OPINION
COFFIN, Circuit Judge.
Rhode Island’s Salary Supplement Act, R.I. Public Laws c. 246 (1969) provides for the payment of state funds to teachers of secular subjects in non-public elementary schools. At present the sole beneficiaries of this statute are teachers in Roman Catholic schools. This suit seeks a declaration that the Act violates the Establishment and Free Exercise Clauses of the First Amendment and an injunction against its continued operation.
Plaintiffs are citizens and taxpayers of the state of Rhode Island. The named defendants include Rhode Island’s Commissioner of Education, the state Treasurer, and the state Controller, all of whom are involved in the administration of the Act. In addition, a couple with children in parochial schools and several teachers eligible for aid under the statute have been permitted to intervene under Rule 24(b), Fed.R.Civ.P. Jurisdiction is based on 28 U.S.C. § 1343(3). Since the complaint seeks to enjoin a state law as repugnant to the Constitution, this three-judge court has been convened pursuant to 28 U.S.C. §§ 2281, 2284.
The complaint alleges in substance that Catholic schools are the primary beneficiaries of the Act, that the goal of these schools is the propagation of the Catholic faith, and that the statute therefore has as its purpose and primary effect the advancement of religion. Plaintiffs also claim that the statute constitutes compulsory taxation in aid of religion and thus violates the Free Exercise Clause. In their answers, defendants and defendant-intervenors have stressed that the Act aids teachers, not schools, and have denied plaintiffs’ conclusion concerning the purpose and effect of the Act. Defendant-intervenors have also alleged that the injunctive relief sought by plaintiffs would infringe
the free exercise of religion of those whose children attend parochial schools.
The case is now before us for final decision on the merits of plaintiffs’ and intervenors’ constitutional claims. We have held a hearing, at which we received depositions and documentary evidence, and heard testimony from the Superintendent of Schools for the Archdiocese of Providence, an Associate Commissioner of Education for the state of Rhode Island, and several teachers eligible for aid under the statute. On the basis of this record, we make the following findings of fact and reach the following conclusions of law in compliance with Rule 52(a), Fed.R.Civ.P.
I.
Findings of Fact
A.
The Statutory Scheme
The Salary Supplement Act opens with a statement that its general purpose is to implement the established state policy of providing “a quality education for all Rhode Island youth”.
This declaration is followed by specific legislative findings : that non-public elementary schools enroll 45,000 students, or 25 per cent of Rhode Island’s elementary school children; that because of the numbers enrolled, these schools are vital to Rhode Island’s educational effort; and that the rising cost of teachers’ salaries makes it increasingly difficult for these schools to maintain their traditional quality. The specific purpose of the Act is, in the language of the statute, “to assist nonpublic schools to provide salary scales which will enable them to retain and obtain teaching personnel who meet recognized standards of quality.” To accomplish this objective, the legislature appropriated $375,000 to pay up to 15 per cent of the salaries of teachers of secular subjects in non-public elementary schools. The Act defines “non-public” school to mean a non-profit school whose average per pupil expenditure on secular education does not exceed the average for the state’s public schools during a specified period. To qualify for aid, a teacher must have a teaching certificate, must teach a course similar to those offered in Rhode Island’s public schools using textbooks approved for public school use,
and must agree not to teach a class in religion.
Regulations implementing the Act call for semi-annual payments to eligible teachers in February and June.
Applicants must sign a “Statement of Eligibility” promising not to teach a course in religion and to use only teaching materials employed in the state’s public schools. The Commissioner of Education also requires non-public schools to submit data concerning enrollments and total expenditures. If this data indicates a per-pupil expenditure in excess of the statutory norm, an agent of the Commissioner must examine the books of the school in question in order to determine how much of its spending was attributable to secular education and how much to religious.
To date, this final breakdown into secular and religious expenses has proved necessary in only one case. Only two non-public schools have reported expenditures in excess of the statutory norm, and teachers at one of these schools failed to apply for aid, thus obviating the need for further inquiry.
Approximately 250 teachers have applied for aid. It is undisputed that all of these applicants are employed by Roman Catholic schools.
B.
Nature of the Crisis Leading to the Statute
The evidence introduced at trial fully corroborates the legislature’s finding of a financial crisis in non-public education, but indicates that the crisis is largely confined to Rhode Island’s Catholic schools. Approximately 95 per cent of the elementary school children attending non-public schools are enrolled in Roman Catholic elementary schools. The financial crisis in these schools stems from the rapidly changing composition of their faculties. As recently as ten years ago, the Archdiocese of Providence relied almost exclusively on nuns to staff its school system. Lay teachers filled only 4 or 5 per cent of the system’s 1200 teaching positions. By 1969, lay teachers constituted one third of the teaching force. In the academic year 1970-1971, the great majority of parishes will require at least one additional lay teacher, with some needing two or three more. This precipitous increase in lay personnel — an increase which is expected to continue — has placed a double burden on the parish budgets which finance almost all of the 98 elementary schools. Each shift from a teaching sister to a lay teacher represents a threefold increase in salary expense (i. e., a shift from approximately $1800 to $5500 at present levels). Moreover, the increasing salary levels in public schools make the task of recruiting lay teachers annually more expensive.
The Salary Supplement Act will not relieve the parishes or parents of their escalating burden, but will temporarily enable parochial schools to compete for qualified teachers. A comparison of past and predicted salary levels confirms the limited impact of the statute. In 1968-1969, a starting lay salary in the parochial schools was $5000 a year. In 1969-1970 the diocesan school system offered $6000, hoping that 15 per cent of this amount, or $900, would be paid by the state under the Supplement Act. In the meantime, however, the standard beginning salary for public elementary school teachers in Providence and elsewhere has increased from $6000 to $7000. Even assuming, as does the Superintendent of the diocesan school system, that the parochial schools can compete effectively for teachers if they come within $500 of public school salaries, the diocese
must offer a starting salary of $6500 in 1970, of which $975 would be paid by the state and $5525 would be paid by the parish — an increase of ten per cent over the $5000 parish expenditure pre-1969.
The financial implications for both the parishes and the state are clear. Even to continue to subsidize the salaries of the present teacher-recipients at the present percentage figure will cost both the state and the parishes substantially more during the coming school years. But already the Superintendent of the diocesan school system has informed elementary school principals that he expects the present ratio of two sisters to one lay teacher to give way to a one to one ratio in the near future. Moreover, the Superintendent acknowledged at trial that almost all of the teaching staff might be lay persons within five years. Since, as we have seen, the substitution of a lay for a religious teacher means, even at present rates, a threefold increase in salary cost, it is obvious not only that the parochial schools face a monumental and deepening financial crisis but also that the objective and rationale of the Salary Supplement Act— to enable non-public schools to compete effectively for qualified teachers — will inescapably require substantially greater appropriations subsidizing substantially greater percentages of the salaries of lay teachers.
C.
The Parochial School System
At trial, the parties went beyond the immediate financial crisis in parochial schools to explore the operations of the diocesan school system as a whole. Plaintiffs’ evidence was designed to show the close financial and pedagogical nexus between Church and school. Intervenors’ testimony explored the teaching of secular subjects in these schools. From this evidence, the following picture emerges.
Rhode Island’s Catholic elementary schools are under the general supervision of the Bishop of Providence, and of his appointed representative, the Diocesan Superintendent of Schools. The financing of these schools varies: in some, substantial tuition is charged covering up to 40 per cent of school costs; others charge only a nominal fee and rely on general parish revenues to make up the balance. In most cases, however, the parish assumes the ultimate financial responsibility for the school, and the pastor assumes the ultimate responsibility for the allocation of parish funds. School principals are, with two exceptions, nuns appointed either by the Superintendent or by the Mother Provincial of the order staffing the school. Lay teachers are first interviewed by the Superintendent’s office, then by the school principal, with the contract being signed by the pastor of the parish. The Diocese sets minimum salary standards for both lay teachers and nuns, but some discretion is left to the individual pastor in negotiating actual salary levels.
The ground rules for school principals are set forth in a “Handbook of School Regulations”, having “the force of synodal law in the diocese.” The impression of tidy central management which the Handbook conveys was somewhat dispelled at trial when the Superintendent testified that the Handbook had been frequently superseded either by informal directive or by common practice. Testimony also indicated that individual principals, whether by design or oversight, felt free to make minor modifications in diocesan programs to fit the particular needs of the school.
The main thrust of the Handbook, however, remains clear. The keynote of education in a parochial school is perhaps best reflected in the Handbook’s remarks about the importance of a teacher: “The prime factor for the success or failure of the school is the spirit and personality, as well as the professional competence, of the teacher. * * * ” Consistent with this view is the Handbook’s view of the scope of religious activities : “Religious formation is not confined to formal courses; nor is it restricted to a single subject area.” Teachers are further advised by the Handbook to try to stimulate interest
in pursuing religious vocations and in missionary work.
This concern for religious values does not necessarily affect the content of secular subjects in diocesan schools. On the contrary, several teachers testified at trial that they did not inject religion into their secular classes, and one teacher deposed that he taught exactly as he had while employed in a public school. This testimony gains added credibility from the fact that several of the teachers were non-Catholics. Moreover, because of the restrictions of Rhode Island’s textbook loan law, (R.I.Gen.Laws §§ 16-23-2, 16-1-9 (Supp.1967), and the explicit requirement of the Salary Supplement Act, teaching materials used by applicants for aid must be approved for use in the public schools.
Nevertheless, the diocesan schools do possess a number of distinctively religious attributes which distinguish them from public schools. Each parochial school is in close proximity to the parish church, allowing a convenient access for Mass or other religious exercises. The exterior of the school often has an identifying cross; each classroom contains a crucifix and school interiors generally exhibit religious paintings and statuary. Classes each day begin with a prayer; some schools schedule additional times for prayer. Roughly thirty minutes a day or one hundred fifty minutes a week are devoted to religious instruction. Assembly programs sometimes include a speaker on religious vocations or missionary activity, and some extra-curricular activities, such as altar or choir boy training and missions programs, are religiously oriented. Students and faculty participate in a yearly retreat and all teachers participate in a two day annual meeting of the Catholic Teachers Institute in which the teaching of religion is occasionally discussed. Occasional guidance counselling, including religious counselling, takes place at the elementary level.
In addition to these factors, the Church gains special advantage from a curriculum which contains both religion and secular subjects. When religion is part of the school day, children can be instructed in their faith more frequently, with less absenteeism, and with greater attention to the religious development of each child. Moreover, religious instruction can be staggered throughout the day to utilize more effectively the limited number of qualified teachers of religion. Finally, we note that the presence of dedicated sisters, though diminishing in numbers, provides an atmosphere where religious instruction and religious vocations seem natural parts of everyday life. Indeed, concern for this intangible “religious atmosphere” has led the Superintendent to decide tentatively to attempt to maintain a one sister to one lay teacher ratio in most of the diocesan schools rather than permitting some to be staffed predominately by lay teachers. To go below this norm would, the Superintendent feared, dilute the Catholic atmosphere of the parochial schools.
Implicit in these findings, we think, is the further finding that the diocesan school system is an integral part of the religious mission of the Catholic Church. It is not that religious doctrine overtly intrudes into all instruction. Rather the combined conveniences of ready access to church and pastor, homogeneous student body, and ability to schedule throughout the day a blend of secular and religious activities makes the parochial school a powerful vehicle for transmitting the Catholic faith to the next generation. It also seems clear to us that this essentially religious enterprise cannot succeed without good teaching in secular subjects. If the quality of teaching falls too low, then not only will Catholic parents be reluctant to enroll their children, but the parochial schools would also run afoul of Rhode Island’s education laws.
Good secular teaching is as es
sential to the religious mission of the parochial schools as a roof for the school or desks for the classrooms.
These determinations point to two ultimate facts. On the one hand, we find that the statute will have the significant if temporary secular effect of aiding the quality of secular education in Rhode Island’s Roman Catholic elementary schools. On the other hand, we think it equally clear that the Act gives significant aid to a religious enterprise.
II.
Conclusions of Law
A.
Plaintiffs’ First Amendment Claims
With these findings in mind, we turn to the central legal issue of the case, whether Rhode Island’s Salary Supplement Act violates the First Amendment’s provision that “Congress shall make no law respecting an establishment of religion, or prohibiting the free exercise thereof”, as applied to the states through the Fourteenth Amendment. Cantwell v. Connecticut, 310 U.S. 296, 60 S.Ct. 900, 84 L.Ed 1213 (1940); Everson v. Board of Education,
supra
n. 1.
Plaintiffs’ free exercise claim need not long detain us. In their complaint, plaintiffs alleged that the statute infringed on the free exercise of their religion by forcing them to pay taxes for religious purposes. But plaintiffs offered no testimony concerning their personal religious beliefs and practices, or lack thereof. Thus they have failed to introduce the kind of particularized evidence necessary “to show the coercive effect of the enactment as it operates against [them] in the practice of [their] religion.” School District of Abington Township, Pennsylvania v. Schempp, 374 U.S. 203, 223, 83 S.Ct. 1560, 1572, 10 L.Ed.2d 844 (1963).
The heart of plaintiffs’ case lies in their claim that the Act violates the Establishment Clause. Much of the dispute at trial centered on the appropriate standard for gauging this claim. As of the date of argument in this case, there were three dominant precedents from which the parties drew sustenance. The first was
Everson,
in which the Court upheld a New Jersey statute which permitted reimbursement to parents for bus fares of children attending both public and parochial schools. The second was
Schempp,
proscribing prayers in public schools. The third was Board of Education of Central School Dist. No. 1 v. Allen, 392 U.S. 236, 88 S.Ct. 1923, 20 L.Ed. 2d 1060 (1968), which upheld a New York law requiring local school boards to loan textbooks to children attending parochial schools.
Plaintiffs have mounted a preliminary attack based on a statement in
Everson
that, “No tax in any amount, large or small, can be levied to support any religious activities or institutions.” 330 U.S. at 16, 67 S.Ct. at 511. Since the Act appropriates tax monies for employees of religious institutions, plaintiffs maintain, it is clearly unconstitutional. Whatever the continued vitality of the
Everson
test, however, we are satisfied that it cannot be applied in such a literal manner. Although in both
Everson
and
Allen,
tax funds were used to support religious institutions, in both the aid was sustained. Plaintiffs distinguish these cases on the grounds that the cost of bus fares in
Everson
and textbooks in
Allen
was originally borne by the parents rather than the schools. But assuming that state subsidies of buses or books otherwise pass constitutional muster, we find it difficult to believe that these laws would fail because prior to enactment parents bore the costs indirectly, in the form of higher school bills, rather than directly.
Accord,
Lemon v. Kurtzman, 310 F.Supp. at 47-48; Tilton v. Finch, 312 F.Supp. 1191, at 1197-1198 (D.Conn. March 19, 1970).
Perhaps recognizing the shortcomings of the
Everson
formulation, both plaintiffs and defendants have focused their argument on the purpose and effect test relied on in
Allen,
392 U.S. at 243, 88 S.Ct. 1923, and first enunciated in
Schempp:
“The test may be stated as follows: what are the purpose and the primary effect of the enactment? If either is
the advancement or inhibition of religion then the enactment exceeds the scope of the legislative power as circumscribed by the Constitution. That is to say that to withstand the strictures of the Establishment Clause there must be a secular legislative purpose and a primary effect that neither advances nor inhibits religion. * * * ” 373 U.S. at 222, 83 S.Ct. at 1571.
The first part of this test — determining the statute’s purpose — presents little difficulty. The Salary Supplement Act, in our opinion, was not intended to advance or interfere with religion. The statute itself recites that its purpose is “to provide a quality education for all Rhode Island youth, those in public and non-public schools alike.” Certainly, the quality of education in non-public schools is a legitimate legislative concern. Board of Education, etc. v. Allen, 392 U.S. at 245-247, 88 S.Ct. 1923. We find nothing in the history and structure of the Act which suggests that the legislative purpose was other than as declared.
In reaching this result, we necessarily reject plaintiffs’ attempt to divine the “true intent” of the legislature by inspecting the activities of lobbyists. As Mr. Justice Frankfurter observed in McGowan v. Maryland, 366 U.S. 420, 469, 81 S.Ct. 1101,1158, 6 L.Ed.2d 393 (1961), “[T]he private and unformulated influences which may work upon legislation' are not open to judicial probing.”
See
United States v. O’Brien, 391 U.S. 367, 383-385, 88 S.Ct. 1673, 20 L.Ed.2d 672 (1968). Nor do we think that the legislature revealed a forbidden motive when it stated that the proposed legislation was designed “to assist non-public
schools
to provide salary scales which will enable them to retain and obtain teaching personnel who meet recognized standards of quality.” (Emphasis added.) Phrasing the test in terms of schools instead of students does not, we think, obscure the legislature’s legitimate concern for the quality of education afforded to Rhode Island youth. Finally, while the restriction of aid to schools whose per pupil expenditure does not exceed that of the public schools may result in excluding, for the most part, private but non-parochial schools from the scope of the statute, this provision also serves the legitimate secular purpose of saving state funds by eliminating schools which can well afford qualified teachers.
The second part of the
Schempp
test— determining the statute’s “primary effect” — presents a more difficult problem of both definition and application. Plaintiffs have argued that “primary” means “essential” or “fundamental”. Defendants and intervenors have taken a more literal position, claiming that “primary” means “first in order of importance”. The problem of definition is critical in this case because, as we have noted in our findings, the Act has two significant effects: on the one hand, it aids the quality of secular education; on the other, it provides support to a religious enterprise. Judicial efforts to decide which of these effects is
“the
primary effect”, as
Schempp
seems to require, are likely to be no more satisfactory than efforts to rank the legs of a table in order of importance.
Intervenors, in an effort to give shape to the
Schempp
test, have urged us to focus solely on the activity subsidized in judging effect. In other words, intervenors propose that we examine only the activities of the teachers receiving aid and ignore the religious nature of the schools in which they teach. Such a narrow perspective, however, strikes us as unrealistic when examining direct financial aid to denominational schools. The expenses of a religious institution may be apportioned in a variety of ways among its “secular” and “religious” activities. Under intervenors’ proposed
test, sophisticated bookkeeping could pave the way for almost total subsidy of a religious institution by assigning the bulk of the institution’s expenses to “secular” activities.
Equally important, government aid to purely secular activity may nevertheless involve the state so deeply in the workings of religious institutions “as to give rise to divisive influences and inhibitions of freedom.” School District of Abington Township, Pennsylvania v. Schempp, 374 U.S. at 307, 83 S.Ct. at 1616 (Goldberg, J. concurring). Such embroilments would escape the narrow focus of intervenors’ test.
These deficiencies in the
Schempp
test were, we think, recognized in the recent case of Walz v. Tax Commission of New York City, 397 U.S. 664, 90 S.Ct. 1409 (May 4, 1970), upholding New York’s tax exemption for property used for religious purposes. The test as phrased by Chief Justice Burger is still two pronged:
“Each value judgment under the Religion Clauses must * * * turn on whether particular acts in question are intended to establish or interfere with religious beliefs and practices or have the effect of doing so.” 398 U.S. at 669, 90 S.Ct. at 1412.
The Chief Justice’s phrasing and application of the effect test, however, mark a. significant refinement. Gone are the intimations of
Schempp
that the single predominant effect of a statute may be isolated by a process of deductive reasoning based on principle and precedent.
Instead, the Court focuses on whether the statute fosters “excessive entanglement” between government and religious institutions. Such entanglement is variously characterized as “sponsorship”, “interference”, and a relationship generating “confrontation and conflicts”. Most pertinently, in discussing the alternatives of taxing or exempting church property, the Chief Justice observes:
“[T]he questions are whether the involvement is excessive, and whether it is a continuing one calling for official and continuing surveillance leading to an impermissible degree of entanglement. Obviously a direct money subsidy would be a relationship pregnant with involvement and, as with most governmental grant programs, could encompass sustained and detailed administrative relationships for enforcement of statutory or administrative standards * * *.” 397 U.S. 675, 90 S.Ct. at 1414.
Walz
thus makes it clear that the test of a statute’s effect is not whether the secular result is more important than the religious result, nor whether the activity aided is in form secular, but whether the degree of entanglement required by the statute is likely to promote the substantive evils against which the First Amendment guards.
When the effect of the Salary Supplement Act is judged by this standard, two significant dangers appear. First, the Act authorizes a subsidy which, unlike the exemptions involved in
Walz,
must be annually renewed. In a nation of many sects, the efforts to obtain government financing of those few denominations which support school systems will inevitably excite bitter controversy. In
some jurisdictions where the majority belong to a single denomination, state subsidies of sectarian schools may amount to an establishment in the classic sense. In Rhode Island itself, sectarian controversy is likely to be exacerbated by the continuing financial crisis in its parochial schools, and by the open-ended nature of the Act. If this grant is constitutional, we would find it difficult to distinguish a 50 or 100 per cent subsidy of teachers’ salaries, or a percentage subsidy of the total cost of “secular” education such as Rhode Island already provides its public schools.
On the other side of the coin, significant state subsidies will inevitably produce significant state limitations on the freedom of denominational schools. A direct subsidy of the educational process such as this statute provides requires greater administrative involvement between church and state than the payment of transportation costs involved in
Everson
or the loaning of textbooks in
Allen.
In
Everson,
the local school committee simply reimbursed the parents of parochial school children for bus fares. In
Allen
public school officials — in theory, at least — performed their customary task of determining which books were suitable for public school use.
Compare
R.I.Gen.Laws § 16-1-9 (Supp. 1967).
Under this statute, by contrast, the Commissioner of Education must monitor parochial school expenditures and in close cases determine which spending is “religious” and which “secular”. This ongoing administrative relationship between government and the Catholic schools creates the possibility of legal disputes over the “secular-religious” distinction,
cf.
Presbyterian Church in the United States v. Mary Elizabeth Blue Hill Memorial Presbyterian Church, 393 U.S. 440, 89 S.Ct. 601, 21 L.Ed.2d 658 (1969), and may well influence the allocation of church resources as the parishes tailor their spending to meet the Commissioner’s criteria.
Equally important, the kind of involvement which this statute promotes may significantly limit the internal freedom of parochial schools. Already the Act has restricted the role of teachers. The evidence before us indicates that some otherwise qualified teachers have stopped teaching courses in religion in order to qualify for aid under the Act. One teacher, in fact, testified that he no longer prays with his class lest he endanger his subsidy. Similar restrictions may be expected on the content of courses. In this very case we have been forced to inquire into the teaching in Rhode Island’s Catholic schools in order to determine whether religion “permeates” the teaching of secular subjects.
See also
Board of Education, etc. v. Allen, 392 U.S. at 248, 88 S.Ct. 1923. Surely such inquiries represent a substantial limitation on the freedom of parochial schools to set their own curriculum. Moreover, as Judge Hastie pointed out dissenting in Lemon v. Kurtzman, 310 F.Supp. at 52, private conduct which is heavily subsidized by the state may be viewed as state action and subjected to the same standards of impartiality which we demand of the government. Burton v. Wilmington Parking Authority, 365 U.S. 715, 81 S.Ct. 856, 6 L.Ed.2d 45 (1961);
cf.
Evans v. Newton, 382 U.S. 296, 86 S.Ct. 486, 15 L.Ed. 2d 373 (1966).
Applying these stand
ards to parochial schools might well restrict their ability to discriminate in admissions policies,
cf.
Pennsylvania v. Board of Directors of City Trusts, 353 U.S. 230, 77 S.Ct. 806, 1 L.Ed.2d 792 (1957), and in the hiring and firing of teachers,
cf.
Keefe v. Geanakos, 418 F.2d 359 (1st Cir. 1969).
At some point the school becomes “public” for more purposes than the Church could wish. At that point, the Church may justifiably feel that its victory on the Establishment Clause has meant abandonment of the Free Exercise Clause.
In short, we see as the necessary effects of the kind of legislation involved here not only substantial support for a religious enterprise, but also the kind of reciprocal embroilments of government and religion which the First Amendment was meant to avoid. We therefore hold that the Salary Supplement Act results in excessive government entanglement with religion and thus violates the Establishment Clause of the First Amendment.
Accord,
Opinion of the Justices, 261 A.2d 58 (Me. Jan. 15, 1970).
B.
Intervenors’ First Amendment and Equal Protection Claims
There remains for our consideration the equal protection claim of the teacherintervenors and the free exercise claim of the parent-intervenors. The equal protection claim is the more elusive of the two. Teacher-intervenors apparently assume a Salary Supplement Act restricted by judicial decree to aiding teachers in non-public but secular schools. Such a statutory scheme, intervenors argue, would violate equal protection because it discriminates among teachers of secular subjects on an impermissible ground — i. e., religion. Their argument remains somewhat hypothetical since, as we have seen, the only applicants for aid have been parochial school teachers. Intervenors are, in other words, complaining of the abstract possibility of discrimination rather than the thing itself. Moreover, intervenors’ hypothetical scheme would not discriminate on the basis of the teachers’ personal beliefs — -surely a suspect classification — but on the basis of the institutions which employ them. This distinction between secular institutions which are eligible for subsidy and religious institutions which are not seems to us commanded by the Establishment Clause. Thus, even though religion in general may be a suspect classification, the mandate of the Establishment Clause provides an overriding justification in this case.
Cf.
Korematsu v. United
States, 323 U.S. 214, 216, 65 S.Ct. 193, 89 L.Ed. 194 (1944); Loving v. Virginia, 388 U.S. 1, 11, 87 S.Ct. 1817, 18 L.Ed.2d 1010 (1967).
The parent-intervenors present a more plausible claim, based on the Free Exercise Clause. Their complaint rests on the conviction that children take their values to a large extent from their school environment; a school without religion conveys to the child the notion that religion is unimportant. Because of this conviction, parent-intervenors feel in conscience bound to send their children to parochial schools which teach both secular subjects and religion. If, however, the quality of secular education falls too low in parochial schools, parentintervenors may well be forced to ignore the dictates of conscience by sending their children to public schools. To avoid this result, intervenors argue that the free exercise benefits which flow from aid to parochial education should prevail over the establishment clause values protected by strict separation, citing in support of their argument Sherbert v. Verner, 374 U.S. 398, 83 S.Ct. 1790, 10 L.Ed.2d 965 (1963).
While sympathetic to intervenors’ dilemma, we note that it stems primarily from their personal evaluation of public school education. We reject the notion that the Free Exercise Clause demands affirmative state action to accommodate such personal evaluations when society at large has accepted the premise that religious and secular education can be successfully separated.
See
Freund, Public Aid to Parochial Schools, 82 Harv. L.Rev. 1680, 1686-1687 (1969). Certainly it would be anomalous if the First Amendment required the state to exclude religion from the public schools but at the same time to support an entire separate school system in order to facilitate the teaching of religion. Such a result is not compelled by Sherbert v. Verner,
supra,
which required a minor change in a welfare program to accommodate Sabbatarians. The Court was careful to note that such minor accommodation did not entail the kind of involvement of religious and secular institutions forbidden by the Establishment Clause. 374 U.S. at 409, 83 S.Ct. 1790. The Court also noted that some burdens on the free exercise of religion could be justified by a compelling state interest requiring uniformity of regulation. 374 U.S. at 406-409, 83 S.Ct. 1790. Our ease falls within both these exceptions. Not only does the Salary Supplement Act embroil religious and secular institutions, but the Establishment Clause values can be preserved only by denial, of direct subsidies to parochial schools. We therefore reject the free exercise claim of parent-intervenors.
III.
Conclusion
Plaintiffs are entitled to a declaratory judgment that R.I.Public Laws c. 246 (1969) violates the First Amendment of the United States Constitution insofar as it authorizes aid to teachers employed by denominational schools, and to an injunction forbidding defendants and their agents from making disbursements to such teachers.
PETTINE, District Judge (concurring in result, concurring in part and dissenting in part with the Opinion of the Court).
The majority decision holds the Salary Supplement Act unconstitutional for two reasons: (1) its effect is substantially to support a religious enterprise; (2) its operation involves those “reciprocal embroilments of government and religion which the First Amendment was meant to avoid.” I concur in the holding of unconstitutionality for the latter reason. However, my disagreement with the former reason requires my separate concurrence.
The Salary Supplement Act does not substantially support a religious enterprise; rather it supports a most important facet of the educational process in modern America, the teaching of all that knowledge presently appropriate for learning, exclusive of religion. That a distinction between the teaching of religion and non-religious subject matter
in sectarian schools may be maintained for Establishment Clause pui'poses seems clear to me from the decision of the Supreme Court in Board of Education of Central School Dist. No. 1 v. Allen, 392 U.S. 236, 88 S.Ct. 1923, 20 L.Ed.2d 1060 (1968). I read
Allen
to invite the inquiry whether the state program under scrutiny has as its effect the advancement of a religion. In answering that inquiry it cannot be presumed that secular aspects of the educational process are permeated by religious values emanating from the climate of the educational environment. Only proof will establish that subsidization of an educational enterprise is subsidization of a “religious enterprise.” In the instant case not only is the record barren of any proof that the teaching of secular subject matter is a “religious enterprise,” but also there is unanimous unrebutted testimony from several teachers of secular subjects in Roman Catholic schools that religion does not enter into their teaching processes.
I cannot accept the majority’s proposal that exclusive focus on the educational function sought to be subsidized is too narrow a standard because “sophisticated bookkeeping could pave the way for almost total subsidy of a religious institution by assigning the bulk of the institution’s expenses to ‘secular’ activities.” Such reasoning equates the legal and bookkeeping definitions of “secular” and “religious.” The law is equipped to distinguish bookkeeping entries from legally distinct entities. The secular-religious distinction is fundamental; it will take more than accounting machinations to elude it.
The majority’s rejection of the plaintiffs’ heavy reliance on the principle stated in Everson v. Board of Education, 330 U.S. 1 at 16, 67 S.Ct. 504 at 511, 91 L.Ed. 711 (1947) that “(n)o tax in any amount, large or small, can be levied to support any religious activities or institutions,” would seem to suggest that the
Allen
test is controlling. Paradoxically, however, the majority, in the face of
Allen
and without legal justification, conducts an extensive exploration of the evidence relating to the religious “atmosphere” in the Roman Catholic schools. In my view,
Allen
renders such an exploration irrelevant.
Finally, I would emphasize anew that I agree with the majority that Walz v. Tax Commission of City of New York, (May 4, 1970) supplements
Allen,
to the extent that multiple effects of a statutory scheme may be considered, including its potential to create continuing controversies inimical to the values of the First Amendment.
Walz
does not, however, provide any precedent for the majority’s assumption that support of the secular educational processes of sectarian schools constitutes “substantial support of a religious enterprise.”
With all respect, I concur with the majority’s result and only with that portion of the majority’s opinion which elaborates the implications of the “excessive entanglements” aspect of the
Walz
decision for the instant case.