual use of peyote would be an exercise of religion (as in Smith), but a parent’s refusal to send a child to school beyond the eighth grade would not (as in Yoder)–a direct reversal of what the Court actually did.

The question is, would our eighteenth century reader have drawn such a distinction? It is certainly possible that the phrase “exercise of religion” had, by the late eighteenth century, developed a precise usage that referred only to acts of worship and ceremony. Even if that usage was not exclusive, it was certainly common; the numerous pre-nineteenth century examples in the OED suggest as much.(31) On the other hand, the distinction drawn between acts of worship and other religiously motivated conduct may not be a distinction that our reader would even contemplate. If our reader were a religious person, for example, he or she might assume that all conduct taken in accord with the will of God is a form of worship and, hence, the exercise of one’s religion. To lead a Christian life would be, therefore, a form of worship. Certainly the language permits this construction. Similarly, I expect a Quaker would see the refusal to bear arms as a form of worship. In addition, some of the “religious” usages of the word exercise do not fit precisely into the worship mode: preaching, prophesying, discussing scripture.

Of course, without more information it is not possible to know which definition our reader would adopt, at least not when the phrase is stated in the abstract. In fact, it is difficult to make an educated guess at this point. From our reader’s perspective, the phrase “exercise of religion” may refer only to acts of worship, ceremony, and the like–at least what those terms were commonly understood to signify–or it may refer more generally to all conduct taken in accord with the principles of a religion. Most likely our reader would not have agonized over these words, but would have read them as referring generally to the practice of religion, without parsing through the possibilities of what that meant. We are left with a significant ambiguity to which we will return.


I will not dwell on the adjective “free.” Although Doctor Johnson provides several definitions, the one most sensibly apropos is the second: “Uncompelled; unrestrained.”(32) This is certainly the usage with which we have become familiar and in the absence of evidence to the contrary it is probably safe to assume that our eighteenth century reader would have assigned this or a similar meaning to the word, at least as used within the context of the Bill of Rights. The OED and Webster’s provide no contrary suggestions. Despite the potential libertarian breadth of this usage, I would not lightly assume that the word “free” connotes an absolute license. A better assumption, taken from our reader’s perspective, would be that freedom within a civil society is inherently limited by the mutual freedom and safety of one’s fellow citizens.

With the foregoing understanding and adding the word “free” to the results of the previous discussion, it would seem that the “free exercise of religion” can be fairly translated as either uncompelled and unrestrained worship of the divine or as uncompelled and unrestrained conduct motivated by divinely inspired beliefs, both within the context of the legitimate standards of a civil society.


This brings us to the word “prohibiting.” How would our reader interpret the word “prohibiting,” within the phrase “prohibiting the free exercise of religion?” The word “prohibit” has two somewhat different usages, both of which are at least potentially relevant to the constitutional text and each of which could send the proscription of that text in slightly different directions. According to Doctor Johnson, the word prohibit means: “1. To forbid; to interdict by authority…. 2. To debar; to hinder.”(33) The first definition suggests an exercise of authority to preclude the doing of a certain thing, as in “the law prohibits speeds in excess of sixty-five miles per hour.” The second definition suggests physical impossibility, as in “the fire prohibited our attempts at rescue.” The OED confirms these understandings by providing parallel, but slightly more specific definitions: “1. To forbid (an action or thing) by or as a command or statute; to interdict…. 2. To prevent, preclude, hinder, or debar (an action or thing) by physical means….”(34) Webster’s is in full accord.(35)

The language that precedes the word “prohibiting,” “Congress shall make no law,” illuminates the likely choice our reader would make between the two possible definitions of that word. Without analyzing this phrase word by word, it seems reasonably clear that these words expressly preclude the federal legislature from exercising its legislative authority in some manner later to be described in the sentence. The first definition of prohibit is premised on the concept of an authority to command, and legislative action in the form of a statute that precludes particular conduct is a prime example of such a prohibition. A legally binding provision that states, “Congress shall make no law prohibiting,” takes from Congress the authority to command in the specified area. If the subsequent language stated, “speeds in excess of sixty-five miles per hour,” we would assume that Congress could not pass legislation making such speeds unlawful. Thus, the first definition of prohibit seems precisely tailored to the design of the First Amendment as a limit on the authority to command through legislation. Congress may not prohibit the specified acts through the exercise of legislative authority. Assuming our reader understood the Bill of Rights to involve limits on governmental authority, it seems quite likely that he or she would interpret the word “prohibiting” in a manner consistent with the first usage.

The likelihood that our reader would adopt the first definition of prohibit is bolstered by further consideration of the second definition. Under that definition, the word “prohibit” is synonymous with “prevent” or “hinder,” often in the sense of a physical obstruction. One could say, “The weather prohibited (prevented) the contractor from completing the construction.” Or one could say, “The high price of the goods prohibited their sale.” In these examples, prohibit is not used in the sense of a command, but in the sense of an obstacle. If we attempt to put a human face behind such a prohibition, we might say, “Jean’s attitude prohibited further discussion.” By so saying, we would not mean that Jean had directly forbidden further discussion–at least the foregoing quote would be a rather inartful way of stating that proposition; rather, we would most likely mean that Jean’s attitude made further discussion pointless. Her attitude indirectly ended the conversation. On the other hand, if we insert Congress into any of the foregoing examples, a very different meaning emerges: “Congress prohibited the contractor from completing the construction,” or “Congress prohibited the sale of those goods,” or “Congress prohibited further discussion.” In each of the revised examples, the sense of direct command is reinserted, reinforcing the importance of the context in which the word “prohibit” is being used. I strongly suspect our reader would have appreciated that context and would have proceeded under the assumption that the word “prohibiting” so used conveyed a sense of a direct, authoritarian command.(36)

In the context of our complete sentence, “Congress shall make no law prohibiting the free exercise of religion,” this means at a minimum that the federal legislature is without power to forbid divinely inspired worship (prayers, religious practices) and perhaps other religiously motivated conduct. Congress could be said to forbid (or prohibit) such practices when the command of the law is directed at the religiously motivated practice, as in “Religious worship on Saturday is forbidden.” On the other hand, a Sunday closing law that did not preclude Saturday worship, but only made Saturday worship more difficult (perhaps even substantially so) would not be covered within the proscription and hence would not be a law prohibiting the exercise of religion. Although Congress may have adversely affected a religious practice, it did not prohibit that practice, at least not in the sense of the first definition of prohibit.

A slightly different question is presented by a law that forbids the taking of specified action that under some circumstances may be religious in nature, such as a neutral law prohibiting the consumption of wine. Does such a law prohibit the free exercise of religion? A law making the consumption of wine illegal clearly prohibits the consumption of wine; it forbids that consumption. Assuming no exceptions to that proscription, the law would also forbid the consumption of wine as part of a religious exercise, even though the law was not directed at suppression of a religious practice. Thus, although the intent of the law may not have been to suppress a religious practice, the law does forbid, in other words, prohibit, something that is a religious practice. To reach a contrary conclusion, one would have to say that the law does not forbid the consumption of wine at religious ceremonies, but, of course, given the defined scope of the law, that is simply not true. These observations suggest that although our reader may not interpret “prohibit” broadly to include all action with an adverse impact upon a religious practice, he or she will not necessarily read “prohibit” narrowly to exclude laws of general applicability that happen to include a religious practice within their proscription.

There is an argument against this slightly more expansive reading of the word prohibit. Congress is precluded from prohibiting “the” free exercise of religion, not all exercises of religion; and although the consumption of wine is sometimes “an” exercise of religion, it does not necessarily follow that a law forbidding the consumption of wine generally is a law that prohibits “the” free exercise of religion. The answer depends on whether one considers the phrase “free exercise of religion” as designed to insulate a catalogue of protected activities from congressional attention, namely, all religious practices, or as descriptive of a concept that embraces a general notion of religious toleration precluding Congress from consciously interfering with religious practices. In other words, could the direction of the entire proscription against congressional action be toward eliminating only those laws that particularly focused on religious practices? It is, after all, “the free exercise of religion” that Congress is precluded from prohibiting. In this sense, a law that prohibited the free exercise of religion would be one that directly outlawed certain conduct because of, and not merely in spite of, that conduct’s religious nature. This more focused reading of the word “prohibiting” would, of course, be consistent with the patterns of religious intoleration with which our reader would have been most familiar, namely, direct and intentional acts of interference with religious practices. Of course, the mere existence of this potential reading does little to establish the choice our reader would make.

My suspicion is that our reader would not likely think about the distinction drawn between laws that specifically prohibit religious practices and laws that generally prohibit a practice that for some persons happens to be religious. Given the historical milieu, our reader would surely read the sentence as necessarily referring to the first sort of prohibition, but would not then make any judgment about the slightly more expansive possibilities for the word. If queried about those possibilities, our reader, being an intelligent person, would probably say, “Interesting. Could that be what they meant?” If potential examples were considered, our reader’s answer might vary according to those examples. Again we are left with an ambiguity. Although “prohibit” would likely be understood in the command sense, there is no obvious answer as to the scope of the command that is being precluded from the congressional arsenal of authority. Is the scope of that proscription premised on the “because of” or the “in spite of” interpretation of the word prohibit? Neither we nor our reader can be certain, at least not without more than is provided by the bare text.(37)

We have uncovered at least two ambiguities that could have reasonably crept into our reader’s attempt to discover meaning in the text of the Free Exercise Clause. The word “prohibiting” can be read broadly to proscribe laws that forbid conduct that happens to be of a religious nature,(38) or narrowly to include only those laws that forbid conduct because of that conduct’s religious nature. Similarly, the phrase “free exercise of religion” can be read broadly to include all religiously motivated conduct, or more narrowly to include only what would be commonly understood to be acts of worship or the like. Applying rules of mathematical probability, these two ambiguities render four possible interpretive combinations:(39)

Prohibiting–Exercise of Religion Broad–Broad Broad–Narrow Narrow–Narrow Narrow–Broad

Of course, each of these interpretive combinations is preceded by the phrase, “Congress shall make no law.” Understanding how that phrase interacts with each combination may bring us slightly closer to our reader’s interpretation or at least give us a clearer idea of the interpretive possibilities.

A Broad/Broad interpretation of the Free Exercise Clause precludes Congress from making any law that proscribes religiously motivated conduct, regardless of whether Congress is aware of the potential religious nature of that conduct and regardless of how one characterizes that conduct (worship or otherwise). Under this interpretation, Congress would, therefore, be required to exempt religiously motivated conduct from the reach of laws of general applicability. At first blush, the Broad/Broad interpretation might seem to lead to absurd results. Thus, if we take the proscription literally, a law outlawing murder in the territories could not be applied to a religiously motivated murder. Although the law prohibited murder only as a general category, it in effect prohibits murder as a religious exercise, that is, as conduct motivated by religious beliefs. But this speculative possibility ignores the limits implicit in the adjective “free.” If, as was suggested earlier, the word free does not connote absolute license, but implicitly includes the basic standards of a civil society such as public safety and health, in other words, if free exercise of religion means freedom of religion within the standards of a civil society, then the Broad/Broad reading would not serve as a vehicle for absurd results. It would, however, require Congress to provide exemptions for religious activity when that activity does not violate the standards of a civil society. One can readily see how the compelling state interest test could be used as a measure of those standards.

A Broad/Narrow reading of the Free Exercise Clause includes the same expansive reading of the word “prohibiting” as does the prior category, but that interpretation is coupled with a more circumspect reading of the phrase “exercise of religion” as including only conduct in the nature of worship. Thus Congress may not either directly or through a law of general applicability forbid any activity that happens to qualify as an act of worship. Again potential absurd results–the protection of ritual human sacrifice–are avoided by a sensible reference to the adjective “free.”

The Broad/Narrow reading does, however, raise two difficulties. The first is the difficulty of distinguishing between religious worship and some other category of religiously motivated conduct. This problem has been alluded to earlier, and the text of the Free Exercise Clause provides no clue as to how one would draw the necessary distinctions. Next, the Broad/ Narrow reading permits Congress to proscribe religiously motivated conduct that cannot be fairly characterized as worship, assuming such a category exists, and Congress could accomplish this through laws of general applicability that happened to include religious activity (for example, no exemptions from a military draft for conscientious objectors, some of whom may object to war on religious grounds) or through laws specifically directed at the religious conduct (for example, no exemptions from a military draft for persons who claim religious scruples against war). In this manner the Broad/ Narrow reading seems to cut against the concept of freedom of religion as a form of religious toleration by permitting Congress to base its treatment of an individual on that individual’s religious motivations and beliefs so long as Congress does not regulate the manner in which that individual worships.

A Narrow/Narrow reading of the Free Exercise Clause provides an interpretation under which Congress is precluded only from making those laws intentionally directed at the suppression of acts of religious worship and the like. Thus Congress could not, consistent with the Free Exercise Clause, create a law that specifically proscribed the drinking of wine as part of a religious ceremony; Congress could, however, pass a law that proscribed the drinking of wine generally, and that law could be lawfully applied to the drinking of wine at a religious ceremony. Also, just as with the Broad/Narrow interpretation, under the Narrow/Narrow interpretation, Congress could pass a law that directly regulated religious conduct not falling into the category of worship. Again, the tension with the concept of freedom of religion is evident.

Finally, a Narrow/Broad reading of the Free Exercise Clause would preclude Congress from directly proscribing any activity falling within the general description of religiously motivated conduct. In other words, Congress could never use one’s religious motivation for conduct as the basis for proscribing that conduct. Since the phrase “free exercise of religion” is being interpreted broadly, no distinction would be drawn between worship and other religiously motivated conduct. Congress would simply be precluded from using religion as a basis for proscribing whatever conduct was being addressed. Congress could, however, pass laws of general applicability that happened to include within their ambit some religious practices.

Of course, we cannot know for certain which of the four potential interpretations our eighteenth century reader would select. Indeed, we do not even know if our reader would be pressed to make any selection among the four; rather, our reader might take from the Free Exercise Clause a somewhat ambiguous understanding of the text, with little concern for the predise scope of either the word “prohibiting” or the phrase “free exercise of religion.” In the absence of a specific factual context, our reader might perceive no pressure to dig for a more precise meaning. It may be sufficient that a commitment to freedom of religion has been stated. Indeed, one of the primary problems with reading a legal text is that it has no concrete meaning until it is applied in a specific factual context, and at that point its meaning is revealed only as a consequence of particular facts and policies. But let us assume that our reader is sufficiently curious about the text to search for a more precise meaning. Do our four interpretive combinations bring us any closer to our reader’s potential interpretation?

There is an interpretive problem that infects both the Broad/Narrow and the Narrow/Narrow combinations. The interpretation of “exercise of religion” as pertaining only to acts of worship and the like seems plausible when the phrase is examined outside the sentence in which it is used. But such an examination looks at words in a relative vacuum. Although “exercise of religion” can mean acts of worship and may often refer to acts of worship in common parlance, its meaning is not necessarily so limited when placed in a sentence that gives the phrase a more particular context and direction. The Free Exercise Clause provides just such a defining context. A general reading of the Free Exercise Clause, that is, a reading that does not carefully parse the clause word by word, certainly suggests that the overall gist of the text is to deny the authority of Congress to interfere with the freedom of religion. As a consequence, a specific interpretation of one component of the clause that runs counter to this general meaning would seem anomalous in the absence of some indication that the component was designed to counter the overall tendency of the sentence. I see nothing in the language of the Free Exercise Clause indicating such a design; and the narrow interpretation of “exercise of religion” does run counter to the general concept of freedom of religion by permitting Congress an authority to use religion as a basis for legal proscription. Thus when the phrase “exercise of religion” is considered within the Free Exercise Clause, the narrow interpretation of that phrase remains a technical possibility but a relatively improbable one. Our reader, being literate and careful (not to mention dependent on my judgment), would likely come to a similar conclusion, thereby interpreting the phrase “exercise of religion” as including a wide array of religiously motivated activities, limited only by the legitimate standards of a civil society.

This leaves us with two interpretive combinations (Broad/Broad and Narrow/Broad). .Under both combinations, the broadest scope is given to the phrase “exercise of religion.” The controversy is over the scope of the word “prohibiting.” I think our reader would assume, at least on first reading, that a law prohibiting the free exercise of religion was a law directly forbidding a specific religious practice or directed at outlawing a particular religious sect. I have two reasons for this conclusion. The historical framework within which our reader’s perceptions were formed was one in which religious persecution was generally accomplished through specific acts directed at religious practices or disfavored religious groups. The Pilgrims came to the new world to flee religious persecution, and so forth, and so on. The words of the clause conjure this obvious history, and without further speculation the image of direct persecution remains dominant. Second, in a relatively unregulated society with a relatively homogeneous religious culture, the possibility that laws of general applicability might burden religious practices was relatively remote and not likely to occur to our reader.

On the other hand, if pressed on the point by a consideration of hypothetical possibilities under laws of general application, such as laws of conscription as they affected Quakers, our reader might well shrug his or her shoulders and say, “Yes, perhaps the language requires Congress to exempt religious practices from the scope of laws of general applicability.” As previously suggested, the word “prohibiting” can easily carry this interpretive weight without contorting the sense of the other words in the Free Exercise Clause. A law prohibiting the consumption of wine also prohibits the consumption of wine as part of a religious observance.

In short, I conclude that a fair case can be made for either the Broad/ Broad or the Narrow/Broad readings of the Free Exercise Clause. I do not think that a convincing case can be made for either the Broad/Narrow or the Narrow/Narrow interpretations. Of course, my conclusions are at best educated guesses as to what a fictional reader would conclude-conclusions tainted by my twentieth century perceptions. Assuming these conclusions have an air of reasonableness, the next step is to measure the decision in Employment Division v. Smith and the RFRA against the backdrop of the two plausible interpretive combinations.


The Smith Court’s exposition of the Free Exercise Clause begins with the observation, “The free exercise of religion means, first and foremost, the right to believe and profess whatever religious doctrine one desires.”(40) In stating this proposition, the Court does not rely upon any particular reading of the text; rather, the Court extrapolates this proposition from modern case law. How the right to believe and profess becomes “first and foremost” in a clause that seems to create protection for the practice of religion is unexplained. Having established this hierarchy, however, it is easier for the Court to place the practice of religion on a somewhat lesser plane. Thus, according to the Court, “It would be true, we think (though no case of ours has involved the point), that a State would be ‘prohibiting the free exercise [of religion]’ if it sought to ban such acts or abstentions only when they are engaged in for religious reasons, or only because of the religious belief that they display.”(41) After this grudging recognition of a protected sphere for religious practices, the Court rejects the proposition that the Free Exercise Clause applies as well to laws of general applicability. In the Court’s words, “As a textual matter, we do not think the words must be given that meaning.”(42) The Court then proceeds to demonstrate that case law supports the narrower interpretation of “prohibiting” as “the correct one.”(43)

Other writers have established that the Court’s exposition of the case law was seriously flawed, especially in light of the Sherbert and Yoder decisions.(44) I agree, that is, assuming one accepts Sherbert and Yoder as describing the appropriate base line. From that base line, the doctrinal discussion in Smith is more of a fabrication than an elucidation; of course, if one is adverse to judicial fabrication, Sherbert and Yoder would be neck and neck with Smith for high honors. My interest here, however, is not doctrinal purity or judicial dexterity, but the relationship between Smith and the text of the Free Exercise Clause. On that score, the Smith decision appears to have adopted a plausible interpretation of the Free Exercise Clause, namely, the Narrow/Broad interpretation in which the word “prohibiting” is interpreted as a command directed at specified conduct. Under this interpretation, a law that prohibits the consumption of wine is not automatically deemed a law that prohibits the free exercise of religion merely because the consumption of wine happens to be a religious practice. Notice, however, that in arriving at this conclusion, the Court did not attempt to discover the meaning of the text by examining the words of the text; rather, meaning is discovered by reference to modern applications of those words (and a somewhat strained interpretation of those applications). The only textual analysis is a sleight of hand dismissal of the broad interpretation–“we do not think the words must be given that meaning.” But in so stating, the Court seems to admit that the words could be given “that meaning.”

In short, the Smith Court chose one plausible interpretation of the Free Exercise Clause over another plausible interpretation, without explaining how the text of the clause supported the choice. When one couples this somewhat cavalier treatment of text with the Court’s less than credible doctrinal circumlocutions, one can only conclude that the “correct interpretation” of the Free Exercise Clause was the product of a policy judgment. Indeed, the opinion says as much:

[I]f “compelling interest” really means what it says (and watering it down here would subvert its rigor in the other fields where it is applied), many laws will not meet the test. Any society adopting such a system would be courting anarchy, but that danger increases in direct proportion to the society’s diversity of religious beliefs, and its determination to coerce or suppress none of them. Precisely because “we are a cosmopolitan nation made up of people of almost every conceivable religious preference,” … and precisely because we value and protect that religious divergence we cannot afford the luxury of deeming presumptively invalid, as applied to the religious objector, every regulation of conduct that does not protect an interest of the highest order.(45)

The Court then describes a litany of laws that might be subjected to this presumption of invalidity,(46) suggesting that the judicial accommodation of free exercise claims asserted against laws of general applicability would prove unworkable in a civil society. This is not necessarily a bad policy judgment; indeed, it has much to commend itself.(47) Most importantly, however, it demonstrates how policy and judgment, applied against the backdrop of modern concerns, can be used to resolve a textual ambiguity. Would that the author of Smith had simply said as much.

The RFRA was adopted as a response to Smith. It represents, in effect, a congressional interpretation of the Free Exercise Clause. That interpretation roughly parallels the Broad/Broad interpretive combination under which laws of general applicability are subject to the strictures of the clause. As such, the RFRA is as textually defensible as is the decision in Smith. One must then inquire into the policy judgment underlying the interpretive choice made by the proponents of the RFRA. But before doing so, one interpretive difficulty must be noted.

The RFRA does go beyond the Broad/Broad interpretation in one important respect. Under the RFRA, government action that “substantially burdens a person’s exercise of religion” is forbidden in the absence of compelling circumstances. This new language broadens the scope of the proscription beyond any sensible interpretation of the word “prohibiting” as that word is used in the context of the Bill of Rights. Prohibit becomes synonymous with interfere and, as discussed earlier, the text does not provide a convincing vehicle for conveying this meaning. A law that merely burdens is not a law that prohibits. Neither the legislative history of the RFRA nor the academic support for its passage provides any clue as to how the “substantially burdens” language comports with the text of the Constitution. Of course, the purpose of that language was to preserve and expand the Court’s decision in Sherbert v. Verner, which, for many proponents of the RFRA, provides more of a foundation for their free exercise jurisprudence than does the language of the Free Exercise Clause. So to this extent, the RFRA is textually indefensible.

Focusing on only the textually defensible aspect of the RFRA, namely, the statute’s reliance on the Broad/Broad interpretation, the best one can say is that Congress relied on one plausible interpretation while the Court relied on another. The text of the Free Exercise Clause, however, played little or no part in the choice made by Congress, i.e., if one can call the congressional decision a choice. The debate over the RFRA did not center upon plausible choices and policy judgments. Rather, the purported need for the RFRA was premised upon an apocalyptic vision of post-Smith freedom of religion in America.(48) The clear illegitimacy of Smith was presumed; the policy judgment and the plausible interpretive choice at the heart of Smith were ignored. Instead, a different policy concern, namely, a belief that “[r]eligious liberty in this country is in very serious crises,”(49) drove the passage of the RFRA and its reliance upon a broadly conceived definition of “prohibit.”

Does the RFRA represent a valid interpretation of text? From the perspective of our eighteenth century reader, the answer is partly yes and partly no. Yes, as to prohibitory laws of general applicability that happen to include religious activities within their proscription; no as to laws that burden, but do not proscribe, religious activities.


An alternative to both Smith and the RFRA would be premised on the twin notions of restraint and patience, and upon the acceptance of ambiguity as a healthy part of constitutional law. If the constitutional text is both relevant and ambiguous, the most sensible method to resolve that ambiguity is by reference to concrete facts arising in specific cases. The ambiguity would not always be resolved in favor of one alternative or the other. Rather, choices would be made based upon relevant facts and good judgment in light of current circumstances. Revision and rethinking would be the only constant. Stated somewhat differently, given two plausible interpretations of the constitutional text, it is both unnecessary and unwise to anoint one alternative as the correct one. Usually both have something to offer. Better, therefore, to permit both alternatives to remain part of a jurisprudence that of necessity must remain flexible given the unknown contingencies to which it will be applied. Which alternative is “correct” in any particular case will depend on the facts and circumstances of that case.

The Smith Court, unfortunately, violated all these principles. It decided broad-based and theoretical questions that were unnecessary to the decision before it. Justice O’Connor’s concurrence demonstrated as much. And it created a rigid structure based upon a policy-driven “correct” interpretation of the Free Exercise Clause that purports to distinguish between religious-specific laws and laws of general applicability. But the policy of one case need not be the policy of another. A policy against insensitivity to less familiar religions might lead to a different “correct” conclusion. Moreover, the structure created by the Court resembles reality only at the obvious endpoints. For example, the antipolygamy law at issue in Reynolds v. United States(50) was a law of general applicability, but may well have been driven by religious-specific animosity. The compelling state interest test is certainly one way to ferret out such animosity. Other techniques are available as well.

The RFRA, although understandable as a reaction to Smith, stands on no better ground. Its “correct” interpretation of the Free Exercise Clause is used to interpose a relatively rigid version of the compelling state interest test between the government and all religious objectors who claim that a law substantially burdens their religious practices.(51) What, by the way, is a substantial burden? This broad sweep does trigger some of the concerns suggested by the Smith Court, namely, the courting of anarchy in a religiously diverse society. For example, to what extent will curriculum decisions at public schools be subject to the compelling state interest test? Realistically, some laws that invade religious practices require close scrutiny and others do not, and the circumstances under which a court ought to be suspicious of any particular government action cannot be defined in advance by any particular test. The Court’s decisions between Sherbert and Smith demonstrate the complex fabric of free exercise jurisprudence. Indeed, the more recent decision in Church of Lukumi Babalu Aye, Inc. v. City of Hialeah,(52) invalidating an animal sacrifice ordinance under the Free Exercise Clause, indicates that Smith may not have been as broad-reaching as first anticipated.

To sum up, two bad ideas do not make a good one. Neither Smith nor the RFRA adequately addresses the significance of the constitutional text. Nor does either of them reflect a realistic assessment of how law actually functions and develops. But not to worry, efforts to establish acontextual meaning are bound to fail. Over time both Smith and the RFRA will be constructed away as text, fact, and policy collide in real cases.

1. Religious Freedom Restoration Act of 1993, Pub. L. No. 103-141, 107 Stat. 1488 (codified in scattered sections of 42 U.S.C., 5 U.S.C.).

2. 494 U.S. 872 (1990).

3. Employment Div. v. Smith, 494 U.S. 872, 890 (1990).

4. Id. at 478-79.

5. See generally Church of Lukumi Babalu Aye, Inc. v. City of Hialeah, 113 S. Ct. 2217 (1993) (holding that law resulted from efforts to eliminate specific religious conduct).

6. Employment Div. v. Smith, 494 U.S. 872, 890 (1990). After the decision in Smith, the State of Oregon adopted legislation that exempted members of the Native American Church from the laws proscribing the use of peyote. OR. REV. STAT. 475.992(5) (1991). Many other states also recognize an exemption. See, e.g., ARIZ. REV. STW. ANN. B 13-3402(B)(1)-(3) (1992); COLO. REV. STAT. 12-22-317(3) (1992); KAN. STAT. ANN. 65-4116(8) (1991); NEV. REV. STAT. 453.541(1991).

7. Citations to the critical literature are noted in James E. Ryan, Smith and the Religious Freedom Restoration Act: An Iconoclastic Assessment, 78 VA. L. REV. 1407, 1409 n.15 (1992). See also Edward M. Gaffney, Jr., The Religion Clause: A Double Guarantee of Religious Liberty, 1993 B.Y.U. L. REV. 189; Douglas Laycock, The Religious Freedom Restoration Act, 1993 B.Y.U. L. REV. 221; John T. Noonan, Jr., The End of Free Exercise?, 42 DEPAUL L. REV. 567 (1992); Alfred J. Sciarrino, The Rehnquist Court’s Free Exercise Collision on the Peyote Rood, 23 CUMB. L. REV. 315 (1993). For dissenting views, see Gerard V. Bradley, Beguiled: Free Exercise Exemptions and the Siren Song of Liberalism, 20 HOFSTRA L. REV. 245 (1991); William P. Marshall, In Defense of Smith and Free Exercise Revisionism, 58 U. CHI. L. REV. 308 (1991); Mark Tushnet, The Rhetoric of Free Exercise Discourse, r993 B.Y.U. L. REV. 117.

8. 374 U.S. 398 (1963).

9. 406 U.S. 205 (1972).

10. Sherbert v. Verner, 374 U.S. 398, 410 (1963).

11. Wisconsin v. Yoder, 406 U.S. 205, 234-35 (1972).

12. The Court left open the possibility that the compelling state interest test would continue to be applied to unemployment compensation laws. Employment Div. v. Smith, 494 U.S. 872, 883-84 (1990).

13. The RFRA provides: The purposes of this Act are to restore the compelling interest test as set forth in Sherbert v. Verner, 374 U.S. 398 (1963) and Wisconsin v. Yoder, 406 U.S. 205 (1972) and to guarantee its application in all cases where free exercise of religion is substantially burdened; and to provide a claim or defense to persons whose religious exercise is substantially burdened by government.

Religious Freedom Restoration Act of 1993, Pub. L. No. 103-141, 107 Stat. 1488 Sec 2(b)(1), (2) (codified in 42 U.S.C. 2000bb).

14. Religious Freedom Restoration Act of 1991: Hearings on H.R. 2797 Before the Subcomm. on Civil and Constitutional Rights, 102d Cong., 2d Sess. 63 (1992) (statement of Nadine Strossen, President, American Civil Liberties Union (ACLU)); Douglas Laycock, Summary and Synthesis: The Crisis in Religious Liberty, 60 GEO. WASH. L. REV. 841, 855 (1992).

15. Sec 3(a), 107 Stat. at 1488.

16. Id. Sec B 3(b)(1), (2) at 1488-89.

17. For a similar critique of RFRA, see generally Ryan, supra note 7.

18. Justice Scalia attempted to distinguish prior, inconsistent cases. See Employment Div. v. Smith, 494 U.S. 872, 881-84 (1990).

19. U.S. CONST. amend. I.


21. Id.


23. Id.

24. Id.


16. Id.

27. JOHNSON, supra note 20.


29. Id. at 401-02.


31. Some writers have suggested that the phrase “free exercise of religion” was used interchangeably with the phrase “rights of conscience,” both of which merely stood for the general concept of freedom of religion. Michael W. McConnell, Free Exercise Revisionism and the Smith Decision, 57 U. CHI. L. REV. 1109, 1114-19 (1990). From a reader’s perspective, this insight only leads to a further abstraction. What is freedom of religion and to what extent does that freedom include the right to practice one’s religion? This question takes us back to the text as written.

32. JOHNSON, supra note 20.

33. Id.

34. 8 OXFORD ENGLISH DICTIONARY 1441 (1970). The OED provides a third definition: “To forbid, stop, or prevent (a person) … from doing something ….” Id. However, that definition appears to be nothing more than a variation of the first definition with a focus of the prohibition on a person rather than on an action or thing. The examples given under each definition bear this out. Under the first definition: “In England an act of parliament was passed in the 5th year of reign of Henry IV prohibiting the attempts at transmutation and making them felonious.” Id. Under the third definition: “There is no Act … prohibiting the Secretary of State for Foreign Affairs from being in the pay of continental powers.” Id. at 1442.


36. I realize that by rejecting the second definition of “prohibition” I have implicitly suggested that modern cases such as Sherbert v. Verner, 374 U.S. 398 (1963), are atextual; I will return to this point in the concluding section.

37. See McConnell. supra note 31. at 1114-16. Professor McConnell comments on the two potential meanings of the word “prohibit.” McConnell suggests that one ought to presume, in the absence of contrary evidence, that the broader interpretation of the word ought to prevail: “While we cannot rule out the possibility that the term ‘prohibiting’ might impliedly be limited to laws that prohibit the exercise of religion in a particular way–that is, in a discriminatory fashion–we should at least begin with the presumption that the words carry as broad a meaning as their natural usage.” Id. I am not sure what it means to refer to a word’s natural usage, and the phrase becomes especially unclear when the word has more than one use, as does the word “prohibit.” I would think that one should attempt to find meaning by examining the context in which the word is being used and by then assessing how the potential usages of the word function within that context.

Professor McConnell also argues that state-court decisions construing state constitutional provisions in the years surrounding the adoption of the Bill of Rights establish, in essence, the broader reading of the word “prohibiting.” Michael McConnell, The Origins and Historical Understanding of Free Exercise of Religion, 103 HARV. L. REV. 1409 (1990). Professor Gerard Bradley, however, has provided a powerful rejoinder to McConnell’s conclusions. Bradley, supra note 7, at 261-307.

38. I assume for present purposes that Congress could avoid this broad proscription by exempting religious practices from the coverage of any particular law.

39. Of Course, if there are more ambiguities, there are more possibilities. For example, if the definition of prohibit as prevent or hinder were used, the interpretive combinations would increase to six (three definitions of prohibit multiplied by two definitions of exercise of religion). Similarly, if we accept the definition of religion that includes secular beliefs our interpretive combinations would be further increased to twelve (three definitions of prohibit multiplied by two definitions of exercise multiplied by two definitions of religion). I think, however, for reasons already explained, the combinations described in the text represent the most likely potential readings of the text.

40. Employment Div. v. Smith, 494 U.S. 872, 877 (1990).

41. Id.

42. Id. at 878.

43. Id.

44. See generally McConnell, supra note 31.

45. Employment Div. v. Smith, 494 U.S. 872, 888 (1990).

46. Id. at 889.

47. See Marshall, supra note 7; Tushnet, supra note 7.

48. See Religious Freedom Restoration Act of 1991: Hearings on H.R. 2797 Before the Subcomm. on Civil and Constitutional Rights, 102d Cong., 2d Sess. 65-67; 330-59 (1992) (Prepared Statements of Nadine Strossen, President, ACLU, and Douglas Laycock, Professor of Law, University of Texas).

49. Laycock, supra note 14, at 850.

50. 98 U.S. 145 (1879).

51. Contrary to some assertions in the legislative history, the phrase “compelling state interest” does not have any set legal definition. “Compelling under the circumstances” would be a more apt phrasing. See, e.g., Maine v. Taylor, 477 U.S. 131 (1986) (finding state’s interest in protecting local baitfish compelling).

52. 113 S. Ct. 2217 (1993).

Copyright Washington & Lee University, School of Law Winter 1994

Provided by ProQuest Information and Learning Company. All rights Reserved