Sandra Day O'Connor [automatically transcribed, may contain inaccuracies]
Thank you so much. Thank you,

Our former national Secretary of Education, he did a great job in that position for the nation. And I really enjoy being in your state and in your beautiful city of Charleston. And being part of a program put on by this relatively new law school, you're doing a super job. And I think this seminar that you put on, is going to register throughout the country, be sure you publish the talks that were given in your law review, or some similar publication, because I think the remarks have been very informative, very helpful, very thorough. So don't let them get away, make sure they're available for circulation nationwide. And I'm so glad to be here. Now, I'm going to bore you with a little more about the endorsement principal, if you can stand it, you think you can for a few minutes.

I'm so glad to be part of the program today. The establishment clause has been very much of interest to me, particularly when I served on the Supreme Court. And I am really pleased that the law review here and the law school has chosen to provide such a marvelous program in it on it today. It's great. And I think the interest in the Establishment Clause is well deserved. religious pluralism lies at the very heart of the American political tradition. And I think it remains a major concern, as our country becomes ever more of our own, have larger and larger communities of people from widely different ethnic and religious backgrounds around the world. I mean, our population today is vastly more diverse than it was 25 years ago, or 50 years ago, it's remarkable.

Less than one third of eighth graders today in the United States can identify the historical purpose of the Declaration of Independence. And it's right there in the title. So I think we're going to have to do, we're going to have to do some reading work first. But I think most Americans do know about our wall, our wall of separation between church and state. And maybe they believe, not correctly, but they believe that those words of separation appear in the Constitution, or maybe they just think that's what the Declaration of Independence was about. I'm not sure. But in any event, most of our citizens know that that concept is somewhere there. And that we've made a commitment, promoting religious freedom and religious pluralism in our country, and have made a commitment to ensure that we have a government that doesn't show favoritism to one particular religious belief or a lack of a religious belief. And how that commitment actually appears in the Constitution.

And the religion clause, as we know, is deceptively simple. "Congress shall make no law respecting an establishment of religion, or prohibiting the free exercise thereof." Those 16 words, that's all that's there. Those 16 words have bedeviled legislators, courts, commentators for more than 220 years. And I'm going to talk today by the elite about at least some of the bedeviling cases that we've had. But I'm not going to comment directly on the present and recent work of the Supreme Court. Now, this always disappoints my audiences. But it seems to me that being a retired justice is a position that requires me, in my opinion, to leave it to the present members of the Supreme Court to figure out the hard cases, and to keep my nose out of it.

So I'm sure it would not surprise you to learn that there are many cases that I would celebrate, and many opinions that I would not endorse, if it were still my job, but it isn't. So I take seriously my commitment to avoid becoming involved by commenting too closely on the merits of questions that may be before the Court or likely to be there. And this is a long-winded way of saying don't get your hopes up for a fiery critique of the last decade of Establishment Clause cases, or predictions, armchair predictions for cases that are coming up. That's your job, not mine.

Now, it's easy to take aim at Establishment Clause jurisprudence, and in general, in some cases, in particular, it's almost a blue good story, for example, in talking about the religion clauses of the Constitution, to mention some of the odd and seemingly inexplicable distinctions drawn by the Court's cases in this area. A prayer offered in an inclusive manner, by a member of the private community at a school graduation ceremony is found to be unconstitutional. But the constitution permits state-employed chaplains to open legislative sessions with a prayer of the same kind. Now what's going on here?

A display on public property of a Christmas tree and a Hanukkah menorah is permissible. But a cross standing alone violates the Constitution. What can I say, states and localities may erect a monument displaying the 10 commandments, except sometimes they can't. Those whose religious observance requires prohibited substances like peyote must comply with generally applicable laws. But the Amish people must be exempted from generally applicable laws regarding compulsory education, because of their religious beliefs. A cross remains standing, because no one has standing to challenge the cross. Though the entire point of the challenge is the message that grace conveys to the blanket.

Now you explain that to me, though moments of silence and invocations of divine providence are both ubiquitous practice in public settings, a law providing schoolchildren a moment of silence for personal medication and prayer goes too far. But now, what is the origin of fine distinctions like this? Do they make any sense? The Court itself has admitted that its establishment clause jurisprudence is marked by considerable internal inconsistency. Well, I'd echo that it certainly is. Justice Scalia once characterized it as embarrassing. And one of his most famous pieces of writing, a colorful running metaphor, comparing the lemon test to a zombie from a horror film, the mound the strange establishment clause, a geometry of crooked lines and wavering shapes that the doctrine has created. Of course, that's actually pretty mild criticism for Justice Scalia. But at some level, I think he's right.

No one would look at the last 40 years of Establishment Clause jurisprudence and think the Court had created a neatly ordered set of rules and precedents to live by. So my intention this afternoon is not to defend the consistency of the Court's jurisprudence, there isn't any, but rather to offer some thoughts on why I think this area of the law is so difficult. Indeed, because I prefer to avoid criticizing or praising the cases on the merits, what I want to talk about is a way of deciding these kinds of difficult cases, and maybe cases more generally. This relates to your topic for this symposium, and the quest for a grand unifying theory of our Establishment Clause jurisprudence. And I don't mean to be a downer, but I'm not sure this search is going to be fruitful, at least on its own terms. I'm going to talk about the particular different faculties of the Establishment Clause in a moment. But first, let me say a few words about the fundamental approach.

It is, of course, important to aspire to clear rules of decision making, and overarching grand theories that organize those rules into a consistent and principal jurisprudence. It's always better when it's possible, to render the law into a form that's easy to apply to the next case, at all levels of the judiciary. My former colleague Justice Scalia is fond of saying, "The rule of law is a law of rules." But my much older colleague with whom I did not actually serve, Justice Oliver Wendell Holmes, taught that, "The life of the law has not been logic, it's been experience. And it cannot be dealt with as if it contained only the axioms and corollaries of a book on mathematics." Perhaps then from Justice Holmes' perspective, a strange geometry of crooked lines and wavering shapes is not the worst thing in the world. It is to be expected from a field that learns by experience in history, rather than from logic and rules.

Now, the law requires that experience in history through the careful plotting process of case by case adjudication, which is what we have in this country. Of course, we try to decide those cases by using somewhat discrete tests and standards. And as someone who proposed an Establishment Clause test that is now somehow occupying all of you on a nice spring day, I can hardly claim to be opposed to searching for a broader doctrinal solution to difficult legal problems and goals. But that tests needs to be flexible, so that it can change and reflect the experience that the Supreme Court gets from each new case. Sometimes a test can teach you what you need to know about a case. Sometimes that case teaches you something you need to know about your test. The eminent political philosopher John Rawls called the process of testing our abstract moral and ethical ideals against our intuitions, reflective equilibrium. Now there is a term for you: you want to be an intellectual, talk about reflective equilibrium. Now, maybe you will forgive a plain-talking cowgirl, if she just calls it, learning from experience.

Now, I hope to explain this process of articulate as a general principle or test, applying it to the cases and allowing it to evolve with experience. I think it's especially suited to the religion classes. But be aware that it will never give you a grand unifying theory matches we would like instead for to get is a guiding idea, like a star up there on the horizon, and a set of doctrinal tools that can help you follow it. I continue to believe that the non-endorsement principle I wrote about in my concurring opinion in Lynch, Lynch v. Donnelly, remains the best guiding principle that we have found for our Establishment Clause jurisprudence. It points the way to the right answers, but it does not find them. And they're certainly not always clear.

So you take it one case at a time, let experience be the life of the law, and either refine or supplement that test you have so that we can talk about what makes the establishment clause in particular, and the religion classes in general, so difficult to pin down. Now, commentators like those we heard today have had a field day in their criticism of the Court's religion clause jurisprudence. I mean, look at all the mess we've made. In my view, though, the Court's jurisprudence is confusing, not because the Court is confused, that may be part of it, but rather because there are inherent tensions in the religion clauses of our constitution that undermine attempts to achieve strict logical consistency in all these areas.

The first difficulty arises, of course, from the tension between the free exercise and establishment clauses themselves. As Chief Justice Burger noted for the Court in the case of Walz v. Tax Commission, either of their two religion clauses expanded to a logical extreme would tend to clash with each other. Our cases have interpreted the Free Exercise Clause to exempt some people from some generally applicable government requirements to permit those people to freely exercise their religion. Meanwhile, another case, which I should say I didn't join, called Smith v. Employment Division suggests that generally applicable laws do not generally require religious exemptions.

Yet even where the first Exercise Clause does not compel a government to grant an exemption, the Supreme Court has suggested that the government may, in some circumstances may voluntarily choose to exempt religious observers without violating the establishment clause. Yet if the government is required or voluntarily, voluntarily chooses to accommodate religion, under their Free Exercise Clause, its purpose in doing it. And the effect of that action is by definition religious. So the government conduct required are permitted under the Free Exercise Clause might well violate us or at least implicate the Establishment Clause because of its concededly religious purpose and effect. This is obviously a pretty careful balancing act, isn't it?

Moreover, the establishment cause itself contains competing values. Even if we all agree the Establishment Clause protects religious liberty, the application of that general principle to concrete cases, has failed to heal consensus on the court. At what point for instance, does prohibition of aid to religion become affirmative hostility toward religion? Although it's much maligned, the Court continues to apply that familiar three-prong lemon test you've talked about today, it requires determining whether the conduct that's challenged has a religious purpose, whether it has a primary effect that advances or inhibits religion, and whether it fosters excessive government entanglement with religion. But the second and third prongs and fat test can be hard to apply to a genuinely well-intentioned school district that wants to open its facilities to religious community groups, promoting their free exercise, but only in such a way as not to impinge too far on the right of other members of the public to be free from government-sponsored religion, which is a legitimate Establishment Clause concern. The religion clause is then have challenged the court to delineate, try to delineate anyway in a principled fashion, the contours of the channel, between the Scylla of what the free exercise Exercise Clause demands and the Charybdis of what the Establishment Clause prohibits, through which any state or federal action has to pass in order to survive constitutional scrutiny.

See, all I'm saving you from now, conflict in this area is unavoidable. Given the obstacles, there is no obvious or really clear roadmap to give you that will be a died for a coherent religious clause jurisprudence. You can't produce it from logic, or from the history of the clause or from the text. Perhaps, then, we might excuse the justices from occasionally making a wrong turn, or getting a bit lost. At the very least, the justices need a guiding star to point the way between dangers that lurk on either side. It's funny, perhaps that metaphor should come So naturally, really unintentionally, to a case about a nativity scene. That alone suggests, I think, how challenging these cases can be in parsing out the interest or intent to endorse religion, from the mere fact of religious ubiquity in our lives in human history.

In any event, in 1984, in a case involving almost a seasonal ritual at the Court, the Court considered an establishment clause challenge to a protected Rhode Island cross that was part of the city's annual Christmas display. That case was Lynch v. Donnelly. And the Court held that notwithstanding the religious significance of the cross, the city had not violated the establishment clause. Now, I joined the majority opinion, which applied the lemon test to the facts of the case. But I also wrote separately to suggest what I consider to be a useful and theoretically sound compass for deciding Establishment Clause cases. I began with the premise that the Establishment Clause prohibits government from making adherence to religion relevant in any way to a person standing in the political community. I then reasoned that government, the government could run afoul of that prohibition in two ways, either by excessive entanglement with the religious institutions, or by government endorsement or disapproval of religion.

The crucial point is that endorsement sends a message to non-adherence that they are outsiders, not full member of the political community. And then an accompanying message to adherents that they are insiders, favorite members of the political community disapproval since the opposite message. That's under this suggested endorsement test. The central issue in the Lynch case was whether this city had endorsed Christianity in its holiday display. I concluded that this city had not intended to convey any message of endorsement of Christianity, or disapproval of non-Christian religions. Because the evident purpose of including the cross in the larger display was celebration of the public holiday through its traditional symbols.

I also concluded that the city's display of the cross did not convey some message of endorsement of Christian beliefs represented by the cross. Because the overall holiday setting of the display negated any message of endorsement. And it was quite a holiday setting. It included a Santa Claus house, reindeer, candy-striped pose, a Christmas tree, Carol or a clown, an elephant, a teddy bear, colored lights and "Seasons Greetings" banner in context and viewed from the perspective of endorsement. I did not see in this wide ranging holiday display and intent or effective message of government endorsement of Christianity. It was a colorful and pluralistic way to mark an important part of secular American community and its holidays.

But consider another close case that came the very next year. In 1985, the court in Wallace v. Jaffree held that Alabama's moment of silence statute violated the Establishment Clause because it was entirely motivated by a purpose to advance religion, and it had no secular purpose. In that opinion, I concurred separately, reiterating my view that the endorsement test gives analytic content to the lemon inquiry into legislative purpose and effect. Accordingly, the crucial question in the case was whether this state had conveyed or attempted to convey the message that children should use the moment of silence for prayer. Specifically, the relevant issue was, an objective observer acquainted with the text's legislative history and implementation of this statute would see it as a state endorsement of prayer in public schools.

Reviewing the evidence in the record about the statute and its intended purpose, I concluded the Alabama statute was invalid on the ground that it was intended to convey a message of state encouragement and endorsement of religion. And it was once again, the context that determined a close case. Alabama already had a law, creating moments of silence. They simply changed the law to expressly permit prayers that were already permitted in practice, broadcasting what I thought was a message that those who chose to use the moment of silence in a particular way were particularly favored.

In my view, the end endorsement test captures the essential command of the establishment clause, namely, that government must not make a person's religious beliefs relevant to his standing in the political community. By conveying a message that religion, or a particular religion, religious belief is favored or prefer preferred. We do live in a pluralistic society, and our citizens do come from diverse religious traditions, or adhere to no particular religious beliefs at all. Religious Liberty, the undisputed central goal of the religion clauses must not be conditioned on forfeiting the individuals status of equal citizenship, if government is going to be neutral in matters of religion, rather than showing either favoritism or disapproval towards citizens, based on their personal religious choices. government cannot endorse the religious practices and beliefs of some citizens without sending a message to non-adherence that they're outsiders, or less than full members of the political community.

I'm saving you. I still believe despite all this rhetoric, that the endorsement tests which focuses on whether government conduct sends a message of separation or inferiority, ask the right question about government practices that are challenged on Establishment Clause grounds. I'm also convinced that the endorsement test is capable of consistent application, at least in the sense that it should always be an object of primary concern. When we evaluate these cases, there is no occasion on which it's appropriate for government to tailor its actions in order for it to express the endorsement of a particular religion, or even a religion in general, as opposed to the celebration of pluralism, free exercise or non hostility.

I still think the non-endorsement principle is a good idea, maybe the best we have in this very difficult area of the law. But is that a grand unified theory of the kind you are searching for? And this seminar? I think the answer is probably not. The endorsement test depends on a sensitivity to the unique circumstances and the context, particular practice that's being challenged. And like any tests that sensitive to context, it will not always yield results, with a unanimous agreement of the decision makers, or FA fits at the margins of the question. Of course, that's true of many standards. In constitutional law, the court can't abide avoid the obligation to draw lines, it has to do that all the time. And they're often very close and very difficult lines. And they have to do that and deciding Establishment Clause cases. And that problem is not unique to the endorsement test.

But if we're going to take a fair account of the endorsement tasks in practice, I think we need to identify some of the hard questions that can't be answered by references to the non endorsement principle alone. And here's some of the hard questions. Do we care about the message that the government it tends to send? The message that gets received? Or what? What if they're not the same message? What kind of listener do we have in mind? And should an ardent atheist the thought to receive a different message from the median citizen? And isn't the ardent? Is it the ardent atheist that we're worried about? Or is it the ordinary citizen? And why does the endorsement test permit evaluation of the government message in its whole context? Instead of trying to get the marginal but avoidable increases in religious messaging? I mean, you could have a holiday display without a cross, or a legislative invocation conducted by someone other than ....

Why doesn't the endorsement test insist on government making those small changes? I'm not sure. Maybe the most challenging are the difficult historical instances of religious mention, in which experience seems to run headlong into the logic of non-endorsement? How only endorsement principle tell us when stamping the words In God We Trust on public institutions, when it conveys a message of religious endorsement? When surely the answer is that at least to some degree, it always does.

Now, when I have a grand unified theory myself on my religion clauses, and maybe about constitutional adjudication, in general, it's a methodological one. It tells me, go slow. Take one step at a time, learning where to place the next foot from the footfalls that have come before, take account of the fundamental values, like non-endorsement, but make our logical tests flexible enough to incorporate the lessons of history and of experience. Let's apply judgment, not rigid legalism to the particular facts of the cases as they come up. Let's not be afraid to rethink what we have thought, or even what we have said.

There is virtue in figuring out these kinds of problems one case at a time. Every case has at its heart, a pair of litigants to pro and con. And they deserve to have their case decided by attention to their particular facts and arguments, rather than purely as a vehicle for fighting over Berlin legal principles.

And those litigants in the case and their particular facts and cases are the teachers who helped us know how and whether the document, that doctrine we're following is working, and why there needs some changes. And there have been questions about whether and how far a non-endorsement principal will survive. Now that I'm not on board anymore, I think the endorsement concept will survive. And I think that it probably is alive. And well, every major Establishment Clause case that's come up since I retired from the court has been resolved primarily on standing or some threshold issue that represents that represents something of the caution that justices think is appropriate in this area of the law, and expresses something of the conflicts among the justices that are faced with really difficult questions, their plurality opinions that have reached the merits, since I left have all attended to the message that the government is sending.

That was particularly true of Justice Kennedy's decision regarding the five-foot cross in the Mojave Desert in the courts most recent case, and I'm not sure whether I would agree or disagree with that analysis on the merits. I haven't decided. But I do think I see the non-endorsement principle they're serving as the guiding star right now. And more importantly, the Court's narrowest opinion in that case, was only a standing question, asking whoever the parties had standing. And that shows the careful and considered face that the court is putting forward now in this very difficult area.

But whatever the court decides to do, in this area of the law, whether the tests I've described will survive or not, it's going to remain one of the most difficult areas of the law, for grand theories are always going to be eroded by the tests of time and experience. I think that's okay. The role of the judge is to do the best she can do. Do the right thing in the case that's before us. And try to explain what we're doing in a way that teaches something about the case that's going to come next. I think that on the merits, the non-endorsement principle, should remain that compass that leads the doctrine onward, through the inevitable thickets that are out there ahead of the court.

But I think even more so that the court does very well to consider each one of these cases, one step at a time, adjusting its theory, perhaps, and its course in very small measures at every step along the way. And thank you so much for tackling some of these issues in this remarkable series of presentations that you've had the last couple of days. It's really been good, high class, very worthwhile, and very impressive for your law schools. So congratulations to all of you, and I'm awfully glad I was here. Thank you .