concurring.

I concur in the Court’s opinion and judgment that Ala. Code §16-1-20.1 (Supp. 1984) violates the Establishment Clause of the First Amendment. My concurrence is prompted by Alabama’s persistence in attempting to institute state-sponsored prayer in the public schools by enacting three successive statutes.1 I agree fully with Justice O’Connor’s assertion that some moment-of-silence statutes may be constitutional,2 a suggestion set forth in the Court’s opinion as well. Ante, at 59.

*63I write separately to express additional views and to respond to criticism of the three-pronged Lemon test.3 Lemon v. Kurtzman, 403 U. S. 602 (1971), identifies standards that have proved useful in analyzing case after case both in our decisions and in those of other courts. It is the only coherent test a majority of the Court has ever adopted. Only once since our decision in Lemon, supra, have we addressed an Establishment Clause issue without resort to its three-pronged test. See Marsh v. Chambers, 463 U. S. 783 (1983).4 Lemon, supra, has not been overruled or its test modified. Yet, continued criticism of it could encourage other courts to feel free to decide Establishment Clause cases on an ad hoc basis.5

*64The first inquiry under Lemon is whether the challenged statute has a “secular legislative purpose.” Lemon v. Kurtzman, supra, at 612. As Justice O’Connor recognizes, this secular purpose must be “sincere”; a law will not pass constitutional muster if the secular purpose articulated by the legislature is merely a “sham.” Post, at 75 (concurring in judgment). In Stone v. Graham, 449 U. S. 39 (1980) (per curiam), for example, we held that a statute requiring the posting of the Ten Commandments in public schools violated the Establishment Clause, even though the Kentucky Legislature asserted that its goal was educational. We have not interpreted the first prong of Lemon, supra, however, as requiring that a statute have “exclusively secular” objectives.6 Lynch v. Donnelly, 465 U. S. 668, 681, n. 6 (1984). If such a requirement existed, much conduct and legislation approved by this Court in the past would have been invalidated. See, e. g., Walz v. Tax Comm’n, 397 U. S. 664 (1970) (New York’s property tax exemption for religious organizations upheld); Everson v. Board of Education, 330 U. S. 1 (1947) (holding that a township may reimburse parents for the cost of transporting their children to parochial schools).

*65The record before us, however, makes clear that Alabama’s purpose was solely religious in character. Senator Donald Holmes, the sponsor of the bill that became Alabama Code § 16-1-20.1 (Supp. 1984), freely acknowledged that the purpose of this statute was “to return voluntary prayer” to the public schools. See ante, at 57, n. 43. I agree with Justice O’Connor that a single legislator’s statement, particularly if made following enactment, is not necessarily sufficient to establish purpose. See post, at 77 (concurring in judgment). But, as noted in the Court’s opinion, the religious purpose of § 16-1-20.1 is manifested in other evidence, including the sequence and history of the three Alabama statutes. See ante, at 58-60.

I also consider it of critical importance that neither the District Court nor the Court of Appeals found a secular purpose, while both agreed that the purpose was to advance religion. In its first opinion (enjoining the enforcement of § 16-1-20.1 pending a hearing on the merits), the District Court said that the statute did “not reflect a clearly secular purpose.” Jaffree v. James, 544 F. Supp. 727, 732 (SD Ala. 1982). Instead, the District Court found that the enactment of the statute was an “effort on the part of the State of Alabama to encourage a religious activity.”7 Ibid. The Court of Appeals likewise applied the Lemon test and found “a lack of secular purpose on the part of the Alabama Legislature.” *66705 F. 2d 1526, 1535 (CA11 1983). It held that the objective of §16-1-20.1 was the “advancement of religion.” Ibid. When both courts below are unable to discern an arguably valid secular purpose, this Court normally should hesitate to find one.

I would vote to uphold the Alabama statute if it also had a clear secular purpose. See Mueller v. Allen, 463 U. S. 388, 394-395 (1983) (the Court is “reluctan[t] to attribute unconstitutional motives to the States, particularly when a plausible secular purpose for the State’s program may be discerned from the face of the statute”). Nothing in the record before us, however, identifies a clear secular purpose, and the State also has failed to identify any nonreligious reason for the statute’s enactment.8 Under these circumstances, the Court is required by our precedents to hold that the statute fails the first prong of the Lemon test and therefore violates the Establishment Clause.

Although we do not reach the other two prongs of the Lemon test, I note that the “effect” of a straightforward moment-of-silence statute is unlikely to “advanc[e] or in-hibi[t] religion.”9 See Board of Education v. Allen, 392 U. S. 236, 243 (1968). Nor would such a statute “foster ‘an excessive government entanglement with religion.’ ” Lemon *67v. Kurtzman, 403 U. S., at 612-613, quoting Walz v. Tax Comm’n, 397 U. S., at 674.

I join the opinion and judgment of the Court.

concurring in the judgment.

Nothing in the United States Constitution as interpreted by this Court or in the laws of the State of Alabama prohibits public school students from voluntarily praying at any time before, during, or after the schoolday. Alabama has facilitated voluntary silent prayers of students who are so inclined by enacting Ala. Code § 16-1-20 (Supp. 1984), which provides a moment of silence in appellees’ schools each day. The parties to these proceedings concede the validity of this enactment. At issue in these appeals is the constitutional validity of an additional and subsequent Alabama statute, Ala. Code § 16-1-20.1 (Supp. 1984), which both the District Court and the Court of Appeals concluded was enacted solely to officially encourage prayer during the moment of silence. I agree with the judgment of the Court that, in light of the findings of the courts below and the history of its enactment, § 16-1-20.1 of the Alabama Code violates the Establishment Clause of the First Amendment. In my view, there can be little doubt that the purpose and likely effect of this subsequent enactment is to endorse and sponsor voluntary prayer in the public schools. I write separately to identify the peculiar features of the Alabama law that render it invalid, and to explain why moment of silence laws in other States do not necessarily manifest the same infirmity. I also write to explain why neither history nor the Free Exercise Clause of the First Amendment validates the Alabama law struck down by the Court today.

I

The Religion Clauses of the First Amendment, coupled with the Fourteenth Amendment’s guarantee of ordered liberty, preclude both the Nation and the States from making any law respecting an establishment of religion or prohibiting *68the free exercise thereof. Cantwell v. Connecticut, 310 U. S. 296, 303 (1940). Although a distinct jurisprudence has enveloped each of these Clauses, their common purpose is to secure religious liberty. See Engel v. Vitale, 370 U. S. 421, 430 (1962). On these principles the Court has been and remains unanimous.

As these cases once again demonstrate, however, “it is far easier to agree on the purpose that underlies the First Amendment’s Establishment and Free Exercise Clauses than to obtain agreement on the standards that should govern their application.” Walz v. Tax Comm’n, 397 U. S. 664, 694 (1970) (opinion of Harlan, J.). It once appeared that the Court had developed a workable standard by which to identify impermissible government establishments of religion. See Lemon v. Kurtzman, 403 U. S. 602 (1971). Under the now familiar Lemon test, statutes must have both a secular legislative purpose and a principal or primary effect that neither advances nor inhibits religion, and in addition they must not foster excessive government entanglement with religion. Id., at 612-613. Despite its initial promise, the Lemon test has proved problematic. The required inquiry into “entanglement” has been modified and questioned, see Mueller v. Allen, 463 U. S. 388, 403, n. 11 (1983), and in one case we have upheld state action against an Establishment Clause challenge without applying the Lemon test at all. Marsh v. Chambers, 463 U. S. 783 (1983). The author of Lemon himself apparently questions the test’s general applicability. See Lynch v. Donnelly, 465 U. S. 668, 679 (1984). Justice Rehnquist today suggests that we abandon Lemon entirely, and in the process limit the reach of the Establishment Clause to state discrimination between sects and government designation of a particular church as a “state” or “national” one. Post, at 108-113.

Perhaps because I am new to the struggle, I am not ready to abandon all aspects of the Lemon test. I do believe, however, that the standards announced in Lemon should be *69reexamined and refined in order to make them more useful in achieving the underlying purpose of the First Amendment. We must strive to do more than erect a constitutional “signpost,” Hunt v. McNair, 413 U. S. 734, 741 (1973), to be followed or ignored in a particular case as our predilections may dictate. Instead, our goal should be “to frame a principle for constitutional adjudication that is not only grounded in the history and language of the first amendment, but one that is also capable of consistent application to the relevant problems.” Choper, Religion in the Public Schools: A Proposed Constitutional Standard, 47 Minn. L. Rev. 329, 332-333 (1963) (footnotes omitted). Last Term, I proposed a refinement of the Lemon test with this goal in mind. Lynch v. Donnelly, 465 U. S., at 687-689 (concurring opinion).

The Lynch concurrence suggested that the religious liberty protected by the Establishment Clause is infringed when the government makes adherence to religion relevant to a person’s standing in the political community. Direct government action endorsing religion or a particular religious practice is invalid under this approach because it “sends a message to nonadherents that they are outsiders, not full members of the political community, and an accompanying message to adherents that they are insiders, favored members of the political community.” Id., at 688. Under this view, Lemon’s, inquiry as to the purpose and effect of a statute requires courts to examine whether government’s purpose is to endorse religion and whether the statute actually conveys a message of endorsement.

The endorsement test is useful because of the analytic content it gives to the Lemon-mandated inquiry into legislative purpose and effect. In this country, church and state must necessarily operate within the same community. Because of this coexistence, it is inevitable that the secular interests of government and the religious interests of various sects and their adherents will frequently intersect, conflict, and combine. A statute that ostensibly promotes a secular interest *70often has an incidental or even a primary effect of helping or hindering a sectarian belief. Chaos would ensue if every such statute were invalid under the Establishment Clause. For example, the State could not criminalize murder for fear that it would thereby promote the Biblical command against killing. The task for the Court is to sort out those statutes and government practices whose purpose and effect go against the grain of religious liberty protected by the First Amendment.

The endorsement test does not preclude government from acknowledging religion or from taking religion into account in making law and policy. It does preclude government from conveying or attempting to convey a message that religion or a particular religious belief is favored or preferred. Such an endorsement infringes the religious liberty of the nonadher-ent, for “[w]hen the power, prestige and financial support of government is placed behind a particular religious belief, the indirect coercive pressure upon religious minorities to conform to the prevailing officially approved religion is plain.” Engel v. Vitale, supra, at 431. At issue today is whether state moment of silence statutes in general, and Alabama’s moment of silence statute in particular, embody an impermissible endorsement of prayer in public schools.

A

Twenty-five states permit or require public school teachers to have students observe a moment of silence in their classrooms.1 A few statutes provide that the moment of silence *71is for the purpose of meditation alone. See Ariz. Rev. Stat. Ann. § 15-522 (1984); Conn. Gen. Stat. § 10-16a (1983); R. I. Gen. Laws § 16—12—3.1 (1981). The typical statute, however, calls for a moment of silence at the beginning of the schoolday during which students may meditate, pray, or reflect on the activities of the day. See, e. g., Ark. Stat. Ann. § 80-1607.1 (1980); Ga. Code Ann. § 20-2-1050 (1982); Ill. Rev. Stat., ch. 122, ¶ 771 (1983); Ind. Code § 20-10.1-7-11 (1982); Kan. Stat. Ann. § 72-5308a (1980); Pa. Stat. Ann., Tit. 24, § 15-1516.1 (Purdon Supp. 1984-1985). Federal trial courts have divided on the constitutionality of these moment of silence laws. Compare Gaines v. Anderson, 421 F. Supp. 337 (Mass. 1976) (upholding statute), with May v. Cooperman, 572 F. Supp. 1561 (NJ 1983) (striking down statute); Duffy v. Las Cruces Public Schools, 557 F. Supp. 1013 (NM 1983) (same); and Beck v. McElrath, 548 F. Supp. 1161 (MD Tenn. 1982) (same). See also Walter v. West Virginia Board of Education, Civ. Action No. 84-5366 (SD W. Va., Mar. 14, 1985) (striking down state constitutional amendment). Relying on this Court’s decisions disapproving vocal prayer and Bible reading in the public schools, see Abington School District v. Schempp, 374 U. S. 203 (1963); Engel v. Vitale, 370 U. S. 421 (1962), the courts that have struck down the moment of silence statutes generally conclude that their purpose and effect are to encourage prayer in public schools.

The Engel and Abington decisions are not dispositive on the constitutionality of moment of silence laws. In those *72cases, public school teachers and students led their classes in devotional exercises. In Engel, a New York statute required teachers to lead their classes in a vocal prayer. The Court concluded that "it is no part of the business of government to compose official prayers for any group of the American people to recite as part of a religious program carried on by the government.” 370 U. S., at 425. In Abington, the Court addressed Pennsylvania and Maryland statutes that authorized morning Bible readings in public schools. The Court reviewed the purpose and effect of the statutes, concluded that they required religious exercises, and therefore found them to violate the Establishment Clause. 374 U. S., at 223-224. Under all of these statutes, a student who did not share the religious beliefs expressed in the course of the exercise was left with the choice of participating, thereby compromising the nonadherent’s beliefs, or withdrawing, thereby calling attention to his or her nonconformity. The decisions acknowledged the coercion implicit under the statutory schemes, see Engel, supra, at 431, but they expressly turned only on the fact that the government was sponsoring a manifestly religious exercise.

A state-sponsored moment of silence in the public schools is different from state-sponsored vocal prayer or Bible reading. First, a moment of silence is not inherently religious. Silence, unlike prayer or Bible reading, need not be associated with a religious exercise. Second, a pupil who participates in a moment of silence need not compromise his or her beliefs. During a moment of silence, a student who objects to prayer is left to his or her own thoughts, and is not compelled to listen to the prayers or thoughts of others. For these simple reasons, a moment of silence statute does not stand or fall under the Establishment Clause according to how the Court regards vocal prayer or Bible reading. Scholars and at least one Member of this Court have recognized the distinction and suggested that a moment of silence in public schools would be constitutional. See Abington, supra, at 281 (Brennan, J., concurring) (“[T]he observance of a mo*73ment of reverent silence at the opening of class” may serve “the solely secular purposes of the devotional activities without jeopardizing either the religious liberties of any members of the community or the proper degree of separation between the spheres of religion and goverment”); L. Tribe, American Constitutional Law § 14-6, p. 829 (1978); P. Freund, The Legal Issue, in Religion and the Public Schools 23 (1965); Choper, 47 Minn. L. Rev., at 371; Kauper, Prayer, Public Schools, and the Supreme Court, 61 Mich. L. Rev. 1031, 1041 (1963). As a general matter, I agree. It is difficult to discern a serious threat to religious liberty from a room of silent, thoughtful schoolchildren.

By mandating a moment of silence, a State does not necessarily endorse any activity that might occur during the period. Cf. Widmar v. Vincent, 454 U. S. 263, 272, n. 11 (1981) (“[B]y creating a forum the [State] does not thereby endorse or promote any of the particular ideas aired there”). Even if a statute specifies that a student may choose to pray silently during a quiet moment, the'State has not thereby encouraged prayer over other specified alternatives. Nonetheless, it is also possible that a moment of silence statute, either as drafted or as actually implemented, could effectively favor the child who prays over the child who does not. For example, the message of endorsement would seem inescapable if the teacher exhorts children to use the designated time to pray. Similarly, the face of the statute or its legislative history may clearly establish that it seeks to encourage or promote voluntary prayer over other alternatives, rather than merely provide a quiet moment that may be dedicated to prayer by those so inclined. The crucial question is whether the State has conveyed or attempted to convey the message that children should use the moment of silence for prayer.2 *74This question cannot be answered in the abstract, but instead requires courts to examine the history, language, and administration of a particular statute to determine whether it operates as an endorsement of religion. Lynch, 465 U. S., at 694 (concurring opinion) (“Every government practice must be judged in its unique circumstances to determine whether it constitutes an endorsement or disapproval of religion”).

Before reviewing Alabama’s moment of silence law to determine whether it endorses prayer, some general observations on the proper scope of the inquiry are in order. First, the inquiry .into the purpose of the legislature in enacting a moment of silence law should be deferential and limited. See Everson v. Board of Education, 330 U. S. 1, 6 (1947) (courts must exercise “the most extreme caution” in assessing whether a state statute has a proper public purpose). In determining whether the government intends a moment of silence statute to convey a message of endorsement or disapproval of religion, a court has no license to psychoanalyze the legislators. See McGowan v. Maryland, 366 U. S. 420, 466 (1961) (opinion of Frankfurter, J.). If a legislature expresses a plausible secular purpose for a moment of silence statute in either the text or the legislative history,3 or if the statute disclaims an intent to encourage prayer over alternatives during a moment of silence,4 then courts should gener*75ally defer to that stated intent. See Committee for Public Education & Religious Liberty v. Nyquist, 413 U. S. 756, 773 (1973); Tilton v. Richardson, 403 U. S. 672, 678-679 (1971). It is particularly troublesome to denigrate an expressed secular purpose due to postenactment testimony by particular legislators or by interested persons who witnessed the drafting of the statute. Even if the text and official history of a statute express no secular purpose, the statute should be held to have an improper purpose only if it is beyond purview that endorsement of religion or a religious belief “was and is the law’s reason for existence.” Epperson v. Arkansas, 393 U. S. 97, 108 (1968). Since there is arguably a secular pedagogical value to a moment of silence in public schools, courts should find an improper purpose behind such a statute only if the statute on its face, in its official legislative history, or in its interpretation by a responsible administrative agency suggests it has the primary purpose of endorsing prayer.

Justice Rehnquist suggests that this sort of deferential inquiry into legislative purpose “means little,” because “it only requires the legislature to express any secular purpose and omit all sectarian references.” Post, at 108. It is not a trivial matter, however, to require that the legislature manifest a secular purpose and omit all sectarian endorsements from its laws. That requirement is precisely tailored to the Establishment Clause’s purpose of assuring that government not intentionally endorse religion or a religious practice. It is of course possible that a legislature will enunciate a sham secular purpose for a statute. I have little doubt that our courts are capable of distinguishing a sham secular purpose from a sincere one, or that the Lemon inquiry into the effect of an enactment would help decide those close cases where the validity of an expressed secular purpose is in doubt. While the secular purpose requirement alone may rarely be determinative in striking down a statute, it nevertheless serves an important function. It reminds government that *76when it acts it should do so without endorsing a particular religious belief or practice that all citizens do not share. In this sense the secular purpose requirement is squarely based in the text of the Establishment Clause it helps to enforce.

Second, the Lynch concurrence suggested that the effect of a moment of silence law is not entirely a question of fact:

“[W]hether a government activity communicates endorsement of religion is not a question of simple historical fact. Although evidentiary submissions may help answer it, the question is, like the question whether racial or sex-based classifications communicate an invidious message, in large part a legal question to be answered on the basis of judicial interpretation of social facts.” 465 U. S., at 693-694.

The relevant issue is whether an objective observer, acquainted with the text, legislative history, and implementation of the statute, would perceive it as a state endorsement of prayer in public schools. Cf. Bose Corp. v. Consumers Union of United States, Inc., 466 U. S. 485, 517-518, n. 1 (1984) (Rehnquist, J., dissenting) (noting that questions whether fighting words are “likely to provoke the average person to retaliation,” Street v. New York, 394 U. S. 576, 592 (1969), and whether allegedly obscene material appeals to “prurient interests,” Miller v. California, 413 U. S. 15, 24 (1973), are mixed questions of law and fact that are properly subject to de novo appellate review). A moment of silence law that is clearly drafted and implemented so as to permit prayer, meditation, and reflection within the prescribed period, without endorsing one alternative over the others, should pass this test.

B

The analysis above suggests that moment of silence laws in many States should pass Establishment Clause scrutiny because they do not favor the child who chooses to pray during a moment of silence over the child who chooses to medi*77tate or reflect. Alabama Code § 16-1-20.1 (Supp. 1984) does not stand on the same footing. However deferentially one ■ examines its text and legislative history, however objectively one views the message attempted to be conveyed to the public, the conclusion is unavoidable that the purpose of the statute is to endorse prayer in public schools. I accordingly agree with the Court of Appeals, 705 F. 2d 1526, 1535 (1983), that the Alabama statute has a purpose which is in violation of the Establishment Clause, and cannot be upheld.

In finding that the purpose of § 16-1-20.1 is to endorse voluntary prayer during a moment of silence, the Court relies on testimony elicited from State Senator Donald G. Holmes during a preliminary injunction hearing. Ante, at 56-57. Senator Holmes testified that the sole purpose of the statute was to return voluntary prayer to the public schools. For the reasons expressed above, I would give little, if any, weight to this sort of evidence of legislative intent. Nevertheless, the text of the statute in light of its official legislative history leaves little doubt that the purpose of this statute corresponds to the purpose expressed by Senator Holmes at the preliminary injunction hearing.

First, it is notable that Alabama already had a moment of silence statute before it enacted § 16-1-20.1. See Ala. Code § 16-1-20 (Supp. 1984), quoted ante, at 40, n. 1. Appellees do not challenge this statute — indeed, they concede its validity. See Brief for Appellees 2. The only significant addition made by § 16-1-20.1 is to specify expressly that voluntary prayer is one of the authorized activities during a moment of silence. Any doubt as to the legislative purpose of that addition is removed by the official legislative history. The sole purpose reflected in the official history is “to return voluntary prayer to our public schools.” App. 50. Nor does anything in the legislative history contradict an intent to encourage children to choose prayer over other alternatives during the moment of silence. Given this legislative history, it is not surprising that the State of Alabama conceded in the *78courts below that the purpose of the statute was to make prayer part of daily classroom activity, and that both the District Court and the Court of Appeals concluded that the law’s purpose was to encourage religious activity. See ante, at 57, n. 44. In light of the legislative history and the findings of the courts below, I agree with the Court that the State intended § 16-1-20.1 to convey a message that prayer was the endorsed activity during the state-prescribed moment of silence.5 While it is therefore unnecessary also to determine the effect of the statute, Lynch, 465 U. S., at 690 (concurring opinion), it also seems likely that the message actually conveyed to objective observers by §16-1-20.1 is approval of the child who selects prayer over other alternatives during a moment of silence.

Given this evidence in the record, candor requires us to admit that this Alabama statute was intended to convey a message of state encouragement and endorsement of religion. In Walz v. Tax Comm’n, 397 U. S., at 669, the Court stated that the Religion Clauses of the First Amendment are flexible enough to “permit religious exercise to exist without sponsorship and without interference.” Alabama Code §16-1-20.1 (Supp. 1984) does more than permit prayer to occur during a moment of silence “without interference.” It *79endorses the decision to pray during a moment of silence, and accordingly sponsors a religious exercise. For that reason, I concur in the judgment of the Court.

HH HH

In his dissenting opinion, post, at 91-106, JUSTICE Rehn-QUIST reviews the text and history of the First Amendment Religion Clauses. His opinion suggests that a long line of this Court’s decisions are inconsistent with the intent of the drafters of the Bill of Rights. He urges the Court to correct the historical inaccuracies in its past decisions by embracing a far more restricted interpretation of the Establishment Clause, an interpretation that presumably would permit vocal group prayer in public schools. See generally R. Cord, Separation of Church and State (1982).

The United States, in an amicus brief, suggests a less sweeping modification of Establishment Clause principles. In the Federal Government’s view, a state-sponsored moment of silence is merely an “accommodation” of the desire of some public school children to practice their religion by praying silently. Such an accommodation is contemplated by the First Amendment’s guarantee that the Government will not prohibit the free exercise of religion. Because the moment of silence implicates free exercise values, the United States suggests that the Lemon-mandated inquiry into purpose and effect should be modified. Brief for United States as Amicus Curiae 22.

There is an element of truth and much helpful analysis in each of these suggestions. Particularly when we are interpreting the Constitution, “a page of history is worth a volume of logic.” New York Trust Co. v. Eisner, 256 U. S. 345, 349 (1921). Whatever the provision of the Constitution that is at issue, I continue to believe that “fidelity to the notion of constitutional — as opposed to purely judicial — limits on governmental action requires us to impose a heavy burden on those who claim that practices accepted when [the provision] was *80adopted are now constitutionally impermissible.” Tennessee v. Garner, 471 U. S. 1, 26 (1985) (dissenting opinion). The Court properly looked to history in upholding legislative prayer, Marsh v. Chambers, 463 U. S. 783 (1983), property tax exemptions for houses of worship, Walz v. Tax Comm’n, supra, and Sunday closing laws, McGowan v. Maryland, 366 U. S. 420 (1961). As Justice Holmes once observed, “[i]f a thing has been practised for two hundred years by common consent, it will need a strong case for the Fourteenth Amendment to affect it.” Jackman v. Rosenbaum Co., 260 U. S. 22, 31 (1922).

Justice Rehnquist does not assert, however, that the drafters of the First Amendment expressed a preference for prayer in public schools, or that the practice of prayer in public schools enjoyed uninterrupted government endorsement from the time of enactment of the Bill of Rights to the present era. The simple truth is that free public education was virtually nonexistent in the late 18th century. See Abington, 374 U. S., at 238, and n. 7 (Brennan, J., concurring). Since there then existed few government-run schools, it is unlikely that the persons who drafted the First Amendment, or the state legislators who ratified it, anticipated the problems of interaction of church and state in the public schools. Sky, The Establishment Clause, the Congress, and the Schools: An Historical Perspective, 52 Va. L. Rev. 1395, 1403-1404 (1966). Even at the time of adoption of the Fourteenth Amendment, education in Southern States was still primarily in private hands, and the movement toward free public schools supported by general taxation had not taken hold. Brown v. Board of Education, 347 U. S. 483, 489-490 (1954).

This uncertainty as to the intent of the Framers of the Bill of Rights does not mean we should ignore history for guidance on the role of religion in public education. The Court has not done so. See, e. g., Illinois ex rel. McCollum v. Board of Education, 333 U. S. 203, 212 (1948) (Frank*81furter, J., concurring). When the intent of the Framers is unclear, I believe we must employ both history and reason in our analysis. The primary issue raised by Justice Rehnquist’s dissent is whether the historical fact that our Presidents have long called for public prayers of Thanks should be dispositive on the constitutionality of prayer in public schools.6 I think not. At the very least, Presidential Proclamations are distinguishable from school prayer in that they are received in a noncoercive setting and are primarily directed at adults, who presumably are not readily susceptible to unwilling religious indoctrination. This Court’s decisions have recognized a distinction when government-sponsored religious exercises are directed at impressionable children who are required to attend school, for then government endorsement is much more likely to result in coerced religious beliefs. See, e. g., Marsh v. Chambers, supra, at 792; Tilton v. Richardson, 403 U. S., at 686. Although history provides a touchstone for constitutional problems, the Establishment Clause concern for religious liberty is dispositive here.

The element of truth in the United States’ arguments, I believe, lies in the suggestion that Establishment Clause analysis must comport with the mandate of the Free Exercise Clause that government make no law prohibiting the free exercise of religion. Our cases have interpreted the Free Exercise Clause to compel the government to exempt persons from some generally applicable government requirements so as to permit those persons to freely exercise their religion. See, e. g., Thomas v. Review Board of the Indiana Employment Security Division, 450 U. S. 707 (1981); Wisconsin v. Yoder, 406 U. S. 205 (1972); Sherbert v. Verner, 374 *82U. S. 398 (1963). Even where the Free Exercise Clause does not compel the government to grant an exemption, the Court has suggested that the government in some circumstances may voluntarily choose to exempt religious observers without violating the Establishment Clause. See, e. g., Gillette v. United States, 401 U. S. 437, 453 (1971); Braunfeld v. Brown, 366 U. S. 599 (1961). The challenge posed by the United States’ argument is how to define the proper Establishment Clause limits on voluntary government efforts to facilitate the free exercise of religion. On the one hand, a rigid application of the Lemon test would invalidate legislation exempting religious observers from generally applicable government obligations. By definition, such legislation has a religious purpose and effect in promoting the free exercise of religion. On the other hand, judicial deference to all legislation that purports to facilitate the free exercise of religion would completely vitiate the Establishment Clause. Any statute pertaining to religion can be viewed as an “accommodation” of free exercise rights. Indeed, the statute at issue in Lemon, which provided salary supplements, textbooks, and instructional materials to Pennsylvania parochial schools, can be viewed as an accommodation of the religious beliefs of parents who choose to send their children to religious schools.

It is obvious that either of the two Religion Clauses, “if expanded to a logical extreme, would tend to clash with the other.” Walz, 397 U. S., at 668-669. The Court has long exacerbated the conflict by calling for government “neutrality” toward religion. See, e. g., Committee for Public Education & Religious Liberty v. Nyquist, 413 U. S. 756 (1973); Board of Education v. Allen, 392 U. S. 236 (1968). It is difficult to square any notion of “complete neutrality,” ante, at 60, with the mandate of the Free Exercise Clause that government must sometimes exempt a religious observer from an otherwise generally applicable obligation. A government that confers a benefit on an explicitly religious basis is not *83neutral toward religion. See Welsh v. United States, 398 U. S. 333, 372 (1970) (White, J., dissenting).

The solution to the conflict between the Religion Clauses lies not in “neutrality,” but rather in identifying workable limits to the government’s license to promote the free exercise of religion. The text of the Free Exercise Clause speaks of laws that prohibit the free exercise of religion. On its face, the Clause is directed at government interference with free exercise. Given that concern, one can plausibly assert that government pursues Free Exercise Clause values when it lifts a government-imposed burden on the free exercise of religion. If a statute falls within this category, then the standard Establishment Clause test should be modified accordingly. It is disingenuous to look for a purely secular purpose when the manifest objective of a statute is to facilitate the free exercise of religion by lifting a government-imposed burden. Instead, the Court should simply acknowledge that the religious purpose of such a statute is legitimated by the Free Exercise Clause. I would also go further. In assessing the effect of such a statute — that is, in determining whether the statute conveys the message of endorsement of religion or a particular religious belief — courts should assume that the “objective observer,” supra, at 76, is acquainted with the Free Exercise Clause and the values it promotes. Thus individual perceptions, or resentment that a religious observer is exempted from a particular government requirement, would be entitled to little weight if the Free Exercise Clause strongly supported the exemption.

While this “accommodation” analysis would help reconcile our Free Exercise and Establishment Clause standards, it would not save Alabama’s moment of silence law. If we assume that the religious activity that Alabama seeks to protect is silent prayer, then it is difficult to discern any state-imposed burden on that activity that is lifted by Alabama Code §16-1-20.1 (Supp. 1984). No law prevents a student who is so inclined from praying silently in public schools. *84Moreover, state law already provided a moment of silence to these appellees irrespective of §16-1-20.1. See Ala. Code § 16-1-20 (Supp. 1984). Of course, the State might argue that § 16-1-20.1 protects not silent prayer, but rather group silent prayer under state sponsorship. Phrased in these terms, the burden lifted by the statute is not one imposed by the State of Alabama, but by the Establishment Clause as interpreted in Engel and Abington. In my view, it is beyond the authority of the State of Alabama to remove burdens imposed by the Constitution itself. I conclude that the Alabama statute at issue today lifts no state-imposed burden on the free exercise of religion, and accordingly cannot properly be viewed as an accommodation statute.

HH I — I I — I

The Court does not hold that the Establishment Clause is so hostile to religion that it precludes the States from affording schoolchildren an opportunity for voluntary silent prayer. To the contrary, the moment of silence statutes of many States should satisfy the Establishment Clause standard we have here applied. The Court holds only that Alabama has intentionally crossed the line between creating a quiet moment during which those so inclined may pray, and affirmatively endorsing the particular religious practice of prayer. This line may be a fine one, but our precedents and the principles of religious liberty require that we draw it. In my view, the judgment of the Court of Appeals must be affirmed.

dissenting.

Some who trouble to read the opinions in these cases will find it ironic — perhaps even bizarre — that on the very day we heard arguments in the cases, the Court’s session opened with an invocation for Divine protection. Across the park a few hundred yards away, the House of Representatives and *85the Senate regularly open each session with a prayer. These legislative prayers are not just one minute in duration, but are extended, thoughtful invocations and prayers for Divine guidance. They are given, as they have been since 1789, by clergy appointed as official chaplains and paid from the Treasury of the United States. Congress has also provided chapels in the Capitol, at public expense, where Members and others may pause for prayer, meditation — or a moment of silence.

Inevitably some wag is bound to say that the Court’s holding today reflects a belief that the historic practice of the Congress and this Court is justified because members of the Judiciary and Congress are more in need of Divine guidance than are schoolchildren. Still others will say that all this controversy is “much ado about nothing,” since no power on earth — including this Court and Congress — can stop any teacher from opening the schoolday with a moment of silence for pupils to meditate, to plan their day — or to pray if they voluntarily elect to do so.

I make several points about today’s curious holding.

(a) It makes no sense to say that Alabama has “endorsed prayer” by merely enacting a new statute “to specify expressly that voluntary prayer is one of the authorized activities during a moment of silence,” ante, at 77 (O’Connor, J., concurring in judgment) (emphasis added). To suggest that a moment-of-silence statute that includes the word “prayer” unconstitutionally endorses religion, while one that simply provides for a moment of silence does not, manifests not neutrality but hostility toward religion. For decades our opinions have stated that hostility toward any religion or toward all religions is as much forbidden by the Constitution as is an official establishment of religion. The Alabama Legislature has no more “endorsed” religion than a state or the Congress does when it provides for legislative chaplains, or than this Court does when it opens each session with an invocation to *86God. Today’s decision recalls the observations of Justice Goldberg:

“[UJntutored devotion to the concept of neutrality can lead to invocation or approval of results which partake not simply of that noninterference and noninvolvement with the religious which the Constitution commands, but of a brooding and pervasive dedication to the secular and a passive, or even active, hostility to the religious. Such results are not only not compelled by the Constitution, but, it seems to me, are prohibited by it.” Abington School District v. Schempp, 374 U. S. 203, 306 (1963) (concurring opinion).

(b) The inexplicable aspect of the foregoing opinions, however, is what they advance as support for the holding concerning the purpose of the Alabama Legislature. Rather than determining legislative purpose from the face of the statute as a whole,1 the opinions rely on three factors in concluding that the Alabama Legislature had a “wholly religious” purpose for enacting the statute under review, Ala. Code § 16-1-20.1 (Supp. 1984): (i) statements of the statute’s sponsor, (ii) admissions in Governor James’ answer to the second amended complaint, and (iii) the difference between § 16-1-20.1 and its predecessor statute.

Curiously, the opinions do not mention that all of the sponsor’s statements relied upon — including the statement “inserted” into the Senate Journal — were made after the legislature had passed the statute; indeed, the testimony that the Court finds critical was given well over a year after the statute was enacted. As even the appellees concede, see Brief for Appellees 18, there is not a shred of evidence that *87the legislature as a whole shared the sponsor’s motive or that a majority in either house was even aware of the sponsor’s view of the bill when it was passed. The sole relevance of the sponsor’s statements, therefore, is that they reflect the personal, subjective motives of a single legislator. No case in the 195-year history of this Court supports the disconcerting idea that postenactment statements by individual legislators are relevant in determining the constitutionality of legislation.

Even if an individual legislator’s after-the-fact statements could rationally be considered relevant, all of the opinions fail to mention that the sponsor also testified that one of his purposes in drafting and sponsoring the moment-of-silence bill was to clear up a widespread misunderstanding that a schoolchild is legally prohibited from engaging in silent, individual prayer once he steps inside a public school building. See App. 53-54. That testimony is at least as important as the statements the Court relies upon, and surely that testimony manifests a permissible purpose.

The Court also relies on the admissions of Governor James’ answer to the second amended complaint. Strangely, however, the Court neglects to mention that there was no trial bearing on the constitutionality of the Alabama statutes; trial became unnecessary when the District Court held that the Establishment Clause does not apply to the states.2 The absence of a trial on the issue of the constitutionality of §16-1-20.1 is significant because the answer filed by the State Board and Superintendent of Education did not make the same admissions that the Governor’s answer made. See 1 Record 187. The Court cannot know whether, if these cases had been tried, those state officials would have offered evidence to contravene appellees’ allegations concerning legislative purpose. Thus, it is completely inappropriate to accord any relevance to the admissions in the Governor’s answer.

*88The several preceding opinions conclude that the principal difference between § 16-1-20.1 and its predecessor statute proves that the sole purpose behind the inclusion of the phrase “or voluntary prayer” in § 16-1-20.1 was to endorse and promote prayer. This reasoning is simply a subtle way of focusing exclusively on the religious component of the statute rather than examining the statute as a whole. Such logic — if it can be called that — would lead the Court to hold, for example, that a state may enact a statute that provides reimbursement for bus transportation to the parents of all schoolchildren, but may not add parents of parochial school students to an existing program providing reimbursement for parents of public school students. Congress amended the statutory Pledge of Allegiance 31 years ago to add the words “under God.” Act of June 14, 1954, Pub. L. 396, 68 Stat. 249. Do the several opinions in support of the judgment today render the Pledge unconstitutional? That would be the consequence of their method of focusing on the difference between § 16-1-20.1 and its predecessor statute rather than examining § 16-1-20.1 as a whole.3 Any such holding would of course make a mockery of our decisionmaking in Establishment Clause cases. And even were the Court’s method correct, the inclusion of the words “or voluntary prayer” in § 16-1-20.1 is wholly consistent with the clearly permissible purpose of clarifying that silent, voluntary prayer is not forbidden in the public school building.4

*89(c) The Court’s extended treatment of the “test” of Lemon v. Kurtzman, 403 U. S. 602 (1971), suggests a naive preoccupation with an easy, bright-line approach for addressing constitutional issues. We have repeatedly cautioned that Lemon did not establish a rigid caliper capable of resolving every Establishment Clause issue, but that it sought only to provide “signposts.” “In each [Establishment Clause] case, the inquiry calls for line-drawing; no fixed, per se rule can be framed.” Lynch v. Donnelly, 465 U. S. 668, 678 (1984). In any event, our responsibility is not to apply tidy formulas by rote; our duty is to determine whether the statute or practice at issue is a step toward establishing a state religion. Given today’s decision, however, perhaps it is understandable that the opinions in support of the judgment all but ignore the Establishment Clause itself and the concerns that underlie it.

(d) The notion that the Alabama statute is a step toward creating an established church borders on, if it does not trespass into, the ridiculous. The statute does not remotely threaten religious liberty; it affirmatively furthers the values of religious freedom and tolerance that the Establishment Clause was designed to protect. Without pressuring those who do not wish to pray, the statute simply creates an opportunity to think, to plan, or to pray if one wishes — as Congress does by providing chaplains and chapels. It accommodates the purely private, voluntary religious choices of the individual pupils who wish to pray while at the same time creating a time for nonreligious reflection for those who do not choose to pray. The statute also provides a meaningful opportunity for schoolchildren to appreciate the absolute constitutional right of each individual to worship and believe as the individual wishes. The statute “endorses” only the view that the religious observances of others should be tolerated and, *90where possible, accommodated. If the government may not accommodate religious needs when it does so in a wholly neutral and noncoercive manner, the “benevolent neutrality” that we have long considered the correct constitutional standard will quickly translate into the “callous indifference” that the Court has consistently held the Establishment Clause does not require.

The Court today has ignored the wise admonition of Justice Goldberg that “the measure of constitutional adjudication is the ability and willingness to distinguish between real threat and mere shadow.” Abington School District v. Schempp, 374 U. S., at 308 (concurring opinion). The innocuous statute that the Court strikes down does not even rise to the level of “mere shadow.” Justice O’Connor paradoxically acknowledges: “It is difficult to discern a serious threat to religious liberty from a room of silent, thoughtful schoolchildren.” Ante, at 73.5 I would add to that, “even if they choose to pray.”

The mountains have labored and brought forth a mouse.6

dissenting.

For the most part agreeing with the opinion of The Chief Justice, I dissent from the Court’s judgment invalidating Ala. Code § 16-1-20.1 (Supp. 1984). Because I do, it is apparent that in my view the First Amendment does not proscribe either (1) statutes authorizing or requiring in so many words a moment of silence before classes begin or (2) a statute that provides, when it is initially passed, for a moment of silence for meditation or prayer. As I read the filed opin*91ions, a majority of the Court would approve statutes that provided for a moment of silence but did not mention prayer. But if a student asked whether he could pray during that moment, it is difficult to believe that the teacher could not answer in the affirmative. If that is the case, I would not invalidate a statute that at the outset provided the legislative answer to the question “May I pray?” This is so even if the Alabama statute is infirm, which I do not believe it is, because of its peculiar legislative history.

I appreciate Justice Rehnquist’s explication of the history of the Religion Clauses of the First Amendment. Against that history, it would be quite understandable if we undertook to reassess our cases dealing with these Clauses, particularly those dealing with the Establishment Clause. Of course, I have been out of step with many of the Court’s decisions dealing with this subject matter, and it is thus not surprising that I would support a basic reconsideration of our precedents.

dissenting.

Thirty-eight years ago this Court, in Everson v. Board of Education, 330 U. S. 1, 16 (1947), summarized its exegesis of Establishment Clause doctrine thus:

“In the words of Jefferson, the clause against establishment of religion by law was intended to erect ‘a wall of separation between church and State.’ Reynolds v. United States, [98 U. S. 145, 164 (1879)].”

This language from Reynolds, a case involving the Free Exercise Clause of the First Amendment rather than the Establishment Clause, quoted from Thomas Jefferson’s letter to the Danbury Baptist Association the phrase “I contemplate with sovereign reverence that act of the whole American people which declared that their legislature should ‘make no law respecting an establishment of religion, or prohibiting the free exercise thereof,’ thus building a wall of separation *92between church and State.” 8 Writings of Thomas Jefferson 113 (H. Washington ed. 1861).1

It is impossible to build sound constitutional doctrine upon a mistaken understanding of constitutional history, but unfortunately the Establishment Clause has been expressly freighted with Jefferson’s misleading metaphor for nearly 40 years. Thomas Jefferson was of course in France at the time the constitutional Amendments known as the Bill of Rights were passed by Congress and ratified by the States. His letter to the Danbury Baptist Association was a short note of courtesy, written 14 years after the Amendments were passed by Congress. He would seem to any detached observer as a less than ideal source of contemporary history as to the meaning of the Religion Clauses of the First Amendment.

Jefferson’s fellow Virginian, James Madison, with whom he was joined in the battle for the enactment of the Virginia Statute of Religious Liberty of 1786, did play as large a part as anyone in the drafting of the Bill of Rights. He had two advantages over Jefferson in this regard: he was present in the United States, and he was a leading Member of the First Congress. But when we turn to the record of the proceedings in the First Congress leading up to the adoption of the Establishment Clause of the Constitution, including Madison’s significant contributions thereto, we see a far different picture of its purpose than the highly simplified “wall of separation between church and State.”

During the debates in the Thirteen Colonies over ratification of the Constitution, one of the arguments frequently used by opponents of ratification was that without a Bill of Rights guaranteeing individual liberty the new general Gov*93ernment carried with it a potential for tyranny. The typical response to this argument on the part of those who favored ratification was that the general Government established by the Constitution had only delegated powers, and that these delegated powers were so limited that the Government would have no occasion to violate individual liberties. This response satisfied some, but not others, and of the 11 Colonies which ratified the Constitution by early 1789, 5 proposed one or another amendments guaranteeing individual liberty. Three — New Hampshire, New York, and Virginia — included in one form or another a declaration of religious freedom. See 3 J. Elliot, Debates on the Federal Constitution 659 (1891); 1 id., at 328. Rhode Island and North Carolina flatly refused to ratify the Constitution in the absence of amendments in the nature of a Bill of Rights. 1 id., at 334; 4 id., at 244. Virginia and North Carolina proposed identical guarantees of religious freedom:

“[A]ll men have an equal, natural and unalienable right to the free exercise of religion, according to the dictates of conscience, and ... no particular religious sect or society ought to be favored or established, by law, in preference to others.” 3 id., at 659; 4 id., at 244.2

On June 8, 1789, James Madison rose in the House of Representatives and “reminded the House that this was the day that he had heretofore named for bringing forward amendments to the Constitution.” 1 Annals of Cong. 424. Madison’s subsequent remarks in urging the House to adopt his drafts of the proposed amendments were less those of a dedicated advocate of the wisdom of such measures than those of a prudent statesman seeking the enactment of meas*94ures sought by a number of his fellow citizens which could surely do no harm and might do a great deal of good. He said, inter alia:

“It appears to me that this House is bound by every motive of prudence, not to let the first session pass over without proposing to the State Legislatures, some things to be incorporated into the Constitution, that will render it as acceptable to the whole people of the United States, as it has been found acceptable to a majority of them. I wish, among other reasons why something should be done, that those who had been friendly to the adoption of this Constitution may have the opportunity of proving to those who were opposed to it that they were as sincerely devoted to liberty and a Republican Government, as those who charged them with wishing the adoption of this Constitution in order to lay the foundation of an aristocracy or despotism. It will be a desirable thing to extinguish from the bosom of every member of the community, any apprehensions that there are those among his countrymen who wish to deprive them of the liberty for which they valiantly fought and honorably bled. And if there are amendments desired of such a nature as will not injure the Constitution, and they can be in-grafted so as to give satisfaction to the doubting part of our fellow-citizens, the friends of the Federal Government will evince that spirit of deference and concession for which they have hitherto been distinguished.” Id., at 431-432.

The language Madison proposed for what ultimately became the Religion Clauses of the First Amendment was this:

“The civil rights of none shall be abridged on account of religious belief or worship, nor shall any national religion be established, nor shall the full and equal rights of conscience be in any manner, or on any pretext, infringed.” Id., at 434.

*95On the same day that Madison proposed them, the amendments which formed the basis for the Bill of Rights were referred by the House to a Committee of the Whole, and after several weeks’ delay were then referred to a Select Committee consisting of Madison and 10 others. The Committee revised Madison’s proposal regarding the establishment of religion to read:

“[N]o religion shall be established by law, nor shall the equal rights of conscience be infringed.” Id., at 729.

The Committee’s proposed revisions were debated in the House on August 15, 1789. The entire debate on the Religion Clauses is contained in two full columns of the “Annals,” and does not seem particularly illuminating. See id., at 729-731. Representative Peter Sylvester of New York expressed his dislike for the revised version, because it might have a tendency “to abolish religion altogether.” Representative John Vining suggested that the two parts of the sentence be transposed; Representative Elbridge Gerry thought the language should be changed to read “that no religious doctrine shall be established by law.” Id., at 729. Roger Sherman of Connecticut had the traditional reason for opposing provisions of a Bill of Rights — that Congress had no delegated authority to “make religious establishments” — and therefore he opposed the adoption of the amendment. Representative Daniel Carroll of Maryland thought it desirable to adopt the words proposed, saying “[h]e would not contend with gentlemen about the phraseology, his object was to secure the substance in such a manner as to satisfy the wishes of the honest part of the community.”

Madison then spoke, and said that “he apprehended the meaning of the words to be, that Congress should not establish a religion, and enforce the legal observation of it by law, nor compel men to worship God in any manner contrary to their conscience.” Id., at 730. He said that some of the state conventions had thought that Congress might rely on *96the Necessary and Proper Clause to infringe the rights of conscience or to establish a national religion, and “to prevent these effects he presumed the amendment was intended, and he thought it as well expressed as the nature of the language would admit.” Ibid.

Representative Benjamin Huntington then expressed the view that the Committee’s language might “be taken in such latitude as to be extremely hurtful to the cause of religion. He understood the amendment to mean what had been expressed by the gentleman from Virginia; but others might find it convenient to put another construction upon it.” Huntington, from Connecticut, was concerned that in the New England States, where state-established religions were the rule rather than the exception, the federal courts might not be able to entertain claims based upon an obligation under the bylaws of a religious organization to contribute to the support of a minister or the building of a place of worship. He hoped that “the amendment would be made in such a way as to secure the rights of conscience, and a free exercise of the rights of religion, but not to patronise those who professed no religion at all.” Id., at 730-731.

Madison responded that the insertion of the word “national” before the word “religion” in the Committee version should satisfy the minds of those who had criticized the language. “He believed that the people feared one sect might obtain a pre-eminence, or two combine together, and establish a religion to which they would compel others to conform. He thought that if the word ‘national’ was introduced, it would point the amendment directly to the object it was intended to prevent.” Id., at 731. Representative Samuel Livermore expressed himself as dissatisfied with Madison’s proposed amendment, and thought it would be better if the Committee language were altered to read that “Congress shall make no laws touching religion, or infringing the rights of conscience.” Ibid.

Representative Gerry spoke in opposition to the use of the word “national” because of strong feelings expressed during *97the ratification debates that a federal government, not a national government, was created by the Constitution. Madison thereby withdrew his proposal but insisted that his reference to a “national religion” only referred to a national establishment and did not mean that the Goverment was a national one. The question was taken on Representative Livermore’s motion, which passed by a vote of 31 for and 20 against. Ibid.

The following week, without any apparent debate, the House voted to alter the language of the Religion Clauses to read “Congress shall make no law establishing religion, or to prevent the free exercise thereof, or to infringe the rights of conscience.” Id., at 766. The floor debates in the Senate were secret, and therefore not reported in the Annals. The Senate on September 3, 1789, considered several different forms of the Religion Amendment, and reported this language back to the House:

“Congress shall make no law establishing articles of faith or a mode of worship, or prohibiting the free exercise of religion.” C. Antieau, A. Downey, & E. Roberts, Freedom From Federal Establishment 130 (1964).

The House refused to accept the Senate’s changes in the Bill of Rights and asked for a conference; the version which emerged from the conference was that which ultimately found its way into the Constitution as a part of the First Amendment.

“Congress shall make no law respecting an establishment of religion, or prohibiting the free exercise thereof.”

The House and the Senate both accepted this language on successive days, and the Amendment was proposed in this form.

On the basis of the record of these proceedings in the House of Representatives, James Madison was undoubtedly the most important architect among the Members of the *98House of the Amendments which became the Bill of Rights, but it was James Madison speaking as an advocate of sensible legislative compromise, not as an advocate of incorporating the Virginia Statute of Religious Liberty into the United States Constitution. During the ratification debate in the Virginia Convention, Madison had actually opposed the idea of any Bill of Rights. His sponsorship of the Amendments in the House was obviously not that of a zealous believer in the necessity of the Religion Clauses, but of one who felt it might do some good, could do no harm, and would satisfy those who had ratified the Constitution on the condition that Congress propose a Bill of Rights.3 His original language “nor shall any national religion be established” obviously does not conform to the “wall of separation” between church and State idea which latter-day commentators have ascribed to him. His explanation on the floor of the meaning of his language— “that Congress should not establish a religion, and enforce the legal observation of it by law” is of the same ilk. When he replied to Huntington in the debate over the proposal which came from the Select Committee of the House, he urged that the language “no religion shall be established by law” should be amended by inserting the word “national” in front of the word “religion.”

It seems indisputable from these glimpses of Madison’s thinking, as reflected by actions on the floor of the House in 1789, that he saw the Amendment as designed to prohibit the establishment of a national religion, and perhaps to prevent discrimination among sects. He did not see it as requiring neutrality on the part of government between religion and irreligión. Thus the Court’s opinion in Everson — while correct in bracketing Madison and Jefferson together in their exertions in their home State leading to the enactment of the *99Virginia Statute of Religious Liberty — is totally incorrect in suggesting that Madison carried these views onto the floor of the United States House of Representatives when he proposed the language which would ultimately become the Bill of Rights.

The repetition of this error in the Court’s opinion in Illinois ex rel. McCollum v. Board of Education, 333 U. S. 203 (1948), and, inter alia, Engel v. Vitale, 370 U. S. 421 (1962), does not make it any sounder historically. Finally, in Abington School District v. Schempp, 374 U. S. 203, 214 (1963), the Court made the truly remarkable statement that “the views of Madison and Jefferson, preceded by Roger Williams, came to be incorporated not only in the Federal Constitution but likewise in those of most of our States” (footnote omitted). On the basis of what evidence we have, this statement is demonstrably incorrect as a matter of history.4 And its repetition in varying forms in succeeding opinions of the Court can give it no more authority than it possesses as a matter of fact; stare decisis may bind courts as to matters of law, but it cannot bind them as to matters of history.

None of the other Members of Congress who spoke during the August 15th debate expressed the slightest indication that they thought the language before them from the Select Committee, or the evil to be aimed at, would require that the Government be absolutely neutral as between religion and irreligión. The evil to be aimed at, so far as those who spoke were concerned, appears to have been the establishment of a national church, and perhaps the preference of one religious sect over another; but it was definitely not concerned about whether the Government might aid all religions evenhandedly. If one were to follow the advice of Justice Brennan, concurring in Abington School District v. Schempp, supra, at 236, and construe the Amendment in the light of what par*100ticular “practices . . . challenged threaten those consequences which the Framers deeply feared; whether, in short, they tend to promote that type of interdependence between religion and state which the First Amendment was designed to prevent,” one would have to say that the First Amendment Establishment Clause should be read no more broadly than to prevent the establishment of a national religion or the governmental preference of one religious sect over another.

The actions of the First Congress, which reenacted the Northwest Ordinance for the governance of the Northwest Territory in 1789, confirm the view that Congress did not mean that the Government should be neutral between religion and irreligión. The House of Representatives took up the Northwest Ordinance on the same day as Madison introduced his proposed amendments which became the Bill of Rights; while at that time the Federal Government was of course not bound by draft amendments to the Constitution which had not yet been proposed by Congress, say nothing of ratified by the States, it seems highly unlikely that the House of Representatives would simultaneously consider proposed amendments to the Constitution and enact an important piece of territorial legislation which conflicted with the intent of those proposals. The Northwest Ordinance, 1 Stat. 50, reenacted the Northwest Ordinance of 1787 and provided that “[r]eligion, morality, and knowledge, being necessary to good government and the happiness of mankind, schools and the means of education shall forever be encouraged.” Id., at 52, n. (a). Land grants for schools in the Northwest Territory were not limited to public schools. It was not until 1845 that Congress limited land grants in the new States and Territories to nonsectarian schools. 5 Stat. 788; C. Antieau, A. Downey, & E. Roberts, Freedom From Federal Establishment 163 (1964).

On the day after the House of Representatives voted to adopt the form of the First Amendment Religion Clauses which was ultimately proposed and ratified, Representative *101Elias Boudinot proposed a resolution asking President George Washington to issue a Thanksgiving Day Proclamation. Boudinot said he “could not think of letting the session pass over without offering an opportunity to all the citizens of the United States of joining with one voice, in returning to Almighty God their sincere thanks for the many blessings he had poured down upon them.” 1 Annals of Cong. 914 (1789). Representative Aedanas Burke objected to the resolution because he did not like “this mimicking of European customs”; Representative Thomas Tucker objected that whether or not the people had reason to be satisfied with the Constitution was something that the States knew better than the Congress, and in any event “it is a religious matter, and, as such, is proscribed to us.” Id., at 915. Representative Sherman supported the resolution “not only as a laudable one in itself, but as warranted by a number of precedents in Holy Writ: for instance, the solemn thanksgivings and rejoicings which took place in the time of Solomon, after the building of the temple, was a case in point. This example, he thought, worthy of Christian imitation on the present occasion . . . .” Ibid.

Boudinot’s resolution was carried in the affirmative on September 25, 1789. Boudinot and Sherman, who favored the Thanksgiving Proclamation, voted in favor of the adoption of the proposed amendments to the Constitution, including the Religion Clauses; Tucker, who opposed the Thanksgiving Proclamation, voted against the adoption of the amendments which became the Bill of Rights.

Within two weeks of this action by the House, George Washington responded to the Joint Resolution which by now had been changed to include the language that the President “recommend to the people of the United States a day of public thanksgiving and prayer, to be observed by acknowledging with grateful hearts the many and signal favors of Almighty God, especially by affording them an opportunity peaceably to establish a form of government for their safety and happiness.” 1 J. Richardson, Messages and Papers of *102the Presidents, 1789-1897, p. 64 (1897). The Presidential Proclamation was couched in these words:

“Now, therefore, I do recommend and assign Thursday, the 26th day of November next, to be devoted by the people of these States to the service of that great and glorious Being who is the beneficent author of all the good that was, that is, or that will be; that we may then all unite in rendering unto Him our sincere and humble thanks for His kind care and protection of the people of this country previous to their becoming a nation; for the signal and manifold mercies and the favorable inter-positions of His providence in the course and conclusion of the late war; for the great degree of tranquillity, union, and plenty which we have since enjoyed; for the peaceable and rational manner in which we have been enabled to establish constitutions of government for our safety and happiness, and particularly the national one now lately instituted; for the civil and religious liberty with which we are blessed, and the means we have of acquiring and diffusing useful knowledge; and, in general, for all the great and various favors which He has been pleased to confer upon us.

“And also that we may then unite in most humbly offering our prayers and supplications to the great Lord and Ruler of Nations, and beseech Him to pardon our national and other transgressions; to enable us all, whether in public or private stations, to perform our several and relative duties properly and punctually; to render our National Government a blessing to all the people by constantly being a Government of wise, just, and constitutional laws, discreetly and faithfully executed and obeyed; to protect and guide all sovereigns and nations (especially such as have shown kindness to us), and to bless them with good governments, peace, and concord; to promote the knowledge and practice of true religion and virtue, and the increase of science among them and *103us; and, generally, to grant unto all mankind such a degree of temporal prosperity as He alone knows to be best.” Ibid.

George Washington, John Adams, and James Madison all issued Thanksgiving Proclamations; Thomas Jefferson did not, saying:

“Fasting and prayer are religious exercises; the enjoining them an act of discipline. Every religious society has a right to determine for itself the times for these exercises, and the objects proper for them, according to their own particular tenets; and this right can never be safer than in their own hands, where the Constitution has deposited it.” 11 Writings of Thomas Jefferson 429 (A. Lipscomb ed. 1904).

As the United States moved from the 18th into the 19th century, Congress appropriated time and again public moneys in support, of sectarian Indian education carried on by religious organizations. Typical of these was Jefferson’s treaty with the Kaskaskia Indians, which provided annual cash support for the Tribe’s Roman Catholic priest and church.5 It was not until 1897, when aid to sectarian edu*104cation for Indians had reached $500,000 annually, that Congress decided thereafter to cease appropriating money for education in sectarian schools. See Act of June 7, 1897, 30 Stat. 62, 79; cf. Quick Bear v. Leupp, 210 U. S. 50, 77-79 (1908); J. O’Neill, Religion and Education Under the Constitution 118-119 (1949). See generally R. Cord, Separation of Church and State 61-82 (1982). This history shows the fallacy of the notion found in Everson that “no tax in any amount” may be levied for religious activities in any form. 330 U. S., at 15-16.

Joseph Story, a Member of this Court from 1811 to 1845, and during much of that time a professor at the Harvard Law School, published by far the most comprehensive treatise on the United States Constitution that had then appeared. Volume 2 of Story’s Commentaries on the Constitution of the United States 630-632 (5th ed. 1891) discussed the meaning of the Establishment Clause of the First Amendment this way:

“Probably at the time of the adoption of the Constitution, and of the amendment to it now under consideration [First Amendment], the general if not the universal sentiment in America was, that Christianity ought to receive encouragement from the State so far as was not incompatible with the private rights of conscience and the freedom of religious worship. An attempt to level all religions, and to make it a matter of state policy to hold all in utter indifference, would have created universal disapprobation, if not universal indignation.

“The real object of the [First] [A]mendment was not to countenance, much less to advance, Mahometanism, or Judaism, or infidelity, by prostrating Christianity; but to exclude all rivalry among Christian sects, and to prevent *105any national ecclesiastical establishment which should give to a hierarchy the exclusive patronage of the national government. It thus cut off the means of religious persecution (the vice and pest of former ages), and of the subversion of the rights of conscience in matters of religion, which had been trampled upon almost from the days of the Apostles to the present age. . . .” (Footnotes omitted.)

Thomas Cooley’s eminence as a legal authority rivaled that of Story. Cooley stated in his treatise entitled Constitutional Limitations that aid to a particular religious sect was prohibited by the United States Constitution, but he went on to say:

“But while thus careful to establish, protect, and defend religious freedom and equality, the American constitutions contain no provisions which prohibit the authorities from such solemn recognition of a superintending Providence in public transactions and exercises as the general religious sentiment of mankind inspires, and as seems meet and proper in finite and dependent beings. Whatever may be the shades of religious belief, all must acknowledge the fitness of recognizing in important human affairs the superintending care and control of the Great Governor of the Universe, and of acknowledging with thanksgiving his boundless favors, or bowing in contrition when visited with the penalties of his broken laws. No principle of constitutional law is violated when thanksgiving or fast days are appointed; when chaplains are designated for the army and navy; when legislative sessions are opened with prayer or the reading of the Scriptures, or when religious teaching is encouraged by a general exemption of the houses of religious worship from taxation for the support of State government. Undoubtedly the spirit of the Constitution will require, in all these cases, that care be taken to avoid discrimination *106in favor of or against any one religious denomination or sect; but the power to do any of these things does not become unconstitutional simply because of its susceptibility to abuse. . . Id., at *470-*471.

Cooley added that

“[t]his public recognition of religious worship, however, is not based entirely, perhaps not even mainly, upon a sense of what is due to the Supreme Being himself as the author of all good and of all law; but the same reasons of state policy which induce the government to aid institutions of charity and seminaries of instruction will incline it also to foster religious worship and religious institutions, as conservators of the public morals and valuable, if not indispensable, assistants to the preservation of the public order.” Id., at *470.

It would seem from this evidence that the Establishment Clause of the First Amendment had acquired a well-accepted meaning: it forbade establishment of a national religion, and forbade preference among religious sects or denominations. Indeed, the first American dictionary defined the word “establishment” as “the act of establishing, founding, ratifying or ordaining,” such as in “[t]he episcopal form of religion, so called, in England.” 1 N. Webster, American Dictionary of the English Language (1st ed. 1828). The Establishment Clause did not require government neutrality between religion and irreligión nor did it prohibit the Federal Government from providing nondiscriminatory aid to religion. There is simply no historical foundation for the proposition that the Framers intended to build the “wall of separation” that was constitutionalized in Everson.

Notwithstanding the absence of a historical basis for this theory of rigid separation, the wall idea might well have served as a useful albeit misguided analytical concept, had it led this Court to unified and principled results in Establishment Clause cases. The opposite, unfortunately, has been *107true; in the 38 years since Everson our Establishment Clause cases have been neither principled nor unified. Our recent opinions, many of them hopelessly divided pluralities,6 have with embarrassing candor conceded that the “wall of separation” is merely a “blurred, indistinct, and variable barrier,” which “is not wholly accurate” and can only be “dimly perceived.” Lemon v. Kurtzman, 403 U. S. 602, 614 (1971); Tilton v. Richardson, 403 U. S. 672, 677-678, (1971); Wolman v. Walter, 433 U. S. 229, 236 (1977); Lynch v. Donnelly, 465 U. S. 668, 673 (1984).

Whether due to its lack of historical support or its practical unworkability, the Everson “wall” has proved all but useless as a guide to sound constitutional adjudication. It illustrates only too well the wisdom of Benjamin Cardozo’s observation that “[mjetaphors in law are to be narrowly watched, for starting as devices to liberate thought, they end often by enslaving it.” Berkey v. Third Avenue R. Co., 244 N. Y. 84, 94, 155 N. E. 58, 61 (1926).

But the greatest injury of the “wall” notion is its mischievous diversion of judges from the actual intentions of the drafters of the Bill of Rights. The “crucible of litigation,” ante, at 52, is well adapted to adjudicating factual disputes on the basis of testimony presented in court, but no amount of repetition of historical errors in judicial opinions can make the errors true. The “wall of separation between church and State” is a metaphor based on bad history, a metaphor which has proved useless as a guide to judging. It should be frankly and explicitly abandoned.

*108The Court has more recently attempted to add some mortar to Everson’s wall through the three-part test of Lemon v. Kurtzman, supra, at 614-615, which served at first to offer a more useful test for purposes of the Establishment Clause than did the “wall” metaphor. Generally stated, the Lemon test proscribes state action that has a sectarian purpose or effect, or causes an impermissible governmental entanglement with religion.

Lemon cited Board of Education v. Allen, 392 U. S. 236, 243 (1968), as the source of the “purpose” and “effect” prongs of the three-part test. The Allen opinion explains, however, how it inherited the purpose and effect elements from Schempp and Everson, both of which contain the historical errors described above. See Allen, supra, at 243. Thus the purpose and effect prongs have the same historical deficiencies as the wall concept itself: they are in no way based on either the language or intent of the drafters.

The secular purpose prong has proved mercurial in application because it has never been fully defined, and we have never fully stated how the test is to operate. If the purpose prong is intended to void those aids to sectarian institutions accompanied by a stated legislative purpose to aid religion, the prong will condemn nothing so long as the legislature utters a secular purpose and says nothing about aiding religion. Thus the constitutionality of a statute may depend upon what the legislators put into the legislative history and, more importantly, what they leave out. The purpose prong means little if it only requires the legislature to express any secular purpose and omit all sectarian references, because legislators might do just that. Faced with a valid legislative secular purpose, we could not properly ignore that purpose without a factual basis for doing so. Larson v. Valente, 456 U. S. 228, 262-263 (1982) (White, J., dissenting).

However, if the purpose prong is aimed to void all statutes enacted with the intent to aid sectarian institutions, whether stated or not, then most statutes providing any aid, such as *109textbooks or bus rides for sectarian school children, will fail because one of the purposes behind every statute, whether stated or not, is to aid the target of its largesse. In other words, if the purpose prong requires an absence of any intent to aid sectarian institutions, whether or not expressed, few state laws in this area could pass the test, and we would be required to void some state aids to religion which we have already upheld. E. g., Allen, supra.

The entanglement prong of the Lemon test came from Walz v. Tax Comm’n, 397 U. S. 664, 674 (1970). Walz involved a constitutional challenge to New York’s time-honored practice of providing state property tax exemptions to church property used in worship. The Walz opinion refused to “undermine the ultimate constitutional objective [of the Establishment Clause] as illuminated by history,” id., at 671, and upheld the tax exemption. The Court examined the historical relationship between the State and church when church property was in issue, and determined that the challenged tax exemption did not so entangle New York with the church as to cause an intrusion or interference with religion. Interferences with religion should arguably be dealt with under the Free Exercise Clause, but the entanglement inquiry in Walz was consistent with that case’s broad survey of the relationship between state taxation and religious property.

We have not always followed Walz’ reflective inquiry into entanglement, however. E. g., Wolman, supra, at 254. One of the difficulties with the entanglement prong is that, when divorced from the logic of Walz, it creates an “in-soluable paradox” in school aid cases: we have required aid to parochial schools to be closely watched lest it be put to sectarian use, yet this Qlose supervision itself will create an entanglement. Roemer v. Maryland Bd. of Public Works, 426 U. S. 736, 768-769 (1976) (White, J., concurring in judgment). For example, in Wolman, supra, the Court in part struck the State’s nondiscriminatory provision of buses for parochial school field trips, because the state supervision *110of sectarian officials in charge of field trips would be too onerous. This type of self-defeating result is certainly not required to ensure that States do not establish religions.

The entanglement test as applied in cases like Wolman also ignores the myriad state administrative regulations properly placed upon sectarian institutions such as curriculum, attendance, and certification requirements for sectarian schools, or fire and safety regulations for churches. Avoiding entanglement between church and State may be an important consideration in a case like Walz, but if the entanglement prong were applied to all state and church relations in the automatic manner in which it has been applied to school aid cases, the State could hardly require anything of church-related institutions as a condition for receipt of financial assistance.

These difficulties arise because the Lemon test has no more grounding in the history of the First Amendment than does the wall theory upon which it rests. The three-part test represents a determined effort to craft a workable rule from a historically faulty doctrine; but the rule can only be as sound as the doctrine it attempts to service. The three-part test has simply not provided adequate standards for deciding Establishment Clause cases, as this Court has slowly come to realize. Even worse, the Lemon test has caused this Court to fracture into unworkable plurality opinions, see n. 6, supra, depending upon how each of the three factors applies to a certain state action. The results from our school services cases show the difficulty we have encountered in making the Lemon test yield principled results.

For example, a State may lend to parochial school children geography textbooks7 that contain maps of the United States, but the State may not lend maps of the United States for use in geography class.8 A State may lend textbooks on American colonial history, but it may not lend a film on *111George Washington, or a film projector to show it in history-class. A State may lend classroom workbooks, but may not lend workbooks in which the parochial school children write, thus rendering them nonreusable.9 A State may pay for bus transportation to religious schools10 but may not pay for bus transportation from the parochial school to the public zoo or natural history museum for a field trip.11 A State may pay for diagnostic services conducted in the parochial school but therapeutic services must be given in a different building; speech and hearing “services” conducted by the State inside the sectarian school are forbidden, Meek v. Pittenger, 421 U. S. 349, 367, 371 (1975), but the State may conduct speech and hearing diagnostic testing inside the sectarian school. Wolman, 433 U. S., at 241. Exceptional parochial school students may receive counseling, but it must take place outside of the parochial school,12 such as in a trailer parked down the street. Id., at 245. A State may give cash to a parochial school to pay for the administration of state-written tests and state-ordered reporting services,13 but it may not provide funds for teacher-prepared tests on secular subjects.14 Religious instruction may not be given in public school,15 but the public school may release students during the day for religion classes elsewhere, and may enforce attendance at those classes with its truancy laws.16

These results violate the historically sound principle “that the Establishment Clause does not forbid governments . . . to [provide] general welfare under which benefits are distributed to private individuals, even though many of those indi*112viduals may elect to use those benefits in ways that ‘aid’ religious instruction or worship.” Committee for Public Education & Religious Liberty v. Nyquist, 413 U. S. 756, 799 (1973) (Burger, C. J., concurring in part and dissenting in part). It is not surprising in the light of this record that our most recent opinions have expressed doubt on the usefulness of the Lemon test.

Although the test initially provided helpful assistance, e. g., Tilton v. Richardson, 403 U. S. 672 (1971), we soon began describing the test as only a “guideline,” Committee for Public Education & Religious Liberty v. Nyquist, supra, and lately we have described it as “no more than [a] useful signpos[t].” Mueller v. Allen, 463 U. S. 388, 394 (1983), citing Hunt v. McNair, 413 U. S. 734, 741 (1973); Larkin v. Grendel’s Den, Inc., 459 U. S. 116 (1982). We have noted that the Lemon test is “not easily applied,” Meek, supra, at 358, and as Justice White noted in Committee for Public Education & Religious Liberty v. Regan, 444 U. S. 646 (1980), under the Lemon test we have “sacrifice[d] clarity and predictability for flexibility.” 444 U. S., at 662. In Lynch we reiterated that the Lemon test has never been binding on the Court, and we cited two cases where we had declined to apply it. 465 U. S., at 679, citing Marsh v. Chambers, 463 U. S. 783 (1983); Larson v. Valente, 456 U. S. 228 (1982).

If a constitutional theory has no basis in the history of the amendment it seeks to interpret, is difficult to apply and yields unprincipled results, I see little use in it. The “crucible of litigation,” ante, at 52, has produced only consistent unpredictability, and today’s effort is just a continuation of “the sisyphean task of trying to patch together the ‘blurred, indistinct and variable barrier’ described in Lemon v. Kurtzman.” Regan, supra, at 671 (Stevens, J., dissenting). We have done much straining since 1947, but still we admit that we can only “dimly perceive” the Everson wall. Tilton, supra. Our perception has been clouded not by the Constitution but by the mists of an unnecessary metaphor.

*113The true meaning of the Establishment Clause can only be seen in its history. See Walz, 397 U. S., at 671-673; see also Lynch, supra, at 673-678. As drafters of our Bill of Rights, the Framers inscribed the principles that control today. Any deviation from their intentions frustrates the permanence of that Charter and will only lead to the type of unprincipled decisionmaking that has plagued our Establishment Clause cases since Everson.

The Framers intended the Establishment Clause to prohibit the designation of any church as a “national” one. The Clause was also designed to stop the Federal Government from asserting a preference for one religious denomination or sect over others. Given the “incorporation” of the Establishment Clause as against the States via the Fourteenth Amendment in Everson, States are prohibited as well from establishing a religion or discriminating between sects. As its history abundantly shows, however, nothing in the Establishment Clause requires government to be strictly neutral between religion and irreligión, nor does that Clause prohibit Congress or the States from pursuing legitimate secular ends through nondiscriminatory sectarian means.

The Court strikes down the Alabama statute because the State wished to “characterize prayer as a favored practice.” Ante, at 60. It would come as much of a shock to those who drafted the Bill of Rights as it will to a large number of thoughtful Americans today to learn that the Constitution, as construed by the majority, prohibits the Alabama Legislature from “endorsing” prayer. George Washington himself, at the request of the very Congress which passed the Bill of Rights, proclaimed a day of “public thanksgiving and prayer, to be observed by acknowledging with grateful hearts the many and signal favors of Almighty God.” History must judge whether it was the Father of his Country in 1789, or a majority of the Court today, which has strayed from the meaning of the Establishment Clause.

The State surely has a secular interest in regulating the manner in which public schools are conducted. Nothing in *114the Establishment Clause of the First Amendment, properly-understood, prohibits any such generalized “endorsement” of prayer. I would therefore reverse the judgment of the Court of Appeals.

delivered the opinion of the Court.

At an early stage of this litigation, the constitutionality of three Alabama statutes was questioned: (1) § 16-1-20, enacted in 1978, which authorized a 1-minute period of silence in all public schools “for meditation”;1 (2) § 16-1-20.1, enacted in 1981, which authorized a period of silence “for meditation or voluntary prayer”;2 and (3) §16-1-20.2, enacted in 1982, which authorized teachers to lead “willing students” in a prescribed prayer to “Almighty God . . . the Creator and Supreme Judge of the world.”3

*41At the preliminary-injunction stage of this case, the District Court distinguished § 16-1-20 from the other two statutes. It then held that there was “nothing wrong” with §16-1-20.4 but that §§16-1-20.1 and 16-1-20.2 were both invalid because the sole purpose of both was “an effort on the part of the State of Alabama to encourage a religious activity.”5 After the trial on the merits, the District Court did not change its interpretation of these two statutes, but held that they were constitutional because, in its opinion, Alabama has the power to establish a state religion if it chooses to do so.6

The Court of Appeals agreed with the District Court’s initial interpretation of the purpose of both § 16-1-20.1 and § 16-1-20.2, and held them both unconstitutional.7 We have already affirmed the Court of Appeals’ holding with respect to §16-l-20.2.8 Moreover, appellees have not questioned the holding that § 16-1-20 is valid.9 Thus, the narrow question for decision is whether §16-1-20.1, which authorizes a period of silence for “meditation or voluntary prayer,” is a *42law respecting the establishment of religion within the meaning of the First Amendment.10

I — I

Appellee Ishmael Jaffree is a resident of Mobile County, Alabama. On May 28,1982, he filed a complaint on behalf of three of his minor children; two of them were second-grade students and the third was then in kindergarten. The complaint named members of the Mobile County School Board, various school officials, and the minor plaintiffs’ three teachers as defendants.11 The complaint alleged that the appellees brought the action “seeking principally a declaratory judgment and an injunction restraining the Defendants and each of them from maintaining or allowing the maintenance of regular religious prayer services or other forms of religious observances in the Mobile County Public Schools in violation of the First Amendment as made applicable to states by the Fourteenth Amendment to the United States Constitution.”12 The complaint further alleged that two of the children had been subjected to various acts of religious indoctrination “from the beginning of the school year in September, 1981”;13 that the defendant teachers had “on a daily basis” led their classes in saying certain prayers in unison;14 that the minor children were exposed to ostracism from their peer group class members if they did not participate;16 and that Ishmael Jaffree had repeatedly but unsuccessfully requested that the devotional services be stopped. The original complaint made no reference to any Alabama statute.

*43On June 4, 1982, appellees filed an amended complaint seeking class certification,16 and on June 30, 1982, they filed a second amended complaint naming the Governor of Alabama and various state officials as additional defendants. In that amendment the appellees challenged the constitutionality of three Alabama statutes: §§16-1-20, 16-1-20.1, and 16-1-20.2.17

On August 2, 1982, the District Court held an evidentiary hearing on appellees’ motion for a preliminary injunction. At that hearing, State Senator Donald G. Holmes testified that he was the “prime sponsor” of the bill that was enacted in 1981 as § 16-1-20.1.18 He explained that the bill was an “effort to return voluntary prayer to our public schools ... it is a beginning and a step in the right direction.”19 Apart from the purpose to return voluntary prayer to public school, Senator Holmes unequivocally testified that he had “no other purpose in mind.”20 A week after the hearing, the District Court entered a preliminary injunction.21 The court held that appellees were likely to prevail on the merits because the enactment of §§ 16-1-20.1 and 16-1-20.2 did not reflect a clearly secular purpose.22

*44In November 1982, the District Court held a 4-day trial on the merits. The evidence related primarily to the 1981-1982 academic year — the year after the enactment of § 16 — 1— 20.1 and prior to the enactment of §16-1-20.2. The District Court found that during that academic year each of the minor plaintiffs’ teachers had led classes in prayer activities, even after being informed of appellees’ objections to these activities.23

In its lengthy conclusions of law, the District Court reviewed a number of opinions of this Court interpreting the *45Establishment Clause of the First Amendment, and then embarked on a fresh examination of the question whether the First Amendment imposes any barrier to the establishment of an official religion by the State of Alabama. After reviewing at length what it perceived to be newly discovered historical evidence, the District Court concluded that “the establishment clause of the first amendment to the United States Constitution does not prohibit the state from establishing a religion.”24 In a separate opinion, the District Court dismissed appellees’ challenge to the three Alabama statutes because of a failure to state any claim for which relief could be granted. The court’s dismissal of this challenge was also based on its conclusion that the Establishment Clause did not bar the States from establishing a religion.25

*46The Court of Appeals consolidated the two cases; not surprisingly, it reversed. The Court of Appeals noted that this Court had considered and had rejected the historical argu*47ments that the District Court found persuasive, and that the District Court had misapplied the doctrine of stare decisis.26 The Court of Appeals then held that the teachers’ religious activities violated the Establishment Clause of the First Amendment.27 With respect to § 16-1-20.1 and § 16-1-20.2, the Court of Appeals stated that “both statutes advance and encourage religious activities.”28 The Court of Appeals then quoted with approval the District Court’s finding that § 16-1-20.1, and §16-1-20.2, were efforts “‘to encourage a religious activity. Even though these statutes are permissive in form, it is nevertheless state involvement respecting an establishment of religion.’”29 Thus, the Court of Appeals concluded that both statutes were “specifically the type which the Supreme Court addressed in Engel [v. Vitale, 370 U. S. 421 (1962)].”30

*48A suggestion for rehearing en banc was denied over the dissent of four judges who expressed the opinion that the full court should reconsider the panel decision insofar as it held § 16-1-20.1 unconstitutional.31 When this Court noted probable jurisdiction, it limited argument to the question that those four judges thought worthy of reconsideration. The judgment of the Court of Appeals with respect to the other issues presented by the appeals was affirmed. Wallace v. Jaffree, 466 U. S. 924 (1984).

hH I — i

Our unanimous affirmance of the Court of Appeals’ judgment concerning § 16-1-20.2 makes it unnecessary to comment at length on the District Court’s remarkable conclusion that the Federal Constitution imposes no obstacle to Alabama’s establishment of a state religion. Before analyzing the precise issue that is presented to us, it is nevertheless appropriate to recall how firmly embedded in our constitutional jurisprudence is the proposition that the several States have no greater power to restrain the individual freedoms *49protected by the First Amendment than does the Congress of the United States.

As is plain from its text, the First Amendment was adopted to curtail the power of Congress to interfere with the individual’s freedom to believe, to worship, and to express himself in accordance with the dictates of his own conscience.32 Until the Fourteenth Amendment was added to the Constitution, the First Amendment’s restraints on the exercise of federal power simply did not apply to the States.33 But when the Constitution was amended to prohibit any State from depriving any person of liberty without due process of law, that Amendment imposed the same substantive limitations on the States’ power to legislate that the First Amendment had always imposed on the Congress’ power. This Court has confirmed and endorsed this elementary proposition of law time and time again.34

*50Writing for a unanimous Court in Cantwell v. Connecticut, 310 U. S. 296, 303 (1940), Justice Roberts explained:

. . We hold that the statute, as construed and applied to the appellants, deprives them of their liberty without due process of law in contravention of the Fourteenth .Amendment. The fundamental concept of liberty embodied in that Amendment embraces the liberties guaranteed by the First Amendment. The First Amendment declares that Congress shall make no law respecting an establishment of religion or prohibiting the free exercise thereof. The Fourteenth Amendment has rendered the legislatures of the states as incompetent as Congress to enact such laws. The constitutional inhibition of legislation on the subject of religion has a double aspect. On the one hand, it forestalls compulsion by law of the acceptance of any creed or the practice of any form of worship. Freedom of conscience and freedom to adhere to such religious organization or form of worship as the individual may choose cannot be restricted by law. On the other hand, it safeguards the free exercise of the chosen form of religion.”

Cantwell, of course, is but one case in which the Court has identified the individual’s freedom of conscience as the central liberty that unifies the various Clauses in the First Amendment.35 Enlarging on this theme, The Chief Justice recently wrote:

*51“We begin with the proposition that the right of freedom of thought protected by the First Amendment against state action includes both the right to speak freely and the right to refrain from speaking at all. See Board of Education v. Barnette, 319 U. S. 624, 633-634 (1943); id., at 646 (Murphy, J., concurring). A system which secures the right to proselytize religious, political, and ideological causes must also guarantee the concomitant right to decline to foster such concepts. The right to speak and the right to refrain from speaking are complementary components of the broader concept of ‘individual freedom of mind.’ Id., at 637.

“The Court in Barnette, supra, was faced with a state statute which required public school students to participate in daily public ceremonies by honoring the flag both with words and traditional salute gestures. In overruling its prior decision in Minersville District v. Gobitis, 310 U. S. 586 (1940), the Court held that ‘a ceremony so touching matters of opinion and political attitude may [not] be imposed upon the individual by official authority under powers committed to any political organization under our Constitution.’ 319 U. S., at 636. Compelling the affirmative act of a flag salute involved a more serious infringement upon personal liberties than the passive act of carrying the state motto on a license plate, but the difference is essentially one of degree. Here, as in Barnette, we are faced with a state measure which forces an individual, as part of his daily life — indeed constantly while his automobile is in public view — to be an *52instrument for fostering public adherence to an ideological point of view he finds unacceptable. In doing so, the State ‘invades the sphere of intellect and spirit which it is the purpose of the First Amendment to our Constitution to reserve from all official control.’ Id., at 642.” Wooley v. Maynard, 430 U. S. 705, 714-715 (1977).

Just as the right to speak and the right to refrain from speaking are complementary components of a broader concept of individual freedom of mind, so also the individual’s freedom to choose his own creed is the counterpart of his right to refrain from accepting the creed established by the majority. At one time it was thought that this right merely proscribed the preference of one Christian sect over another, but would not require equal respect for the conscience of the infidel, the atheist, or the adherent of a non-Christian faith such as Islam or Judaism.36 But when the underlying principle has been examined in the crucible of litigation, the *53Court has unambiguously concluded that the individual freedom of conscience protected by the First Amendment embraces the right to select any religious faith or none at all.37 This conclusion derives support not only from the interest in respecting the individual’s freedom of conscience, but also from the conviction that religious beliefs worthy of respect are the product of free and voluntary choice by the faithful,38 *54and from recognition of the fact that the political interest in forestalling intolerance extends beyond intolerance among Christian sects — or even intolerance among “religions” — to encompass intolerance of the disbeliever and the uncertain.39 *55As Justice Jackson eloquently stated in West Virginia Board of Education v. Barnette, 319 U. S. 624, 642 (1943):

“If there is any fixed star in our constitutional constellation, it is that no official, high or petty, can prescribe what shall be orthodox in politics, nationalism, religion, or other matters of opinion or force citizens to confess by word or act their faith therein.”

The State of Alabama, no less than the Congress of the United States, must respect that basic truth.

III

When the Court has been called upon to construe the breadth of the Establishment Clause, it has examined the criteria developed over a period of many years. Thus, in Lemon v. Kurtzman, 403 U. S. 602, 612-613 (1971), we wrote:

“Every analysis in this area must begin with consideration of the cumulative criteria developed by the Court over many years. Three such tests may be gleaned from our cases. First, the statute must have a secular legislative purpose; second, its principal or primary effect must be one that neither advances nor inhibits religion, Board of Education v. Allen, 392 U. S. 236, 243 (1968); finally, the statute must not foster ‘an excessive *56government entanglement with religion.’ Walz [v. Tax Comm’n, 397 U. S. 664, 674 (1970)].”

It is the first of these three criteria that is most plainly implicated by this case. As the District Court correctly recognized, no consideration of the second or third criteria is necessary if a statute does not have a clearly secular purpose.40 For even though a statute that is motivated in part by a religious purpose may satisfy the first criterion, see, e. g., Abington School District v. Schempp, 374 U. S. 203, 296-303 (1963) (Brennan, J., concurring), the First Amendment requires that a statute must be invalidated if it is entirely motivated by a purpose to advance religion.41

In applying the purpose test, it is appropriate to ask “whether government’s actual purpose is to endorse or disapprove of religion.”42 In this case, the answer to that question is dispositive. For the record not only provides us with an unambiguous affirmative answer, but it also reveals that the enactment of § 16-1-20.1 was not motivated by any clearly secular purpose — indeed, the statute had no secular purpose.

IV

The sponsor of the bill that became §16-1-20.1, Senator Donald Holmes, inserted into the legislative record — appar*57ently without dissent — a statement indicating that the legislation was an “effort to return voluntary prayer” to the public schools.43 Later Senator Holmes confirmed this purpose before the District Court. In response to the question whether he had any purpose for the legislation other than returning voluntary prayer to public schools, he stated: “No, I did not have no other purpose in mind.”44 The State did not present evidence of any secular purpose.45

*58The unrebutted evidence of legislative intent contained in the legislative record and in the testimony of the sponsor of § 16-1-20.1 is confirmed by a consideration of the relationship between this statute and the two other measures that were considered in this case. The District Court found that the 1981 statute and its 1982 sequel had a common, nonsecular purpose. The wholly religious character of the later enactment is plainly evident from its text. When the differences between § 16-1-20.1 and its 1978 predecessor, § 16-1-20, are examined, it is equally clear that the 1981 statute has the same wholly religious character.

There are only three textual differences between § 16— 1-20.1 and §16-1-20: (1) the earlier statute applies only to grades one through six, whereas §16-1-20.1 applies to all grades; (2) the earlier statute uses the word “shall” whereas § 16-1-20.1 uses the word “may”; (3) the earlier statute refers *59only to “meditation” whereas § 16-1-20.1 refers to “meditation or voluntary prayer.” The first difference is of no relevance in this litigation because the minor appellees were in kindergarten or second grade during the 1981-1982 academic year. The second difference would also have no impact on this litigation because the mandatory language of § 16-1-20 continued to apply to grades one through six.46 Thus, the only significant textual difference is the addition of the words “or voluntary prayer.”

The legislative intent to return prayer to the public schools is, of course, quite different from merely protecting every student’s right to engage in voluntary prayer during an appropriate moment of silence during the schoolday. The 1978 statute already protected that right, containing nothing that prevented any student from engaging in voluntary prayer during a silent minute of meditation.47 Appellants have not identified any secular purpose that was not fully served by § 16-1-20 before the enactment of § 16-1-20.1. Thus, only two conclusions are consistent with the text of § 16 — 1—20.1: (1) the statute was enacted to convey a message of state endorsement and promotion of prayer; or (2) the statute was enacted for no purpose. No one suggests that the statute was nothing but a meaningless or irrational act.48

We must, therefore, conclude that the Alabama Legislature intended to change existing law49 and that it was moti*60vated by the same purpose that the Governor’s answer to the second amended complaint expressly admitted; that the statement inserted in the legislative history revealed; and that Senator Holmes’ testimony frankly described. The legislature enacted §16-1-20.1, despite the existence of § 16— 1-20 for the sole purpose of expressing the State’s endorsement of prayer activities for one minute at the beginning of each schoolday. The addition of “or voluntary prayer” indicates that the State intended to characterize prayer as a favored practice. Such an endorsement is not consistent with the established principle that the government must pursue a course of complete neutrality toward religion.50

The importance of that principle does not permit us to treat this as an inconsequential case involving nothing more than a few words of symbolic speech on behalf of the political majority.51 For whenever the State itself speaks on a religious *61subject, one of the questions that we must ask is “whether the government intends to convey a message of endorsement or disapproval of religion.”52 The well-supported concurrent findings of the District Court and the Court of Appeals — that § 16-1-20.1 was intended to convey a message of state approval of prayer activities in the public schools — make it unnecessary, and indeed inappropriate, to evaluate the practical significance of the addition of the words “or voluntary prayer” to the statute. Keeping in mind, as we must, “both the fundamental place held by the Establishment Clause in our constitutional scheme and the myriad, subtle ways in which Establishment Clause values can be eroded,”53 we conclude that §16-1-20.1 violates the First Amendment.

The judgment of the Court of Appeals is affirmed.

It is so ordered.