Although not primarily concerned with the Newdow case, Richard Ellis's book, To the Flag: The Unlikely History of the Pledge of Allegiance, sheds some light on it, and in particular on Justice O'Connor's opinion. For more than a generation, the Court's Establishment Clause rulings have been a series of ever-changing doctrines or tests, each supposedly explicating the First Amendment's words but each identified, more tellingly, with its author's words in his or her controlling judicial opinion. In its much criticized 2002 decision, the Ninth Circuit Court of Appeals employed O'Conner's so-called "endorsement" test, first announced by her in 1984 in Lynch v. Donnelly, holding that the phrase "under God" unconstitutionally endorses monotheism. To void that result, O'Connor devised a new four-part subtest to govern "ceremonial deism." Under her amended doctrine, actions like the Pledge pass constitutional muster if they: 1) are historical and ubiquitous, 2) are not worship or prayerful, 3) do not refer to a particular religion, and 4) involve minimal religious content.
O'Connor had to revise her approach if she wanted to uphold the Pledge because, as Ellis's history reveals, Congress clearly meant to endorse religion when it added the words "under God." Ellis, a professor of politics at Willamette University, traces the creation of the Pledge to Francis Bellamy, a Christian socialist who drafted a recitation for use in a nationwide 1892 Columbus Day celebration. Bellamy's original text—which read, "I pledge allegiance to my Flag and to the republic for which it stands—one Nation indivisible—with Liberty and Justice for all" —was revised in 1923 by a national flag conference. Congress then inserted "under God" in 1954 as part of an ongoing series of bipartisan steps intended, in Ellis's words, to "inject religious faith into public life." In 1952, Congress had mandated an annual National Day of Prayer. In 1955, Congress placed "In God We Trust" on paper money (it had been on coins since Lincoln's presidency). In 1956, the same words became the official national motto, replacing "E Pluribus Unum." None of these actions could have passed an honest application of O'Connor's original endorsement test.
Ellis explains that, with the emergence of the Soviet Union, Congress became concerned about the religious character of the American people. He reminds us that in the early years of the Cold War, many citizens saw Communism as a potentially dangerous rival for the American people's affections. Congress sought to strengthen American patriotism by distinguishing our attachment to God from atheistic Communism's rejection of Him. In 1955 President Eisenhower perhaps most clearly expressed the spirit of the "under God" movement when he declared in a nationally broadcast address, "Without God there could be no American form of Government, nor an American way of life. Recognition of the Supreme Being is the first—the most basic—expression of Americanism."
It's hard to imagine a statement that more flagrantly violates the spirit of Justice O'Connor's endorsement test; yet such expressions have always been part of American political rhetoric. Even Thomas Jefferson—who among the founders was perhaps the most skeptical of the salutary effects of religion—asked, "Can the liberties of a nation be thought secure when we have removed their only firm basis, a conviction in the minds of the people that these liberties are the gift of God?"
Notice that Jefferson's and Eisenhower's statements, like the phrase "under God" itself, contain a certain ambivalence. Recognition of God may be the most basic expression of Americanism because we believe America to be a part of God's providential order—we are "under God" in the sense that we are a chosen people. Alternatively, recognition of God might express our understanding that a higher law exists and that that law limits the scope and authority of our law—America is "under God" in the sense of being subject to His judgment. In this latter sense, recognition of God would lead America, unlike the Soviet Union, say, to limit political power and to respect the rights with which human beings are endowed by their Creator. Consequently, governmental recognition and even endorsement of God do not necessarily enhance political power or lead to the union of church and state. As Tocqueville observed long ago, religion can increase democratic citizens' affection for their nation while also limiting their demands on it.
Unlike Tocqueville, Justice O'Connor seems only dimly aware that self-government requires self-governing individuals, and that religion and moral character have always, therefore, been legitimate objects of political concern. It is not illiberal and it should not be unconstitutional for the state to endorse, however mildly, those religions that nurture qualities of good citizenship.
Perhaps because O'Connor's opinion in the Pledge case was so obviously arbitrary and ad hoc, both Rehnquist and Thomas shunned it. The Chief Justice cited, instead, the "psychological coercion" doctrine. That standard, first articulated by Justice Kennedy in the graduation prayer case Lee v. Weisman (1992), holds that public schools cannot psychologically pressure school children to participate in religious exercises. Rehnquist determined that the phrase "under God" is in no sense a prayer. Iing it in the Pledge, he wrote, does not convert the recital into a religious exercise and, accordingly, does not violate the psychological coercion test.
Justice Thomas disagreed. He argued that a straightforward application of the psychological coercion test would demand that "under God" be found unconstitutional. After all, if, as Lee v. Weisman held, high school students cannot be exposed to a single end-of-the-year non-denominational prayer, then they certainly cannot be led in a daily recital that affirms God's existence. Thomas proposed jettisoning Kennedy's framework and replacing it with Justice Scalia's "legal coercion" standard, Ih would prohibit the state from coercing religious belief or practice by force of law or threat of penalty—e.g., by suspending from school students who fail to recite the words "under God." More radically, Thomas argued that the original intent of the Establishment Clause was to prevent Congress from meddling with religion in the states; and thus the federal courts had no business parsing California's law concerning the Pledge of Allegiance to begin with.
Although they reached the same conclusion, Rehnquist and Thomas set forth two fundamentally different visions for the future of jurisprudence on the Establishment Clause. Rehnquist would have the Court operate within existing precedents; Thomas would overturn them.
Newdow was not the first time that this difference surfaced. In Zelman v. Simmons-Harris (2002), the Chief Justice found constitutional a Cleveland program that allowed parents to use publicly funded tuition vouchers at private religious schools. Applying, in effect, the Court's religion decisions of the 1960s and '70s, he argued the program passed muster because it did not have the purpose or effect of advancing religion. Rehnquist employed O'Connor's endorsement test, too, contending that because the state did not discriminate on the basis of religion in distributing vouchers or dictating their use, a reasonable observer could not perceive state endorsement of religion over irreligion. Thomas voted with the Chief Justice in Zelman, but he directly questioned the propriety of evaluating state programs as Rehnquist did.
One could make reasonable arguments on behalf of either position. In one obvious respect, the Chief Justice's deference to precedent would seem to be more conservative, inasmuch as respecting prior decisions is a canon of traditional jurisprudence. Stare decisis generally promotes the predictable and consistent application of legal principles, encourages reliance on judicial decisions, and contributes to the actual and perceived integrity of the judicial process. If the Court does not respect its own precedents, it can hardly expect others to follow them. Not all precedents, alas, are equally worthy of respect. The problem of misguided precedents is particularly rampant in Establishment Clause jurisprudence, which seems to put a premium on unfair, unprincipled, and historically fictitious constructions of the text. Justice Thomas's call to overturn these bizarre decisions may seem "activist," but it is also, in a principled way, conservative. He would confess that the Court has erred and then return Establishment Clause jurisprudence to its constitutional foundations.
The argument for Rehnquist's approach is that it reflects the realities of the Court's current composition. For a decade, the Court has been mired in a 4-2-1-1-1 divide on establishment matters. Justices Stevens, Souter, Breyer, and Ginsburg embrace versions of the old "wall of separation" interpretation. They find suspect any form of government support for any kind of religion. Justices Scalia and Thomas favor a legal coercion standard. O'Connor uses her endorsement test; Kennedy, his psychological coercion framework. The Chief Justice's position fluctuates, but given that O'Connor and Kennedy are most likely to join the four strict-separationists, he tends to adopt one of their approaches in an attempt to assemble a five-member, moderately conservative majority. Such flexibility has allowed Rehnquist to deliver important victories, including one on school vouchers. Whatever the problems with O'Connor's and Kennedy's tests, the Chief Justice has demonstrated that he can use them to reach desirable results.
But this strength comes with a troubling weakness. O'Connor's endorsement test and Kennedy's psychological coercion standard have forced Rehnquist to employ less-than-compelling legal arguments, which often deny religion's legitimate place in public discourse. To uphold the Pledge, for example, the Chief Justice had to all-but-read religion out of it. His opinion implicitly conceded that in the public square religion (in a prayerful sense) is constitutionally suspect. He also adopted the false dichotomy that something that is patriotic cannot be religious. The Pledge is patriotic and religious; the categories are not mutually exclusive. Rehnquist's opinion, furthermore, implied that there exists a clear and significant distinction between religious affirmations, like "under God," and religious exercises, like a non-denominational graduation prayer. Even if such a distinction does exist—and if it does, defining it would be more appropriate for scholastic theologians than judges—it is by no means clear why the state should be allowed to lead students to recite "under God" if it cannot ask them to stand for a non-denominational prayer. In short, the Chief Justice's reasoning was problematic, at best.
The strongest argument supporting Thomas's call to overturn precedent may be that it would allow the Court to adopt a clear standard for what the Establishment Clause forbids. A jurisprudence that "respects" precedent by devising nuanced distinctions and exceptions in order to reach a desired result undermines the rule of law. Over time, it creates a body of decisions so complex and so multifaceted that, for all practical purposes, precedent ceases to exist and judges can use the past to construct whatever future they desire. For evidence of this, consider that Justice O'Connor's approach now permits legislation that, as Richard Ellis shows beyond any conceivable doubt, had the explicit purpose of publicly endorsing religion.
Much more so than Rehnquist, Justice Thomas is willing to clean up the Court's Establishment Clause mess. Although Thomas lacks the votes to elevate his criticism of the Court's church-state precedents into the Opinion of the Court, the Court's composition will likely change soon. Thomas then could be in a position to rewrite church-state jurisprudence by forging a lasting majority willing to make a clean break with Justice Kennedy's and Justice O'Connor's temporizing. Only then would the Establishment Clause start to look familiar to its authors and ratifiers.