I am concerned about the actual operation of the program, not about how many corporations have been formed.

There is some support in the cases for this notion that two corporations matter— but not much. Bradfield v. Roberts, the 1899 opinion upholding government money to a religious hospital, is written on the ground that the hospital is not the church, but merely a corporation controlled by the church. This has always struck me as classic nineteenth-century formalism, but at any rate, the opinion does not create a requirement of separate incorporation. It simply decides the case before it, in which separate incorporation was one of the facts.

In the cases on religious schools, the Court has created a category of institutions it calls "pervasively sectarian." Even at the height of restrictions on aid to religious schools, some forms of aid could go to pervasively sectarian institutions, but aid to those institutions was more tightly restricted than aid to other religious institutions that were not pervasively sectarian. This doctrine is said to support the requirement of two corporations; opponents of charitable choice presume that the church itself is pervasively sectarian, but that its affiliate may not be. The presumption is fallacious; a church might operationally separate its delivery of social services from its purely religious functions, whether or not it separately incorporates them, and the separately incorporated affiliate might combine its religious and secular work.

With respect to social services, the Court reserved the question of pervasively sectarian providers in Bowen v. Kendrick. See 487 U.S. at 611, following cases which it characterized as having "left open the consequences which would ensue if they allowed federal aid to go to institutions that were in fact pervasively sectarian.' More recently, four justices in a school case repudiated the whole concept of pervasively sectarian, correctly noting that the Court had steadily reduced its reliance on the concept, that the concept had originated as a code word for Catholic, and that it had grown directly out of virulent nineteenth-century anti-Catholicism. Mitchell v. Helms, 530 U.S. 793, 828-29 (2000) (plurality opinion). Two more Justices, concurring, did not join in the concept's overt repudiation, but neither did they rely on it. Id. at 836-67 (O'Connor, J., concurring). It seems quite unlikely that the distinction between pervasively sectarian institutions and other religious institutions will be revived and actually extended to control cases about social services. Charitable choice legislation should not codify this discredited concept.

Whether there is one corporation or two, the real question is whether the religious provider must secularize the part of its operation that delivers government-funded services. Certainly it must fund any religious elements itself; government can pay only for secular services. But must it abandon religious elements altogether? Charitable choice proposals say no, and that is the right answer.

To say that a religious provider must conceal or suppress its religious identity, refrain from religious speech, remove religious symbols from its work area, or hire people who are not committed to its mission, is an indirect way of saying that government can contract only with secular providers. Attaching such conditions to a government contract uses the government's power of the purse to coerce people to abandon religious practices. Such coercion is just as indefensible as if the government coerced people to participate in religious practices. Charitable choice provisions that protect the religious liberty of religious providers are pro-separation; they separate the religious choices and commitments of the American people from government influence.

The ultimate irony in this debate are the people who oppose charitable choice on the ground that if religious organizations take government money, they will eventually be regulated and secularized-and then also oppose charitable choice on the ground that it protects religious providers against secularizing regulation. They cannot have it both ways. The status quo, in which bureaucrats have discretion to contract with religious providers or boycott them, on whatever conditions the executive chooses to impose, is far more dangerous to religious organizations than a charitable choice bill with clear protections against discrimination and against secularization. C. Protecting Beneficiaries.

The third change in charitable choice is that it provides explicit protection for the religious liberty of the beneficiaries of government programs. They are entitled by statute to a secular provider on demand. If they choose to accept a religious provider, they may be exposed to religious exercises, but they cannot be required to actively participate.

These are important protections, and I would not support any bill that omitted them. They do not exist in present law. When a bureaucrat chooses to contract with Catholic Charities, no current law requires that he have a secular provider available for all those who request it. And any constitutional protections for program beneficiaries are, like the protections for providers, little known and undeveloped.

IV. IMPLEMENTATION.

Charitable choice is in principle a great improvement for religious liberty. But the difficulties of implementation are serious. Those difficulties are not new; they exist under the status quo, where they have received no serious attention from either side. These difficulties are more visible under charitable choice, because contracts with religious providers are more visible, and both sides have begun thinking about the difficulties. I doubt that either side has thought enough.

I am no expert on government grants and contracts or on the delivery of social services. I cannot offer full solutions to these problems, but I can flag some of the more obvious risks.

A. Ending Government Discrimination.

Charitable choice says government cannot discriminate in the award of grants and contracts. How do you enforce that? Legislatures have found it necessary to enact procurement laws with so many protections against corruption that the process of buying anything for the government has come to be a standard source of jokes. To the usual risks of government contracting, add the religious biases of the general public and of the officers awarding the grants and contracts. Some of them are deeply religious; some of them are strongly secularist; nearly all of them like some religions more than others, and have some religions they really mistrust. Choosing someone to deliver social services is more complex than picking the low bidder on a pencil contract. How do you keep thousands of government employees, federal, state, and local, from discriminating on religious grounds when they award grants and contracts?

I don't know the answer to that question. We are learning that just telling them not to discriminate doesn't work. It appears that open and obvious religious discrimination continued under the limited charitable choice provisions enacted in 1996. Amy Sherman's study, reported at a House hearing in April, found that some states are contracting frequently with religious providers, and that others are not doing so at all.

I don't know how you police bureaucrats, but I think you have to assume that many of them will continue to engage in religious discrimination despite the enactment of charitable choice. Some will refuse to deal with religious providers; some will refuse to deal with non-Christian religions, or non-Western religions; some will prefer religious providers and discriminate against secular providers. You at least need a reporting requirement, so that implementation can be monitored, and you may need to require explanations of any obvious over-or-under representation of religious providers. As we have learned from the civil rights experience, resolving claims of subtle discrimination is a difficult task.

Decentralization reduces the risk of discrimination. For those services that can feasibly be delivered through vouchers, vouchers privatize the choice of providers and thus deprive government employees of the opportunity to discriminate. Decentralized contract awards, with many government employees choosing providers, spreads the risk of discrimination better than centralized contract awards with one or a few employees choosing providers.

B. Deregulating Providers.

Charitable choice proposals have made the most conceptual progress with respect to deregulating providers. Existing legislation and other pending proposals have clear and specific provisions to protect the religious liberty of providers who accept government grants or contracts.

These protections have to be in the statute, because no one can count on the courts to provide them constitutionally. The federal courts systematically underprotect the free exercise of religion, and the Supreme Court believes that when the government awards a contract, it can define the job very precisely and attach all sorts of conditions to ensure that the contractor adheres to the job specifications. Rust v. Sullivan, 500 U.S. 173 (1991). When Congress means to deregulate, it has to say

SO.

It would be better to vote down charitable choice than to remove the deregulation of religious providers. From a religious liberty perspective, the worst outcome would be to codify a rule that government offers money to religious providers but only on condition that they agree to secularize themselves. An unambiguous and highly visible offer of government payments to change one's religious practice would be worse than the muddled, regulated, and discriminatory status quo.

These protections will be somewhat easier to enforce than the basic rule of no discrimination in the award of contracts, because victims of violations will know immediately when government asks them to change their hiring rules or downplay their religious message. Still, you have to assume that there will be political and bureau

cratic resistance to the deregulation of religious providers, and that continued vigilance will be necessary to make it work.

C. Protecting Beneficiaries.

Most charitable choice proposals provide equally clear protections for program beneficiaries. Beneficiaries should be entitled to a secular provider on demand, to decline to actively participate in religious exercises, and to clear notice of these rights. But these rights may be very difficult to implement.

Social service programs have never been funded sufficiently to meet the need, and recent legislation ensures that these programs will be even more severely starved for funds in the future. We have not succeeded in guaranteeing even one provider for all the people who need these services. How can we plausibly guarantee a choice of providers?

The problem is hard enough in big cities; it is far worse in small towns and rural areas. It is hard to envision religious and secular providers operating side by side with government funds in New York City. It is impossible to imagine in Waxahachie, Texas. Nor do I think it is just a matter of sending one or a few dissenters to a private practitioner. Private practitioners tend not to locate in low-income areas, and anyway, there may be many beneficiaries who don't want a religious provider. The beneficiaries are vulnerable and dependent and may be afraid to assert their rights, but government and government-funded providers should not take advantage of that. The goal should be to give each beneficiary his free choice of a religious or secular provider, and at the very least, not to push a religious provider on anyone. I suspect that is a much bigger challenge than the sponsors of charitable choice have talked about in public.

Again, these problems are probably no worse than under the status quo; they are just more visible. When government contracts with religious providers today, I am not aware that it makes any effort to provide secular alternatives. Once gain, charitable choice is an improvement in concept. But implementation is likely to be difficult.

D. Program Efficacy.

A frequent policy question about charitable choice is whether religious providers will help more beneficiaries than secular providers. I don't know; social services are not my field. But my work on religious liberty and the associated experience of religious diversity makes me nearly certain that that is the wrong question.

The right question is whether religious providers will help different beneficiaries than secular providers. If some people in need respond to religious messages but not secular ones, and other people in need respond to secular messages but not religious ones, then the only way to help both groups is to make available both religious and secular providers.

Whether there are significant numbers of people in both groups is an empirical question, but the answer will surely be yes. There are many Americans for whom God is the only source of ultimate meaning and for whom religious messages are more motivating than any secular message ever could be. There are many others for whom stories of God are a giant fraud or a giant game of pretend. And there are yet many others in between, whose views of God are not strong enough to motivate either reform or resistance. Given the enormous diversity of religious views in the country, it seems almost inevitable that there will be a similar diversity of responses to religious and secular providers of social services, and that each type of provider may reach some beneficiaries that the other type of provider could not.

In any event, the question to ask is not whether religious providers will help more people than secular providers, or vice versa. The question to ask is whether offering people a choice of religious or secular providers will help more people than exclusive reliance on one or the other.

V. CONCLUSION.

Religion should not be forced on any American, but neither should any American be excluded from the operation of social welfare programs because of his religion, or lack thereof. The Religion Clauses are designed to let people of fundamentally different views about religion live together in peace, in mutual liberty, and in equality. Religious choices and commitments are left to the private sector, and to that end, government should neither prefer the religious nor prefer the secular. In its own operations, it must necessarily be secular. But when it chooses to contract out to the private sector, it should contract without regard to religion. This principle minimizes government influence on religion and thus maximizes religious liberty, and this is the true meaning and purpose of separation of church and state.

Minimizing government influence is easier said than done. Charitable choice is admirable in its commitments to nondiscrimination on the basis of religion, to deregulating religious providers, and to protecting program beneficiaries. But each of these commitments will be difficult to implement; each of them requires careful attention from the Congress and from those expert in the delivery of social services. Mr. CHABOT. Professor Saperstein.

STATEMENT OF DAVID N. SAPERSTEIN, ADJUNCT PROFESSOR OF LAW; DIRECTOR, RELIGIOUS ACTION, CENTER OF REFORM JUDAISM, GEORGETOWN UNIVERSITY LAW CENTER Mr. SAPERSTEIN. Good morning, Mr. Chairman, and distinguished Members of the Subcommittee. I am here today to urge that you reject charitable choice. Charitable choice is bad for religion. It is bad public policy. It is unconstitutional and it is socially divisive. And all this for a program that may not result in one more needy person being helped.

Let me address some of the policy issues first. With government money comes government rules, regulations, audits, monitoring, interference and control. It will result in a limitation of religious autonomy and freedom.

Second, with government money comes compromises in the religious mission of the churches, synagogues, mosques of America.

Third, by opening up our Nation's limited funding for social services to, potentially, scores of thousands of houses of worship, millions of dollars will be diverted from and thus weaken what is widely regarded as the finest, most effective social service providers today: the superb, albeit overwhelmed, religiously affiliated social service providers such as Catholic Charities, Jewish Federation, Lutheran Social Services, all of which abide by the vast majority of regulations applicable to all charities.

Fourth, charitable choice will lead to increased religious competition and divisiveness in America. Choosing between professional social service agencies is one thing. Choosing between focal houses of worship comes much closer to choosing between religion. You are going to find the Episcopal Church, the AME Zion Church, and the local mosque competing against each other for grants, and they are going to come to you here on Capitol Hill for assistance. One will receive it in the end, and one won't; and they will want to know from you why. And, of course, politics will determine who gets these grants, and that means it will be the smaller minority religious groups that are likely to be left out.

Indeed these very political realities already result in many minority religious groups facing particular problems, for example, in finding locations to build houses of worship. In the recent Pew poll, it shows that substantial majorities of Americans feel that, for example, Buddhist and Muslim social service providers should not receive government Federal funds. This is not helpful to either religion or poor people in America.

And fifth, such funding violates the religious rights of taxpayers. As Jefferson said, to compel a man to furnish contributions of money for the propagation of opinions which he disbelieves is sinful and tyrannical. And this helps explain why so many religious leaders on the left and the right oppose the program.

Let me then address the legal and constitutional issues. First, in all of the discussions of all the cases that you will be reading about

and hearing about, there is one central principle, one legal standard you must keep in mind. The Supreme Court of the United States, and the vast majority of lower courts as well, has never upheld direct government cash support for pervasively sectarian institutions. Indeed, in cases where it has given indirect support, it has distinguished it; either direct versus indirect, cash versus inkind, or pervasively sectarian institution against religiously affiliated or other religious organization.

Secondly, the rights of beneficiaries would inevitably be infringed. As Professor Laycock and others have noted, in the real world, protecting beneficiaries will be difficult and, I might add, all but impossible.

Third, churches and synagogues have rightly been exempted from many laws that would compromise their religious freedom, including the right to discriminate whom they hire on religious grounds. Major government funding for programs with such exemptions is a debatable constitutional point, but if constitutional, such federally funded programs will be part of a campaign that will weaken civil rights in America and give government sanction for unintentionally dividing America along religious lines.

So in deciding on charitable choice, you on Capitol Hill are faced with a wrenching tension between two valid moral principles. The first is that government should accommodate the ability of religious organizations to function, and it is only that exemption that is given in the Amos decision-as the Amos decision described itonly to further the religious function of the entity. And to take away that exemption is to curtail that religious freedom in a manner that will threaten other exemptions.

On the other hand, we have an equally valid principle that with government money we shouldn't be discriminating against people. The notion that under these programs, government-funded social service programs run by a Protestant church might run ads that read, "Jews, Catholics, Muslims need not apply", "no unmarried mothers will be hired;" and, unless I read Watts-Hall incorrectly, even that "no blacks need apply," should be deeply troubling to all. Mr. Chairman, there is much in the President's program on which we can work together with everyone across the board on ways to stimulate through the tax system more money to religious organizations, as well as others providing aid to the poor; much in terms of technical assistance. There is much that we can do. What we shouldn't do is tear this country apart over this fight about whether to directly fund religious organizations. We have programs in effect. Let's stop, take a look at how they work in real life, and then come back and address this very problematic issue. Mr. CHABOT. Thank you, Professor.

[The prepared statement of Rabbi Saperstein follows:]

PREPARED STATEMENT OF RABBI DAVID SAPERSTEIN

Good morning Mr. Chairman, distinguished members of the Committee. I am Rabbi David Saperstein, Director of the Religious Action Center of Reform Judaism which represents over 1,700 rabbis and 900 synagogues with 1.5 million members. I am also an attorney and for many years have taught church-state law on the faculty of Georgetown University Law School.

I am honored to share this hearing today with three such distinguished constitutional law scholars. Each one is a friend and dedicated champion of religious free