YOLANDA AGUILAR, ET AL., APPELLANTS v. BETTY-LOUISE FELTON, ET AL.
SECRETARY, UNITED STATES DEPARTMENT OF EDUCATION, APPELLANT V.
BETTY-LOUISE FELTON, ET AL.  CHANCELLOR OF THE BOARD OF EDUCATION OF
THE CITY OF NEW YORK, APPELLANT V. BETTY-LOUISE FELTON, ET AL.

   No. 84-237, No. 84-238, No. 84-239

   In the Supreme Court of the United States

   October Term, 1984

   On Appeal From the United States Court of Appeals for the Second
Circuit

   Brief for the Secretary of Education

                       PARTIES TO THE PROCEEDING

   The Secretary of Education and the Chancellor of the Board of
Education of the City of New York were named as defendants and were
appellees in the court of appeals.  Yolanda Aguilar, Lillian Colon,
Miriam Martinez, and Belinda Williams intervened as defendants in the
district court and were appellees in the court of appeals.
Betty-Louise Felton, Charlotte Green, Barbara Hruska, Meryl A.
Schwartz, Robert H. Side, and Allen H. Zelon were the plaintiffs in
the district court and appellants in the court of appeals.

            TABLE OF CONTENTS
   QUESTION PRESENTED
   PARTIES TO THE PROCEEDING
   Opinions below
   Jurisdiction
   Constitutional, statutory, and regulatory provisions involved
   Statement
   Introduction and summary of argument
   Argument:
      New York's Title I program of remedial educational
      services for educationally deprived children, provided by
      public school professionals on private school premises,
      does not violate the Establishment Clause
      A. This Court has appellate jurisdiction under 28
         U.S.C. 1252
      B. The Title I program of on-premises instruction
         for disadvantaged nonpublic school students has, as the
         court of appeals recognized, "done much good, and little,
         if any, detectable harm"
      C. An analysis of the record demonstrates that
         New York's Title I program has a legitimate
         secular purpose, does not impermissibly advance
         religion, and has not occasioned an excessive
         entanglement between Church and State
         1. Purpose
         2. Effect
         3. Entanglement
         4. The authority of Meek v. Pittenger
      D. The court of appeals' reasons for refusing to
         consider the evidentiary record were misconceived
      E. To deny comparable services to nonpublic school
         students would be inconsistent with the Religion
         Clauses' principles of evenhandedness and
         accommodation
   Conclusion

                            OPINIONS BELOW

   The opinion of the court of appeals (J.S. App. 1a-54a) is reported
at 739 F.2d 48.  The opinion of the district court (J.S. App. 55a-57a)
is unreported.  The opinion in National Coalition for Public Education
& Religious Liberty v. Harris (J.S. App. 60a-103a), on which the
district court relied, is reported at 489 F. Supp. 1248.

                             JURISDICTION

   The judgment of the court of appeals (J.S. App. 104a-105a) was
entered on July 9, 1984.  A notice of appeal (J.S. App. 106a-107a) was
filed on August 2, 1984.  The appeal was docketed on August 13, 1984,
and the Court, on October 9, 1984, postponed further consideration of
jurisdiction to the hearing on the merits.  The jurisdiction of this
Court is invoked under 28 U.S.C. 1252.

     CONSTITUTIONAL, STATUTORY, AND REGULATORY PROVISIONS INVOLVED

   The relevant constitutional, statutory, and regulatory provisions
are set out at J.S. App. 108a-127a.

                          QUESTION PRESENTED

   Whether Title I of the Elementary and Secondary Education Act of
1965, which authorizes federal funding of remedial education for
educationally deprived children (in public or private schools) in low
income areas, violates the Establishment Clause of the First Amendment
insofar as it authorizes the funding of secular remedial classes
taught by public school teachers under public school control on the
premises of religious schools.

                               STATEMENT

   1. a. Title I of the Elementary and Secondary Education Act of
1965, 20 U.S.C. 2701 et seq., is the "cornerstone" of the national
effort to provide special educational assistance to needy children.
H.R. Rep. 95-1137, 95th Cong., 2d Sess. 4 (1978).  Over the almost 20
years of its operation, Title I has provided federal grants-in-aid "to
local educational agencies serving areas with concentrations of
children from low-income families to expand and improve their
educational programs by various means * * * which contribute
particularly to meeting the special educational needs of educationally
deprived children." 20 U.S.C. 2701.  /1/

   Title I was designed by Congress to "'bring better education to
millions of disadvantaged youth who need it most.'" S. Rep. 89-146,
89th Cong., 1st Sess. 5 (1965) (citation omitted).  Educational
services typically made available under Title I include remedial
reading, remedial mathematics, and English as a second language.  See
H.R. Rep. 95-1137, supra, at 6.  By helping these children develop
these basic skills, it is hoped that the cycle of poverty can be
broken at its source.  S. Rep. 89-146, supra, at 5.

   Local programs must satisfy specific statutory criteria in order to
qualify for Title I funds.  20 U.S.C. 3805(b).  The programs must
provide services to students (i) who are educationally deprived, that
is, who perform at a level below normal for their age (see 34 C.F.R.
200.3(b)), and (ii) who live in an area that has a high concentration
of families with incomes below the poverty level.  20 U.S.C. 3805(b).
The Title I funds may be used only to provide services that otherwise
would not be available to the participating students.  20 U.S.C.
3807(b).  /2/

   Since passage of Title I in 1965, it has been universally
recognized as the largest and most successful federal educational
effort.  The program, which typically involves special instruction,
small classes, and time with specially trained teachers (H.R. Rep.
95-1137, supra, at 6), has been successful in reaching "areas that
have the highest proportions of children from low-income families" (S.
Rep. 95-856, 95th Cong., 2d Sess. 7 (1978);  H.R. Rep. 95-1137, supra,
at 5), and has been "extremely effective in enhancing * * *
achievement." Id. at 6.  /3/ For example, participating first graders
studied during the period 1975-1978 registered reading gains averaging
12 months, and mathematics gains averaging 11 months, during seven
months of Title I instruction.  Third graders gained an average of
eight months in reading and 12 months in mathematics.  These compare
with ordinary gains of only seven months for each school year of
instruction for disadvantaged children not served by Title I.  H.R.
Rep. 95-1137, supra, at 6.  In contrast to earlier studies showing
that disadvantaged students "fall more and more behind in their
achievement levels," students participating in Title I tended to keep
pace with "their less-needy, unassisted peers * * * , and in some
instances (achieved) dramatic improvements in reading." Id. at 6-7.

   These results confirm that Title I has been, as hoped, "a very
potent instrument to be used in the eradication of poverty and its
effects." S. Rep. 89-146, supra, at 5.  See H.R. Rep. 95-1137, supra,
at 7.

   b. The equitable participation of eligible children, whether they
attend public or nonpublic schools, has been a fundamental principle
of Title I since its inception.  Congress recognized that many
families in low-income urban areas send their children to nonpublic
schools.  See Wheeler v. Barrera, 417 U.S. 402, 405-406 (1974).
Congress therefore specifically provided that eligible schoolchildren
were not to be excluded from the benefits of Title I merely because
they attend private rather than public school.  See id. at 406.  20
U.S.C. 2740(a), 3806(a) provides:

         To the extent consistent with the number of educationally
      deprived children in the school district of the local
      educational agency who are enrolled in private elementary and
      secondary schools, such agency shall make provisions for
      including special educational services and arrangements * * * in
      which such children can participate * * * .

This provision was interpreted to mean that nonpublic school students
must receive services that are "comparable in quality, scope and
opportunity for participation to those provided to public school
children with needs of equally high priority." Wheeler v. Barrera, 417
U.S. at 407 (citation omitted);  see 45 C.F.R. 116.19(b) (1974).

   The question of providing services to students in nonpublic schools
has been revisited by later Congresses.  Each has, without significant
dissent or controversy, decided to maintain or strengthen the
provisions for equitable participation of nonpublic school children in
Title I.

   In 1974, Congress became concerned that the then-existing mechanism
for enforcement of the nonpublic school student participation
requirement (disapproval of grant applications or withholding of funds
under 20 U.S.C. (1976 ed.) 241f(b), 241(j) was not adequate.  S. Rep.
93-763, 93d Cong., 2d Sess. 29-30 (1974).  Congress therefore enacted
a new "by-pass" provision that would require the Secretary (then the
Commissioner of Education) to supply services to eligible nonpublic
school students in instances where the local educational agency was
prohibited by state or local law, or otherwise substantially failed,
to do so.  Pub. L. No. 93-380, Sec. 101, 88 Stat. 497-499;  see 20
U.S.C. 2740(b), 3806(b).

   In 1978, Congress again considered the matter, having been informed
by a National Institute of Education study that many children in
nonpublic schools were not being served adequately.  H.R. Rep.
95-1137, supra, at 32-33.  Congress's solution (Pub. L. No. 95-561,
Sec. 101, 92 Stat. 2171-2172) was to stiffen the requirement of
comparability.  Now, the statute requires (20 U.S.C. 2740(a), 3806(a)
(emphasis added)) that:

      Expenditures for educational services and arrangements pursuant
      to this section for educationally deprived children in private
      schools shall be equal (taking into account the number of
      children to be served and the special educational needs of such
      children) to expenditures for children enrolled in the public
      schools of the local educational agency.

   The requirement of participation by nonpublic school children has
never encountered significant opposition or controversy, and there has
never been a serious attempt in Congress to eliminate it.  /4/

   c. The Department of Education's regulations implementing the Title
I program (now Chapter 1) are set forth in 34 C.F.R. Pt. 200.  Several
of these regulations specifically address participation in Title I
programs by educationally deprived children enrolled in private
schools.  See 34 C.F.R. 200.70-200.85.  The regulations require that
the local public school boards that administer these programs
(denominated by the regulations as "local educational agencies" or
"LEAs") shall provide eligible private school children with Title I
services that "assure participation on an equitable basis" (34 C.F.R.
200.70(a)(1)).  The LEA must allow such children "to participate in a
manner that is consistent with the(ir) number and special educational
needs" (34 C.F.R. 200.70(b)).  While so doing, however, the LEA must
"exercise administrative direction and control over (Title I) funds
and property" used in such programs (34 C.F.R. 200.70(c)).  See S.
Rep. 89-146, supra, at 12.  The Title I services to private school
children must be provided either by public employees or by contract
with a person or organization "independent of the private school and
of any religious organizations" (34 C.F.R. 200.70(d)(1)).

   Several regulatory provisions are specifically designed to
effectuate the congressional intent that no financial aid or services
be provided "to a private institution," as distinguished from the
educationally deprived schoolchildren who attend a private
institution.  See S. Rep. 89-146, supra, at 11.  Thus, one provision
stipulates, in accordance with express congressional intent (S. Rep.
89-146, supra, at 12), that Title I funds may be used only "to provide
services that supplement the level of services that would, in the
absence of (Title I) services, be available to children in private
schools" (34 C.F.R. 200.72(a)).  Another provision specifies that
Title I funds may be used only "to meet the special educational needs
of children in private schools" and not to meet any "needs of the
private schools" themselves or any "general needs" of the private
school children (34 C.F.R. 200.72(b), (1) and (2)).

   Although the LEAs are permitted to make public employees available
on the premises of private schools, this approach may be used only as
"necessary to provide equitable (Title I) services" and only "(i)f
those services are not normally provided by the private school" (34
C.F.R. 200.73(a) and (b)).  This provision accords with Congress's
original statement of its intent.  Congress expressly indicated that a
program of assistance by public school teachers to nonpublic school
children in their own schools would be proper, but "only to provide
specialized services which contribute particularly to meeting the
special educational needs of educationally deprived children (such as
therapeutic, remedial or welfare services) and only where such
specialized services are not normally provided by the nonpublic
school" (S. Rep. 89-146, supra, at 12).  See also 111 Cong. Rec.
5745-5748 (1965) (remarks of Reps. Carey and Perkins, managers of the
House bill).

   The regulations also provide that "a public agency must keep title
to and exercise continuing administrative control of all equipment and
supplies that the LEA acquires with (Title I) funds" (34 C.F.R.
200.74(a)).  This provision, too, was anticipated by Congress in the
original legislative history.  S. Rep. 89-146, supra, at 12.  Finally,
the regulations expressly prohibit use of Title I funds "for repairs,
minor remodeling, or construction of private school facilities" (34
C.F.R. 200.75).  Cf. S. Rep. 89-146, supra, at 11.

   d. According to the most recent data in the record, over $3 billion
($3,104,317,000) was appropriated for Title I program expenditures in
Fiscal Year 1982 (J.A. 228;  C.A. App. A251).  /5/ During the
1980-1981 school year, Title I services were provided to 5,170.935
public school children and 192,994 private school children (ibid.).
Thus, about 3.7% of the Title I students were in private schools.  And
approximately $105,200,000 -- about 4% of the total Fiscal Year 1980
appropriation -- was expended on Title I services for children
attending private schools (ibid.).

   2. a. This case concerns the largest Title I program in the nation,
that operated by the Board of Education of the City of New York.  This
program has been in operation for 18 years, and the facts concerning
it have been developed in detail in the record of this case.  Those
facts are essentially undisputed (J.S. App. 10a, 56a n.1).

   Approximately 13% of the over 300,000 students eligible for Title I
programs in New York City attend nonpublic schools, most of which are
religiously oriented (J.S. App. 7a;  J.A. 38, 81-82;  C.A. App. A32,
A80-A81).  Title I students are taught remedial reading, remedial
mathematics, and English as a second language and are provided a
clinical and guidance program designed to enhance achievement in those
subjects (J.S. App. 10a-11a).  In accordance with Title I and its
implementing regulations (see 20 U.S.C. 3807(b);  34 C.F.R. 200.72(a),
200.73(b)), Title I funds are used to provide supplemental educational
services for nonpublic school students only when these are not
available in the nonpublic school.

   Initially, the Board did not offer Title I instruction on the
premises of nonpublic schools.  Instead, it required nonpublic school
students who wished to participate in Title I programs to travel to
public schools after regular school hours.  This approach proved, in
the Board's view, to be "a total failure," largely because of the
participants' fatigue and lack of attendance, J.A. 63;  C.A. App. A60.
 /6/ The Board then decided to hold some Title I classes in the
nonpublic schools, but after regular school hours.  Ibid.  This
approach proved unsuccessful for similar reasons:  "both students and
teachers were tired, * * * there was concern about the safety of
children traveling home after dark or in inclement weather, and * * *
communication between Title I teachers and other professionals and the
regular classroom teachers of the nonpublic schools was virtually
impossible" (J.S. App. 8a).  See J.A. 64;  C.A. App. A60.

   The Board then considered holding the remedial classes for
nonpublic school students in the public schools during school hours.
This plan was abandoned because of concerns that it would violate the
New York Constitution (J.S. App. 8a;  J.A. 64;  C.A. App. A60-A61).
In addition, a study in 1977-1978 showed that the transportation and
other noninstructional costs of conducting Title I classes for
nonpublic school students at sites away from their schools would have
amounted to $4.2 million -- an amount equivalent to 42% of the entire
Title I budget for nonpublic school children.  /7/ In order to absorb
these costs, the Board would have had to deny Title I services to over
5,000 children who had been receiving them (J.S. App. 8a-9a, 72a-73a;
J.A. 66-67;  C.A. App. A63-A64).  /8/

   New York's experience with attempts to provide Title I services off
the premises of the participants' regular schools, or after regular
school hours, has not been unique.  Many other school districts have
similarly found that on-premises programs are virtually the only
effective means for providing comparable services to nonpublic school
children.  See Barnes v. Bell, Civ. No. C-80-0501-L(B) (W.D. Ky. filed
Oct. 1, 1980);  Wamble v. Bell, Civ. No. 77-0254-CV-W-8 (W.D. Mo.
filed Apr. 4, 1977);  Dane Affidavit Paras. 12-17, DX U, Tab A-1, at
6-9.  And some communities providing state or local funds for
assistance to nonpublic school students have had the same experience.
See Brief for the United States at 10-11, School District of Grand
Rapids v. Ball, cert. granted, No. 83-990 (Feb. 27, 1984).  Indeed, it
is significant that the Department of Education has, on at least one
occasion, concluded after full investigation that Title I services
provided at times other than during the school day "were not, and
probably could not, be comparable in quality, scope and opportunity
for participation, to services provided during the regular school
hours".  Missouri Study, J.A. 35;  C.A. App. A29.  /9/

   After unsuccessfully "experimenting with (these) alternative
programs" (J.S. App. 71a), New York decided, in 1966, to provide Title
I instruction on the premises of nonpublic schools during school
hours.  All the teachers and other professionals who provide Title I
services, with the exception of some physicians under special
contract, are regular full-time employees of the Board (J.A. 46-47;
C.A. App. A41-A42).  Teachers who are willing are assigned by the
Board to teach nonpublic school Title I students.  The Board does not
inquire into teachers' religious affiliations when making assignments.
 J.A. 48;  C.A. App. A43.  The undisputed evidence is that the vast
majority (73%) of the Title I teachers work in nonpublic schools with
a religious affiliation different from their own.  J.A. 49-50;  C.A.
App. A44-A45;  see J.S. App. 11a-12a, 74a.  In addition, 78% of the
teachers, and all of the non-teacher professionals, spend fewer than
five days a week in any one school and work in more than one school in
the course of the week.  J.A. 49, 82;  C.A. App. A44, A80.

   The New York on-premises program is designed to "'create() the
unusual situation in which an educational program may operate within
the private school structure but be totally removed from the
administrative control and responsibility of the private school'"
(J.S. App. 14a (citation omitted)).  Students for the program are
picked, not by the private school, but by the Board.  J.A. 50-51;
C.A. App. A46.  The Board issues Title I teachers detailed written and
oral instructions that emphasize that they are independent public
school employees who are in no way responsible to the nonpublic school
authorities.  J.A. 50;  C.A. App. A45.  The private school principals
are also informed of the requirement that the Title I teachers' role
be kept distinct from the schools' religious aspects.  J.A. 59-60;
C.A. App. A55-A56.  Title I teachers are instructed not to introduce
any religious matters into their programs.  They are also instructed
not to engage in team teaching or cooperative instructional
activities.  They may consult with a nonpublic school teacher about a
student's needs, but, if they do, they are not to engage in any
religious discussion.  J.A. 50-52;  C.A. App. A45-A47;  see J.S. App.
12a, 74a.

   Pursuant to instructions given by the Board to participating
nonpublic schools, Title I teachers use classrooms that are
specifically designated for Title I instruction and that are free from
any religious symbols.  J.A. 59;  C.A. App. A55.  The nonpublic
schools are not reimbursed for classroom space.  The materials used in
the classes have no religious content.  J.A. 56;  C.A. App. A52.  The
Board retains title to the materials and equipment used in Title I
classes.  The teachers are instructed to keep the materials locked in
storage cabinets when they are not in use;  and the materials are
subject to an annual inventory.  J.A. 56-57;  C.A. App. A52-A53;  see
J.S. App. 13a, 74a-75a.

   Each Title I teacher is supervised by a field supervisor, employed
by the Board, who is to make at least one unannounced visit a month to
the Title I classroom.  J.A. 53;  C.A. App. A48.  The field
supervisors answer to the Board's program coordinators, who also make
occasional unannounced visits.  In addition, the Board holds monthly
training sessions for those employees serving as Title I
professionals.  No Title I teacher, in the entire time that
on-premises instruction has been provided, has complained that
nonpublic school authorities were attempting to interfere in his work
for religious reasons;  nor is there any recorded complaint that a
teacher was injecting religious matters into a class.  J.A. 52, 55,
80-81;  C.A. App. A47, A50, A79;  see J.S. App. 13a-14a, 87a.

   b. The record shows that the disadvantaged children who have
participated in New York's on-premises Title I program have made
significant educational progress.

   New York City's Title I Corrective Reading Program served 11,789
disadvantaged students in the nonpublic schools in the 1979-1980
school year.  The program consisted of small group instruction by
full-time public school reading specialists (in groups of 10 or fewer
children), using multi-media as well as conventional reading
materials.  Participating students at all grade levels achieved gains
in performance well above those of a nationally representative group.
While the Title I nonpublic school students tested at the 14th to 18th
percentile in reading comprehension prior to participation, at the end
of the school year they had advanced to between the 27th and the 30th
percentile.  See Final Evaluation Report, J.A. 120-124;  C.A. App.
A127-A131.

   New York City's Reading Skills Program, a one-on-one program for
children with severe reading problems, served some 431 nonpublic
school students during 1979-1980.  The results here, too, were highly
positive.  The students advanced from between the 10th and 16th
percentiles, to between the 22d and 32d percentiles.  See Final
Evaluation Report, J.A. 159-161;  C.A. App. A171-A174.  Similarly, the
Corrective Mathematics Program, which served 8,547 nonpublic school
students, and the English as a Second Language Program, which served
3,360 nonpublic school students, saw participants improve their skills
dramatically.  See Final Evluation Report, J.A. 89-91, 187-189;  C.A.
App. A90-A94, A205-A207.  /10/

   3. Appellees, six federal taxpayers, brought this suit in 1978 in
the United States District Court for the Eastern District of New York.
 They contended that the Constitution prohibits public employees from
providing remedial education on the premises of religious schools
during regular school hours.  They sought declaratory and injunctive
relief against the operation of New York City's Title I program.  Four
individuals whose children attend private elementary schools in New
York City and receive Title I educational assistance subsequently
intervened as defendants (J.S. App. 9a-10a;  see J.A. 2, 9-13;  C.A.
App. A2, A3-A7).

   The district court stayed proceedings in this case pending the
outcome of another suit, also challenging New York City's program of
on-premises Title I instruction, which was pending before a
three-judge court in the United States District Court for the Southern
District of New York.  National Coalition for Public Education &
Religious Liberty (PEARL) v. Harris, 489 F. Supp. 1248, appeal
dismissed for want of jurisdiction, 449 U.S. 808 (1980) (J.S. App.
60a-103a).  /11/ The three-judge court in PEARL unanimously upheld the
constitutionality of the program.  The parties to this case then
stipulated that this case would be heard on the record developed in
PEARL, as supplemented by various affidavits and documents (J.S. App.
10a, 56a).

   The district court granted summary judgment for the defendants
(J.S. App. 55a-59a).  The court agreed with the reasoning of the
three-judge court in PEARL and rejected the plaintiffs' argument that
Meek v. Pittenger, 421 U.S. 349, 367-373 (1975), compelled the
conclusion that the Title I program was unconstitutional (J.S. App.
56a-57a).

   The court of appeals, relying principally on Meek v. Pittenger,
supra, reversed (J.S. App. 1a-54a).  The court did not question any of
the factual conclusions reached by the district courts that had
considered New York's Title I program (see id. at 10a).  The court
explicitly acknowledged the considerable good the program had
accomplished (id. at 4a, 52a).  The court nevertheless held that "the
Establishment Clause, as it has been interpreted by the Supreme Court,
constitutes an insurmountable barrier to the use of federal funds to
send public school teachers and other professionals into religious
schools to carry on instruction, remedial or otherwise" (J.S. App. 4a
(citations omitted);  see id. at 36a-39a, 50a-53a).

   The court of appeals interpreted Meek as creating a per se rule
that the supervision needed to ensure that public employees not
further a religious school's religious purposes necessarily creates "a
constitutionally excessive entanglement of church and state" (J.S.
App. 36a (footnote omitted)).  The court specifically stated that it
was not ruling on "the merits of the argument" that the supervision of
public school teachers in a nonpublic school did not, on this record,
create an unconstitutional degree of entanglement (id. at 53a);  it
made clear that it simply felt itself to be bound by the dictates of
Meek.

                 INTRODUCTION AND SUMMARY OF ARGUMENT

   This case touches on the national commitment to equal educational
opportunity.  It raises the question whether that commitment is
hobbled by a constitutional rule that precludes Congress and the
states from continuing a highly successful program to provide remedial
education to disadvantaged children on an even-handed basis --
including those whose beliefs lead them to exercise their
constitutional right to attend private religious schools.

   In a day when skepticism and controversy surround virtually every
governmental undertaking, a remarkable consensus supports both the
ends of this program and the means adopted to achieve them.  The
purpose of Title I is to enhance equal opportunity -- to give more
Americans a chance to achieve the material and intellectual
fulfillment that only education can provide.  The method adopted is to
enlarge access to specialized instruction in educational fundamentals.
 There is nothing fancy or trendy in these programs;  they bring the
basics (remedial English, reading, mathematics) to youngsters who are
educationally handicapped by poverty and who otherwise would be
deprived of this expert help because they attend schools that do not
provide it.

   The key to the program is that the aid is for children.  It is not
schools that are eligible, but children.  And, from the beginning,
Congress has been insistent that all children who need the program
must have equal access to it, no matter what school they go to.  A
program that seeks to enlarge equal opportunity is not to be
distributed in a way that discriminates.

   It is in pursuit of these aims that the widely admired program at
issue here has developed.  For almost 20 years New York City, funded
by the federal government, has been providing basic remedial education
to educationally handicapped poor children -- children who would
otherwise go without it.  More than 20,000 of these children attend
private (including private religious) schools -- as of course they
have a right to do.  New York has worked conscientiously to satisfy
the statute's central command of equitable participation for all
eligible children.  And by a process of trial and error New York has
discovered what common sense would in any event teach -- that the
program can be truly successful and cost-effective only if the
remedial instruction takes place during school hours in the school
that the child actually attends.  In this setting of poverty and
cultural deprivation, the chance that help will actually reach the
student who needs it is dramatically impaired if the remedial
instruction is available only at a distance or outside of normal
hours.  So New York has adopted the educationally correct and sensible
solution:  it has brought the remedial program to the schools that the
eligible students actually attend.  And on this basis the program has
flourished, an unquestionable success.

   But the court of appeals tells us that all this must cease.  That
court concedes that this program "has done * * * much good and little,
if any, detectable harm" (J.S. App. 52a), and that "other ways of
using Title I funds for the benefit of students in religious schools *
* * are almost certain to be less effective, more costly, or both"
(ibid.).  But we cannot have this admirable and desirable program
because, we are told, the Constitution contains a rigid per se rule
that the use of public funds for secular instruction by secular
teachers under secular control on the premises of religiously oriented
schools constitutes an "establishment" of religion.

   The question we raise is:  Why?  Is this per se rule compelled by
the language or history of the Establishment Clause?  Nobody so
asserts.  Rather, the rule rests on a theory about facts.  The theory
is that publicly funded remedial instruction, provided to
educationally disadvantaged children by public school teachers under
public school control on premises borrowed from religious schools,
will inevitably be subverted and turned to religious ends -- unless
there exists a level and intensity of surveillance over the enterprise
that itself constitutes an impermissible "entanglement" between Church
and State.

   So we turn, then, to the facts.  Is there a body of actual
experience that has been put forward to support these empirical
assertions?  The record is clear:  elaborate and conscientious inquiry
has been made by two district courts into the actual New York
experience.  And that experience shows that the program has in fact
done only what it is supposed to do -- benefit the secular education
of deprived youngsters -- and has not been subverted or turned to
religious or sectarian ends.  Further, the record shows that this has
been accomplished without any extraordinary or impermissible
surveillance -- that no detectable "entanglement" has in fact
occurred.  But, we are told by the court of appeals, the actual record
of experience of the New York schools is irrelevant and may not be
examined.  Rather, the proposition that -- absent extraordinary and
constitutionally impermissible precautions -- the program will be
turned to impermissible ends is simply to be assumed, a priori.  And
this, we are told, is required by this Court's cases.

   In the face of this holding, our submission is straightforward.
Nothing in the Constitution requires us to adopt the unwelcome rule
that Congress and the states are rigidly disabled from rendering
secular remedial assistance to educationally deprived children in
their own schools.  Nothing in this Court's cases should be read to
impose on us the self-defeating ordinance that remedial English and
mathematics can be brought to needy private school children only under
conditions which guarantee that the effort is likely to be a costly
failure.  Nothing forces us to the perverse principle that actual
facts and actual experience must be deemed irrelevant in determining
whether a given program actually constitutes a threat to the values of
the Establishment Clause.

   1. We submit that an analysis of this record in light of this
Court's decisions will reveal, as the district courts concluded, that
New York's on-premises Title I program for nonpublic school students
has a secular purpose, has a primary effect that neither advances nor
inhibits religion, and occasions no excessive entanglement between
governmental and religious authorities.  /12/ The question of purpose
is, indeed, conceded:  no one doubts that improving the education of
needy youngsters is a worthy and legitimate governmental function.
The question of effect, though not addressed by the court of appeals,
was resolved in favor of the program by the district courts.  The
program maintains a thorough neutrality, providing educational
assistance on an equal basis to all eligible children whether they
attend public or private schools.  It provides no direct or indirect
assistance to the nonpublic schools themselves.  Because the remedial
instruction is provided directly to the participating students, and
because only such instructional services as would not otherwise be
available are federally funded, the schools reap no benefit.  The
Title I courses themselves are entirely secular, and the record shows
that the Title I teachers have not allowed religious influences to
affect their teaching.

   2. The court of appeals held the program unconstitutional because
of its potential for excessive entanglement.  In particular, the court
concluded that, under Meek v. Pittenger, 421 U.S. 349 (1975), the
degree of surveillance of Title I teachers that would be needed to
guarantee that religious influences did not insinuate themselves into
the teaching would itself constitute excessive entanglement.  However,
the court did not find, and the record would not permit a finding,
that the present level of surveillance in fact is extensive, let alone
impermissibly "entangling." In the absence of a factual finding that
more extensive surveillance would be needed in order to obviate
impermissible effects, the court should have upheld the program.

   Meek does not compel a contrary conclusion.  Meek should not be
read as committing this Court to a rigid per se rule against
on-premises instruction.  Indeed, decisions before and after Meek
belie any such interpretation.  And Meek itself involved a different
program -- one that lacked adequate guidance and safeguards to prevent
controversies over the respective authority of Title I and private
school personnel, that required the instruction of nonpublic school
students to be conducted on the premises of private schools whether or
not it would be educationally necessary, and that involved annual
appropriations battles with attendant political divisiveness.  Most
importantly, the decision in Meek was made on the basis of an abstract
claim of reliance on the good faith and professionalism of the
teachers rather than, as here, on a 16-year documented history of
successful avoidance of religious influences or entanglements.

   3. The reasons given by the court of appeals for declining to
assess its holding in light of the factual record are not persuasive.
Indeed, they suggest that the court of appeals has improperly reversed
the burden of proof, in effect applying a presumption of
unconstitutionality to this statutory program.  Further, the court's
approach seems quite contrary to the values of evenhandedness and
accommodation that infuse the Religion Clauses.  The effect of the
decision below is to penalize and discourage the exercise of
constitutional rights -- certainly an ironic reading of the First
Amendment.

                               ARGUMENT

 NEW YORK'S TITLE I PROGRAM OF REMEDIAL EDUCATIONAL SERVICES FOR
EDUCATIONALLY DEPRIVED CHILDREN, PROVIDED BY PUBLIC SCHOOL
PROFESSIONALS ON PRIVATE SCHOOL PREMISES, DOES NOT VIOLATE THE
ESTABLISHMENT CLAUSE

     A. This Court Has Appellate Jurisdiction Under 28 U.S.C. 1252

   The court of appeals held that the Establishment Clause of the
First Amendment forbids the expenditure of funds appropriated under
Title I of the Elementary and Secondary Education Act of 1965, 20
U.S.C. 2701 et seq., on remedial instruction for students of nonpublic
religiously oriented schools, if that instruction occurs on the
premises of those schools.  As this Court has held (Wheeler v.
Barrera, 417 U.S. 402, 422-423 (1974)) and as the court of appeals
explicitly recognized (J.S. App. 6a n.2, 24a), Title I authorizes such
expenditures.  Indeed, the legislative history of Title I shows that
Congress specifically contemplated on-premises instruction (S. Rep.
89-146, 89th Cong., 1st Sess. 12 (1965)), and a regulation specifies
that such instruction is to be provided only "(t)o the extent
necessary to" satisfy the statutory mandate that comparable services
be supplied to public and nonpublic school students (34 C.F.R.
200.73(a)).  The record in this case amply demonstrates that
on-premises instruction was in fact required in order to satisfy this
mandate.

   Since the holding of the court of appeals is that the Constitution
prevents Congress from achieving a central purpose of the statute --
that private and public school students have equal access to remedial
instruction -- our submission is that the court of appeals has "held
(Title I) unconstitutional as applied to a particular circumstance"
(United States v. Darusmont, 449 U.S. 292, 293 (1981)) and an appeal
lies to this Court under 28 U.S.C. 1252.  See California v. Grace
Brethren Church, 457 U.S. 393, 404-407 (1982).  Cf. United States v.
Rock Royal Co-operative, Inc., 307 U.S. 533 (1939).  Since the
Secretary is a party to this litigation, he would be bound by a
holding that an on-premises Title I program is unconstitutional.
McLucas v. DeChamplain, 421 U.S. 21, 31 (1975).  While the court of
appeals did not explicitly state that Title I was unconstitutional as
applied (as is normally required to found an appeal under Section
1252), this case is one of those unusual instances in which such a
determination "was a necessary predicate to the relief" that is
granted (United States v. Clark, 445 U.S. 23, 26 n.2 (1980)).  /13/

   We note that in Flast v. Cohen, 392 U.S. 83, 88-91 (1968), this
Court held that a claim that New York City's Title I program violated
the Establishment Clause -- the same claim that is made by plaintiffs
here -- was properly brought before a three-judge court convened
pursuant to 28 U.S.C. (1970 ed.) 2282.  The interpretation of Section
2282 sheds light on the meaning of Section 1252 because Section 2282
provided for a three-judge court when an injunction was sought against
the enforcement of an Act of Congress "on ground of
unconstitutionality," and both Section 1252 and Section 2282 were
enacted as part of the same statute (Judiciary Act of 1937, ch. 754,
Secs. 2, 3, 50 Stat. 752).

   If the Court determines that it lacks appellate jurisdiction in
this case, we request that it decide this case on writ of certiorari
(see 28 U.S.C. 2103).  The importance and national impact of the
question of constitutional law decided by the court of appeals
warrants such review.

 B. The Title I Program Of On-Premises Instruction For Disadvantaged
Nonpublic School Students Has, As The Court Of Appeals Recognized,
"Done Much Good, And Little, If Any, Detectable Harm"

   This case raises important questions about our national commitment
to equal educational opportunities for all of our country's children
-- whether rich or poor, gifted or slow, devout or indifferent -- all
alike.  That commitment rests on insights that today command virtually
unanimous assent:  education, in our modern society, is indispensable
to a successful and fulfilled life;  it is a fundamental obligation of
government to assure every child an equal opportunity to receive an
adequate education;  breaking the poverty-related barriers to
education is central to breaking other aspects of the poverty cycle
itself;  this entails the creation of special educational programs for
those children who start the course hobbled by the fetters of poverty
and educational deprivation.  See, e.g., Brown v. Board of Education,
347 U.S. 483, 493 (1954).

   Accordingly, Congress has established as a high national priority
the provision of special educational assistance -- especially in the
basic skills of reading, language, and mathematics -- to disadvantaged
youngsters whose schools would not otherwise provide it.  The
"cornerstone" of this effort is Title I of the Elementary and
Secondary Education Act of 1965, 20 U.S.C. 2701 et seq. (see H.R. Rep.
95-1137, supra, at 4), which is based on the recognition that there is
"a close relationship between conditions of poverty and lack of
educational development and poor academic performance." S. Rep.
89-146, 89th Cong., 1st Sess. 5 (1965)).

   From the beginning, Congress had determined that otherwise
qualified disadvantaged youngsters attending nonpublic schools should
be able to participate in Title I programs on an equitable basis;  and
successive Congresses have affirmed and strengthened this
determination.  See pages 5-6, supra.  The concept of parity between
public and private school students reflects two fundamental concerns.

   First, Congress intended Title I to provide aid to children -- not
institutions -- and recognized that many educationally disadvantaged
students in low income areas attend nonpublic schools, often at great
financial sacrifice to their families.  To deny services to classes
and categories of needy students would undermine the objectives of
Title I.  As this Court observed in Wheeler v. Barrera, 417 U.S. 402,
405-406 (1974) (emphasis in original;  footnotes omitted):

      The Congress, by its statutory declaration of policy, and
      otherwise, recognized that all children from educationally
      deprived areas do not necessarily attend the public schools, and
      that, since the legislative aim was to provide needed assistance
      to educationally deprived children rather than to specific
      schools, it was necessary to include eligible private school
      children among the beneficiaries of the Act.

   Second, Congress intended to treat all eligible children equally,
thereby remaining neutral as between public and nonpublic schools.  To
deny Title I services to students who attend nonpublic schools would
discriminate against some children -- those exercising a
constitutional right.

   Indeed, Congress has placed an ever greater emphasis on equality as
experience has shown that, in many instances, disadvantaged students
in nonpublic schools were not receiving their fair share of Title I
assistance.  Thus, in 1978, Congress required for the first time that
expenditures for eligible students in nonpublic schools shall be
"equal" to expenditures for public school children.  See pages 5-6,
supra.  The purpose of this was not to benefit religion or religious
schools, but simply to ensure the "equitable participation of children
enrolled in nonpublic schools." S. Rep. 93-763, 93d Cong., 2d Sess. 30
(1974).

   In guaranteeing equitable participation to all, Congress has
followed the path set by Everson v. Board of Education, 330 U.S. 1
(1947), and Board of Education v. Allen, 392 U.S. 236 (1968), in which
this Court held that "the Establishment Clause does not prevent a
State from extending the benefits of state laws to all citizens
without regard for their religious affiliation" (id. at 242).  Cf.
Widmar v. Vincent, 454 U.S. 263 (1981);  Zorach v. Clauson, 343 U.S.
306, 313-314 (1952).

   While insisting that disadvantaged nonpublic school students be
permitted to participate equitably in the program, Congress has also
been aware that programs of educational assistance to religiously
oriented private school students must be carefully structured to avoid
impermissible advancement of religion or excessive entanglement
between Church and State.  See S. Rep. 89-146, supra, at 11-12.
Congress left decisions on the details of the Title I program to the
discretion of local authorities (within statutory and regulatory
limits).  However, it anticipated that public school teachers would be
made available to disadvantaged students on the premises of nonpublic
schools, and specified that this may be done "only to provide
specialized services which contribute particularly to meeting the
special educational needs of educationally deprived children (such as
therapeutic, remedial or welfare services) and only where such
specialized services are not normally provided by the nonpublic
school." Ibid.  Congress expected that the program could thus be
"interpreted and administered in a manner consistent with the
establishment of religious provision of the first amendment of the
Constitution." Id. at 37 (quoting memorandum from Department of
Health, Education, and Welfare).

   It is in pursuit of these aims, and with these safeguards, that the
program at issue here developed.  For almost 20 years New York City,
funded by the federal government, has been providing basic remedial
education to hundreds of thousands of educationally deprived poor
children -- children who would otherwise go without it.  Some 20,000
of these children (in the 1981-1982 school year) attended private
school.  New York has been conscientious in satisfying the statute's
central command of equitable treatment for all eligible children.

   The record in this case demonstrates that the Title I program in
New York City, like that nationwide, has been highly successful.  The
improved performances of participants in the on-premises program bear
eloquent testimony to the impact this program is having on
disadvantaged children in New York's nonpublic schools.  See pages
13-14, supra.

   In its opinion, the court of appeals recognized the success of the
program, observing that "(w)e have no doubt that the program here
under scrutiny has done much good." J.S. App. 4a.  It also recognized
the practical consequences of its ruling that disadvantaged
schoolchildren attending religiously oriented schools are foreclosed
from receiving remedial assistance from Title I teachers at their
school.  The court "ha(d) no doubt" that the local school authorities
"could reasonably have regarded" the on-premises program "as the most
effective way to carry out the purposes of the Act." J.S. App. 4a.
Indeed, the court concluded that "(w)hile other ways of using Title I
funds for the benefit of students in religious schools can be found,
these * * * are almost certain to be less effective, more costly, or
both." J.S. App. 52a (emphasis added).

   The record graphically substantiates and quantifies the court of
appeals' conclusion.  See pages 9-11, supra.  Additional
transportation and other noninstructional costs alone would consume
some $4.2 million (in 1977-1978 figures) -- more than 42% of the
budget for the nonpublic school Title I program.  In human terms, this
would mean that more than 5,000 children -- a number equivalent to 36
out of every hundred nonpublic school students now being served --
would be forced out of the program.  /14/ And for those children who
continued to be served, as the court of appeals recognized (J.S. App.
7a-8a, 52a), the Title I services would almost certainly be less
effective.  See J.A. 34-37, 63-64, 66-68;  C.A. App. A28-A31, A60,
A62-A64;  Dane Affidavid Paras. 12-17, DX U, Tab A-1, at 6-9.
Moreover, if -- as seems likely /15/ -- remedial instruction under a
revised program would have to take place after regular school hours,
experience documented in the record shows that attendance may be
expected to be poor, and that both teachers and students may be
expected to be too tired for the program to be effective.  See pages
9-11, supra.  The judgment below would thus result in clear and
well-documented injury to thousands of disadvantaged youngsters.

   On the other hand, the court of appeals recognized that, under the
detailed, 16-year record compiled and examined by two district courts,
the efforts of the program "to prevent the public school teachers and
other professionals whom it sends into religious schools from giving
sectarian instruction or otherwise fostering religion" have been
"largely successful." J.S. App. 4a.  In particular, the court did not
overturn the factual finding of the district court in PEARL (J.S. App.
97a), adopted by the district court below (J.S. App. 56a), that the
Title I program in New York City had successfully avoided "an
'impermissible fostering of religion'" without "the 'continuing
surveillance' deemed necessary in Meek." The court of appeals
accurately summarized the evidence in the case when it commented that
the program "apparently has done * * * much good and little, if any,
detectable harm." J.S. App. 52a.

   This case does not involve a casual program, lightly entered into
or easily abandoned.  The program invalidated by the court below
represents a central feature of federal educational policy, now almost
20 years old, established and refined with rare unanimity by
successive Congresses whose deliberations demonstrated full
understanding and sensitivity to the church-state problems involved.
It is a program of major significance to New York and like communities
across the nation, which have, after extended experience and
experimentation, determined that an on-premises program is virtually
the only effective means for providing comparable services to
disadvantaged nonpublic school students.  And thousands of
disadvantaged children have gained precious opportunity from this
program;  they stand to lose grievously if the decision below is
affirmed.

   Under these circumstances, appellees bear a heavy burden of
persuasion that this major program, which has been shown on the record
to have been so needed and successful and at the same time to have
occasioned no "detectable harm" to First Amendment values, should be
held unconstitutional.  If "'the measure of constitutional
adjudication'" in the Establishment Clause area "'is the ability and
willingness to distinguish between real threat and mere shadow'"
(Marsh v. Chambers, No. 82-23 (July 5, 1983), slip op. 11, quoting
Abington School District v. Schempp, 374 U.S. 203, 308 (1963)
(Goldberg, J., concurring)), appellants here are entitled to wonder
where the "real threat" from this program lies.

      C.  An Analysis Of The Record Demonstrates That New York's

               Title I Program Has a Legitimate Secular Purpose, Does

               Not Impermissibly Advance Religion, And Has Not

               Occasioned An Excessive Entanglement Between Church And

               State

   1. Purpose.  It is neither disputed nor disputable that Title I is
animated by a legitimate and admirable secular purpose:  providing
enhanced educational opportunities for disadvantaged children.  See
PEARL, J.S. App. 76a-77a.  Appellees have so conceded (C.A. Br. 20).
/16/

   The specific decision to provide Title I services to nonpublic
school students on the premises of the nonpublic schools is also the
product of solely secular considerations.  As detailed above (see
pages 9-11, supra), the New York City Board of Education decided to
provide Title I services on the premises of private schools only after
other alternatives were considered and unsuccessfully attempted.
These alternatives -- providing services off the premises or after
school hours -- would serve fewer students and serve them less
effectively.  Dane Affidavit Paras. 12-17, DX U, Tab A-1, at 6-9.  In
some circumstances, the alternatives have been found to fall short of
the statutory requirements of comparable services.  See page 10 & n.9,
supra.  Consequently, the Board turned to the present method as the
only feasible means to provide children in nonpublic schools with
Title I educational services comparable to those in public schools.
See J.A. 63-68;  C.A. A59-A64;  J.S. App. 7a-9a, 71a-73a.

   2. Effect.  The Title I program "is working as Congress intended it
should" (H.R. Rep. 95-1137, supra, at 5 (quoting testimony before the
committee)).  The primary effect of the program has been exactly that
envisioned by Congress:  to improve the educational opportunities of
educationally disadvantaged youngsters in low-income areas.  The
results of the New York on-premises program, discussed at pages 13-14,
supra, fully substantiate the beneficial secular effects of the
program.  /17/

   It is also undeniable that the primary effect of the program has
been neither to advance nor to inhibit religion.  This is because, as
is required by the governing statute, comparable services are provided
to eligible schoolchildren whether they attend public or nonpublic
schools.  If Title I services were provided solely to nonpublic school
students, it might be said that the program had the effect of
promoting religious education.  Conversely, if Title I services were
withheld from students attending religiously oriented schools, it
might be said that the program had the effect of discriminating
against religion.  As it is, the effect of the program is absolutely
neutral.  See Mueller v. Allen, No. 82-195 (June 29, 1983), slip op.
8-10, distinguishing Committee for Public Education & Religious
Liberty v. Nyquist, 413 U.S. 756 (1973).  /18/

   Before this Court is a record documenting the actual operation of
the New York Title I program since its inception.  That record reveals
that, during the program's 16-year documented history, there has been
no known instance of any Title I teacher or professional succumbing to
religious influences by introducing religious materials into the Title
I classroom, becoming involved in the nonpublic schools' religious
activities, or being pressured by nonpublic school personnel to
compromise the secular character of the Title I program.  J.A. 52,
80-81;  C.A. App. A47, 479.  Indeed, the danger of Title I employees
inculcating religion is remote because most teachers do not share the
religion of the schools to which they are assigned and are itinerant
-- teaching in several schools, often of differing religious
sponsorship.  See page 11, supra.  /19/

   It must be stressed, in connection with the "effects" inquiry, that
Title I services are provided directly "to educationally deprived
children rather than to specific schools." Wheeler v. Barrera, 417
U.S. at 406 (emphasis in original).  No tax-funded resources are
placed at the disposal of religious school authorities.  /20/ This
distinguishes the Title I program from many others invalidated by this
Court.  As the Court pointed out in Mueller v. Allen, slip op. 10, it
is "noteworthy that all but one of (the Court's) recent cases
invalidating state aid to parochial schools have involved the direct
transmission of assistance from the state to the schools themselves."
/21/ See also Lemon v. Kurtzman, 403 U.S. 602, 621 (1971) (provision
of direct aid to nonpublic schools was a "factor (that) distinguishes
both Everson and Allen, for in both those cases the Court was careful
to point out that state aid was provided to the student and his
parents -- not to the church-related school");  Walz v. Tax
Commission, 397 U.S. 664, 675 (1970) (noting difficulties presented by
"direct money subsidy" program).  /22/

   As discussed at pages 7-8, supra, Title I and the implementing
regulations are carefully designed to ensure that no financial aid,
materials, or services are provided to the nonpublic schools
themselves.  This is in accordance with clear congressional directives
from the beginning of the program.  See S. Rep. 89-146, supra, at 11.
Most important is the requirement that Title I funding be used "only
to provide services that supplement the level of services" that would
otherwise be provided to the eligible schoolchildren.  34 C.F.R.
200.72(a);  see 20 U.S.C. 2736(c), 3807(b).  /23/ The "supplement, not
supplant" provisions of the law ensure that Title I will not relieve
the nonpublic /24/ schools of any of their educational and financial
obligations.  The record shows that, in deciding what Title I services
to provide, the Board "has been guided by the statutory requirement
that those services supplement, and not supplant, instruction and
instructional services already being provided by the nonpublic schools
themselves." J.A. 44;  C.A. App. A39.  /25/

   In conclusion:  two district courts have engaged in a searching
investigation and analysis of the actual "effects" of the New York
City program at issue here.  Guided by the precedents of this Court,
they concluded that that program does not have the impermissible
effect of advancing religion.  "(T)he substantial evidentiary record
compiled in this case * * * demonstrates to the Court's satisfaction
that the program has remained free from religious influence and has
not promoted the religious mission of the nonpublic schools" (PEARL,
J.S. App. 85a).  The court of appeals did not disturb this finding,
and it should be upheld by this Court.

   3. Entanglement.  The Title I program is so structured that there
is no jurisdictional overlap between public and religious authorities,
and little contact between them.  Public authorities exercise no
control over the activities or personnel of the religious schools (in
contrast to Lemon v. Kurtzman, supra) (J.A. 62;  C.A. App. A58;  see
J.S. App. 99a), and the religious school authorities exercise no
control whatsoever over the activities or personnel of the Title I
program (J.A. 46-49, 52-55;  C.A. App. A41-A44, A48-A51).  The result
is that there is no "excessive entanglement" between religious and
governmental authorities.

   The three-judge district court in PEARL aptly described the
provision of services to nonpublic school students here as a "school
within a school." J.S. App. 98a, quoting Larkin Affidavit Para. 107,
DX U, Tab A-4.  The court found that the Title I classrooms "are free
from religious influences and are distinct from the facilities used by
the nonpublic school for regular instructional activities." J.S. App.
98a.  The court further found that "Title I supervisors maintain a
complete separation between Title I activities and the regular
programs of the parochial schools." Id. at 98a-99a.

   The testimony of Title I administrators, teachers, and support
professionals in this case shows that Title I personnel are given
guidelines stressing the program's independence from the nonpublic
schools.  They are instructed that the Title I program is separate
from and independent of the nonpublic schools' program, and that, as
public employees, their role is purely public and secular.  This
uniform testimony, of record in this case, demonstrates that the Title
I personnel fully comprehend and strictly adhere to these guidelines
restricting their own activities and the use of materials, and that
they view the Title I remedial program and their role as independent
of the instructional program of the nonpublic schools.  See J.A.
50-51, 52-55;  C.A. App. A45-A46, A48-A51;  PEARL, J.S. App. 97a.

   The few administrative contacts between Title I personnel and the
teachers or officials of the nonpublic schools do not constitute
excessive entanglement.  These contacts fall into three categories:
(1) disseminating information about Title I to the administrators, (2)
processing requests for services, and (3) resolving scheduling
problems and other questions concerning the implementation of the
program (J.A. 59;  C.A. App. A55).  These routine administrative
interactions do not give parochial school officials any control over
the program, nor involve Title I personnel in the affairs of the
parochial schools.  PEARL, J.S. App. 100a.  The Title I supervisors
have only casual and periodic contact with nonpublic school
administrators and little, if any, contact with nonpublic school
teachers.  Title I supervisors review and evaluate only the
performance of the Title I teacher in the Title I classroom, and
neither the parochial school program nor its instructional material is
subject to public examination.  PEARL, J.S. App. 99a;  see generally
J.A. 59-63;  C.A. App. A55-A59.

   These minimal contacts scarcely reflect the degree of intrusiveness
by government authority that this Court has cautioned against as
amounting to impermissible entanglement.  Cf. Walz v. Tax Commission,
397 U.S. at 674-676 (routine administrative contacts between
government and religiously affiliated institutions do not constitute
entanglement);  Mueller v. Allen, slip op. 14 (same).  And, as the
Court explained in Committee for Public Education & Religious Liberty
v. Regan, 444 U.S. 646, 660-661 (1980) (footnote omitted), when the
administrative contacts are of the kind that are "straightforward and
susceptible to * * * routinization," there is no excessive
entanglement on the face of the plan, "and we are not prepared to read
into the plan as an inevitability the bad faith upon which any future
excessive entanglement would be predicated." /26/

   There remains the question whether the prevention of impermissible
effects in this context has brought about "(a) comprehensive,
discriminating, and continuing state surveillance," such as would
constitute an excessive entanglement under this Court's decision in
Lemon v. Kurtzman, 403 U.S. at 619.  It is this aspect of the
entanglement inquiry -- and this aspect alone -- that caused the court
of appeals to conclude that, despite the practical educational
considerations and the weight of congressional judgment, remedial
instruction and related counseling services may be provided to
students attending religiously oriented schools "only if such
instruction or services are afforded at a neutral site off the
premises of the religious school" (J.S. App. 35a-36a).

   The court of appeals concluded that to "be sufficiently certain
that public employees, in a program like the present one, will
maintain strict religious neutrality, they and the institutions in
which they work must be subjected to 'comprehensive, discriminating
and continuing state surveillance.'" J.S. App. 36a, quoting Lemon v.
Kurtzman, 402 U.S. at 619.  This need for surveillance arises,
according to the court, because "(t)eachers and those who provide
clinical and guidance services of the sort here at issue are engaged
in occupations in which the performance of one's duties may be subtly
and all too easily influenced by a sectarian milieu." J.S. App. 36a
(footnote omitted).  See also J.S. App. 38a (problem is that a public
school teacher may "succumb() to, or indeed embrace(), the religious
influences that bear on anyone offering instruction or guidance in
rooms that are part and parcel of a religious school").

   In so holding, the court did not rely on the record evidence.  The
opinion of the court includes no reference to any evidence concerning
the actual level of "surveillance" involved in the Title I program.
/27/ Indeed, the court expressly declined to address "the merits of
th(e) argument" that the record here affirmatively negates any
supposition of an excessive degree of entanglement stemming from such
surveillance (J.S. App. 53a).  The holding below was based entirely on
the court's "reasoned apprehension of potentials" (J.S. App. 41a) and
-- more directly -- on an analogy to this Court's decision in Meek v.
Pittenger, supra.

   There is a striking incongruity between the weighty considerations
supporting this important program and the court's speculative, even
chimerical, basis for striking it down.  With all respect, we believe
that the esoterica of the court of appeals' First Amendment
jurisprudence have gotten in the way of the ultimate inquiry --
whether the program here "establishes a religious faith, or tends to
do so." Lynch v. Donnelly, No. 82-1256 (Mar. 5, 1984), slip op. 8-9.
The concerns of the court of appeals, if they are real at all, seem
exceedingly remote from the purposes of the Establishment Clause.  As
explained in Lemon v. Kurtzman, 403 U.S. at 614, the purpose of the
"entanglement" inquiry is "to prevent, as far as possible, the
intrusion of either (church or state) into the precincts of the
other." Justice Brennan has further explained that this purpose is "to
keep the state from interfering in the essential autonomy of religious
life." Marsh v. Chambers, slip op. 9 (dissenting opinion).  The
concern about intrusive surveillance is simply one element bearing on
this purpose.  It was first identified in Lemon v. Kurtzman, supra, in
the context of cash salary supplements for full-time religious school
teachers.  There, the surveillance in question pertained to
individuals "under religious control and discipline" (403 U.S. at
617);  in a very real and dirrect sense, the surveillance constituted
"excessive and enduring entanglement between state and church" (id. at
619).  The government has no business overseeing the agents of the
church to ensure that they do not commit religion.

   By contrast, the surveillance at issue here involves only public
school teachers and public school authorities.  /28/ Such surveillance
obviously entails no relationship between the government and religious
authorities.  It is no different, in principle, from the
"surveillance" that may be necessary to ensure that teachers in the
public schools comply with this Court's school prayer decisions.  The
Constitution does not prohibit entanglement between the government and
its employees.  /29/

   The court of appeals' approach to the entanglement issue was marked
by another, analytical, flaw.  The question of "excessive
entanglement" is, inevitably, a correlative of what is necessary to
prevent impermissible "effects." The question is whether the
prevention of an impermissible effect ("advancing religion") entails
an excessive degree of entanglement.  In this case, this reduces to
the specific issue of what sort and degree of "surveillance" is
necessary to prevent trained public school professionals from
"succumbing" to the "sectarian milieu" of the nonpublic school.  This
is fundamentally an empirical question.  It simply cannot be answered
on an a priori basis.  And where, as here, there is an extensive
factual record bearing on the issue, it is a profound mistake to
refuse to examine the evidence and to rely on conclusory presumptions
about the degree of surveillance that might be necessary to prevent
the teachers from "succumbing" and turning the Title I program into a
forbidden advancement of religion.  See J.S. App. 39a.

   The court of appeals did not seriously suggest (and appellees do
not contend) that the degree and kind of surveillance that was
actually carried out in New York City over the Title I teachers now
teaching in private schools constituted an excessive entanglement.
/30/ Nor did the court find any impermissible "succumbing." But if the
present degree of surveillance does not constitute an excessive
entanglement, and has not in fact had impermissible effects, why
should we presume that much more surveillance is necessary to prevent
these effects?

   A purely hypothetical approach to both effects and entanglement has
the result of invalidating any program of assistance to religious
institutions or individuals attending them.  Cf. Wolman v. Walter, 433
U.S. at 262 (Powell, J., concurring in part).  Where a program has
operated for 16 years or more without any untoward consequences, and
this has been accomplished without "comprehensive, discriminating and
continuing state surveillance" (Lemon v. Kurtzman, 403 U.S. at 619),
it suggests very strongly that entanglement problems are largely
imaginary.

   4. The Authority of Meek v. Pittenger.  The court of appeals based
its conclusion that the New York City Title I program is
unconstitutional on what it deemed to be the compelling authority of
Meek v. Pittenger, supra.  J.S. App. 4a.  /31/ Moreover, the court
felt that Meek required that it so hold without examining "the merits
of th(e) argument" that the surveillance of the public school teachers
in this case did not rise to the level of impermissible entanglement
(J.S. App. 53a).

   Even if the court of appeals were correct that Meek compels a lower
court to reach so unfortunate a result (and we think it was not), this
Court is not so constrained.  /32/ Meek may well provide a close
factual analogy to this case.  But there is every reason to hesitate
before reading Meek as an irrevocable commitment to a rigid per se
rule barring the lower courts from conducting an inquiry into the
actual facts and realities of particular on-premises programs.  Just
one year before Meek, in Wheeler v. Barrera, supra, the Court was
faced with the question of the constitutionality, in the abstract, of
an on-premises Title I program.  The Court declined to decide the
question in the abstract, stating that it should be decided on the
basis of "a careful evaluation of the facts of the particular case."
417 U.S. at 426.  The Court observed that "the First Amendment
implications may vary according to the precise contours of the plan
that is formulated." Ibid.  The Court stated that "a program whereby a
former parochial school teacher is paid with Title I funds to teach
full time in a parochial school undoubtedly would present quite
different problems than if a public school teacher, solely under
public control, is sent into a parochial school to teach special
remedial courses a few hours a week." Ibid.  A rigid interpretation of
Meek would, however, render any such considerations irrelevant.  We
seriously doubt that the Court, one year after being so cautious in
Wheeler, intended in Meek to adopt a per se rule that effectively
preordains the outcome of this and every other on-premises case.

   Two terms after Meek, in Wolman v. Walter, supra, the Court
demonstrated that it did not view Meek as precluding analysis of the
facts of the particular case.  There, the Court addressed the
constitutionality of providing diagnostic services to nonpublic school
students on the premises of their schools.  An absolutist
interpretation of Meek -- i.e., that the danger of public school
personnel succumbing to the sectarian atmosphere of the church-related
school is so great that constitutionally excessive surveillance is
necessarily required to prevent it -- would logically have resulted in
invalidation of the program.  But the program was in fact upheld.  433
U.S. at 241-244.  And the provision of therapeutic services in
"religiously neutral locations" was upheld simply because of the
Court's judgment, on the record, that "the danger perceived in Meek
does not arise" in such circumstances.  Id. at 247.  The Court
recognized the importance of a "practical response to the logistical
difficulties of extending needed and desired aid to all the children
of the community" (id. at 247 n.14) (emphasis added)).

   More recently, in Committee for Public Education & Religious
Liberty v. Regan, 444 U.S. at 661 (citations omitted), this Court
explicitly rejected the broad interpretation of Meek on which the
court below relied:

      Meek v. Pittenger * * * is said to have held that any aid to
      even secular educational functions of a sectarian school is
      forbidden, or more broadly still, that any aid to a sectarian
      school is suspect since its religious teaching is so pervasively
      intermixed with each and every one of its activities.  * * * (T)
      hat case was simply not understood by this Court to stand for
      (such a) broad proposition.

   We therefore seek an understanding of Meek that is narrower and
more consistent with this surrounding legal texture.  In this spirit,
we note several significant differences between Meek and the instant
case.  First, the Title I program is governed by detailed regulations
and guidelines that are made known to Title I personnel and the
nonpublic schools as well.  See pages 6-8, 12, supra.  This has the
beneficial effect of routinizing the conduct of the program and
eliminating the need for ad hoc judgments about the relations between
the religious school and the Title I program.  See Regan, 444 U.S. at
660 (upholding program where it is "straightforward and susceptible to
routinzation").  In sharp contrast, the program in Meek (as the Court
observed) had "the potential for provoking controversy between the
Commonwealth and religious authorities over the extent of the
teachers' responsibilities and the meaning of the legislative and
administrative restrictions on the content of their instruction." 421
U.S. at 372 n.22.

   Second, the Title I program is designed to be a flexible means for
accomplishing an important public purpose.  On-premises programs are
permitted only to the limited extent necessary to fulfill those
purposes.  The record here shows that New York adopted the on-premises
approach only as a last resort.  See pages 9-11, supra.  By contrast,
the Pennsylvania program at issue in Meek was limited by statute to
on-premises services.  See 421 U.S. at 367 ("the services are provided
only on the nonpublic school premises").  This factor bears heavily on
whether the decision to provide on-premises services is made on truly
educational grounds, rather than for the convenience or assistance of
the nonpublic schools.  See Wheeler v. Barrera, 417 U.S. at 428
(Powell, J., concurring) (emphasis added) ("I would have serious
misgivings about the constitutionality of a statute that required the
utilization of public school teachers in sectarian schools.").

   Third, the Title I program is a single statutory scheme that
provides aid to students in both public and nonpublic schools
according to a fixed statutory formula.  See 20 U.S.C. 3806(a).  This
obviates any possibility that the program might foster chronic
controversy over the amount of aid to be given to nonpublic school
students.  In fact, in its nearly 20 years of operation, Title I has
not precipitated religious debate in the political arena;  the concept
of equitable treatment of nonpublic school students has enjoyed
widespread and virtually unanimous support.  /33/ By contrast, the
program in Meek established aid for public and nonpublic school
students through separate statutory schemes requiring separate annual
appropriations decisions.  See 421 U.S. at 352, 372.  This created
risks of "repeated confrontation between proponents and opponents of
the * * * program (and) * * * political fragmentation and division
along religious lines" (id. at 372).

   Finally, the most important distinction between this case and Meek
is the procedural posture in which the two cases arose.  This case
comes to the Court after a full trial on the merits by two district
courts, with an extensive factual record covering a period of 16 years
of Title I operations.  By contrast, Meek came to this Court on review
of a decision by a three-judge district court, upholding the program
on a sparse evidentiary record.  Because the Meek program had been in
existence for barely a year at the time of the sole evidentiary
hearing, the record contained little about its actual operation.

   In Meek, therefore, this Court was asked to rely on the "good
faith" and professionalism of the Pennsylvania teachers.  421 U.S.
369.  Here, the Court has the benefit of a record that demonstrates
that Title I personnel "have not fostered religious views or advanced
religious activities" (PEARL, J.S. App. 97a (emphasis in original)).
As the district court below stated (J.S. App. 56a-57a):

      although arguably some of the circumstances of the Title I
      program parallel the State program in Meek, the direct evidence
      demonstrates that the concerns of the Meek Court about the
      potential for the unconstitutional mingling of government and
      religion in the administration of this type of program have not
      materialized.  Undoubtedly, the Supreme Court will not ignore
      direct evidence of how Title I has functioned and operated in
      New York City's nonpublic schools for some seventeen (17) years
      in favor of plaintiffs' conjecture about the possibility of
      unconstitutional governmental activity inherent in the
      arrangements of this program.

 D. The Court Of Appeals' Reasons For Refusing To Consider The
Evidentiary Record Were Misconceived

   The court of appeals offered a litany of reasons why its
hypothetical conclusion of excessive entanglement should not "yield to
the considerable evidence concerning the actual workings of the City's
provision of on-premises remedial instruction and guidance services"
(J.S. App. 37a).  None of them can withstand analysis.

   The court first opined (J.S. App. 38a) that the failure of the
record to reflect evidence that Title I was in fact used to foster
religion merely shows that it is difficult to "prove a negative";  in
the same vein, the court later (J.S. App. 41a) asserted that to rely
on the fact that "no harm has been proved to have been done in the
past is a fundamentally wrong approach," that there is "no guarantee"
that the secular approach of the current Title I administrators will
continue, and that the plaintiffs cannot be "expected to mount
perpetual guard." These statements demonstrate, we believe, that the
court of appeals arbitrarily reallocated the burden of proof in this
case to defendants, in order to spare the plaintiffs the necessity of
proving, now or in the future, the existence of unconstitutional
practices.  But there is no warrant in law for such a reallocation of
the burden of proof.  It is particularly inappropriate here because it
leads to a presumption of the unconstitutionality of a major
congressional program.  Cf. Ashwander v. Tennessee Valley Authority,
297 U.S. 288, 354-355 (1936) (Brandeis, J., concurring);  Trustees of
Dartmouth College v. Woodward, 17 U.S. (4 Wheat.) 518, 625 (1819).
/34/ And, as already detailed, this Court has clearly indicated in
Wheeler v. Barrera, supra, that a "careful evaluation of the facts" of
the particular case is not only appropriate but necessary.  /35/

   The court further expressed the fear that "to relax" the "ban on
sending public school teachers and counselors into religious schools *
* * would let the genie out of the bottle" (J.S. App. 42a (footnote
omitted)).  Without "a principled basis on which this Court can impose
a limit," the court of appeals predicted, "(c)onsiderable segments of
the curriculum of the religious schools could be turned over to public
school teachers" (J.S. App. 42a-43a).  But there is no substantial
basis for the assertion that, if the on-premises Title I program is
upheld, there will remain no "principled basis" for imposing a limit.
The correct line, we believe, is to permit federal aid to go on an
evenhanded basis to all students as long as the provision of services
is strictly supplemental and totally in the control of public
authorities.  The vice of promoting religion is avoided so long as the
program is supplemental and evenhanded and resources that might be
diverted to religious ends are not placed in the control of religious
institutions.  And the vice of excessive entanglement is avoided so
long as there is no mixture or overlap of jurisdictions.  The
government must have no control over religious institutions or
personnel;  the religious institution must have no control over public
programs or personnel.  So long as these principles -- which lie at
the heart of this Court's Establishment Clause jurisprudence -- are
honored, there is little danger of the slippery slope.

   The court of appeals also stated (J.S. App. 43a) that "the symbolic
significance of the regular appearance of public school teachers in
religious schools" may be cause for invalidating the program,
suggesting that this may violate "the 'neutrality' test toward
religion." But the answer is that given by the Court in Everson,
Allen, Zorach, Mueller, Widmar, and like decisions:  the government
does not violate its obligation of neutrality by treating all students
equally.  The "symbolic significance" of evenhanded treatment of all
students, we believe, exemplifies rather than subverts the spirit of
the Religion Clauses.  In fact, what would be deeply troubling is the
"symbolic significance" of denying access to effective remedial
education for deprived youngsters who have exercised their
constitutional right to choose a religious education.

 E. To Deny Comparable Services To Nonpublic School Students Would Be
Inconsistent With The Religion Clauses' Principles Of Evenhandedness
And Accommodation

   The decision of parents to send their children to nonpublic schools
is, in part, an exercise of religious liberty, protected under the
Free Exercise Clause of the Constitution.  Pierce v. Society of
Sisters, 268 U.S. 510 (1925).  Their decision is entitled to respect,
to toleration, to accommodation.  Here they ask no more than that they
not be excluded from a major program otherwise open to all -- that
they be accorded the "equitable participation" in Title I that the
statute demands.  Cf. Widmar v. Vincent, supra.  Accommodation of
these children by providing remedial services in the manner that is
most effective as an educational matter is not special preference;  it
is simple justice.  It promotes "pluralism and diversity" (Lynch v.
Donnelly, slip op. 8) -- the very goal and hallmark of the Religion
Clauses -- in the field of education.

   Conversely, to deny these needed services to disadvantaged
religious school pupils is to burden their exercise of a
constitutional right.  It generates more constitutional problems than
it solves.  Cf. Sherbert v. Verner, 374 U.S. 398, 404-406 (1963).

   The Establishment Clause plays a harmonious role in the regime of
religious liberty enshrined in our Constitution.  To interpret it in
so rigid and intolerant a spirit as to interfere with, rather than
promote, that liberty would surely affront the intentions of the
framers.  It would be the ultimate constitutional irony if, in the
name of one of the Religion Clauses, educationally deprived children
were to be denied benefits which are central to many aspects of a
successful life solely because of choices that are themselves
constitutionally protected.

                              CONCLUSION

   The judgment of the court of appeals should be reversed.

   Respectfully submitted.

   Rex E. Lee

      Solicitor General

   Richard K. Willard

      Acting Assistant Attorney General

   Paul M. Bator

      Deputy Solicitor General

   Michael W. McConnell

      Assistant to the Solicitor General

   Anthony J. Steinmeyer

   Michael Jay Singer

      Attorneys

   OCTOBER 1984

   /1/ Effective July 1, 1982, Title I was superseded by Chapter 1 of
the Education Consolidation and Improvement Act of 1981, 20 U.S.C.
3801 et seq.  Chapter 1 continues to provide federal financial
assistance to meet the special educational needs of the educationally
deprived children served under Title I while, at the same time,
eliminating burdensome and unnecessary federal supervision, direction,
and control.  See 20 U.S.C. 3801.  Although Chapter 1 generally
contains fewer and less restrictive program requirements than Title I,
the provisions concerning the participation of children in private
schools are virtually identical.  Compare 20 U.S.C. 2740 (former Title
I provision) with 20 U.S.C. 3806 (current Chapter 1 provision).  See
also J.S. App. 3a n.1.  Moreover, Chapter 1, at 20 U.S.C. 3803,
expressly incorporates by reference several sections from Title I, as
follows:  20 U.S.C. 2711-2713, 2721-2722, 2761-2763, 2771-2772,
2781-2783, 2791-2792, 2841-2844, 2851-2854.  Because there are no
material differences between the two statutes with respect to the
issues in this litigation, and because the declaratory and injunctive
relief ordered by the court of appeals (see J.S. App. 54a) is not
affected by the changes made by Chapter 1, this case is not moot.
See, e.g., Schall v. Martin, No. 82-1248 (June 4, 1984), slip op. 2
n.2.  Like the court of appeals, we will continue to refer to the
program as "Title I."

   /2/ Misapplication of Title I funds by state and local educational
agencies subjects those funds to recoupment by the Secretary of
Education.  See Bell v. New Jersey, No. 81-2125 (May 31, 1983).  In
addition, a state or local authority's failure to comply with program
requirements could result in the Secretary's withholding of funds.
See 20 U.S.C. 3872.

   /3/ The conclusions of the House and Senate committees quoted in
text were predicated largely on a comprehensive, three-year study of
the effects of Title I, undertaken by the National Institute of
Education pursuant to Pub. L. No. 93-380, Sec. 821, 88 Stat. 599.  The
results of the study were corroborated by other sources and studies
canvassed by the congressional committees.

   /4/ In 1981, Congress substantially streamlined the program by
enacting the Education Consolidation and Improvement Act of 1981, Pub.
L. No. 97-35, Sec. 552-558, 95 Stat. 464-469, 20 U.S.C. 3801-3807.
Although the purpose of this Act was to eliminate needlessly
prescriptive federal requirements (see 20 U.S.C. 3801), Congress
carried forward the requirement of participation by nonpublic school
children in full force and effect.  See 20 U.S.C. 3806.

   /5/ The Title I appropriation for Fiscal Year 1984 is
$3,480,000,000.  Pub. L. No. 98-138, Sec. 209, 97 Stat. 888.

   /6/ These factors are of more than usual importance because, as
evidence in the record shows, educationally deprived children
generally have less than average motivation to participate in school
programs.  The added degree of fatigue and inconvenience decreases the
receptivity of these children who already have learning problems,
making it very unlikely that the Title I instruction would be
educationally effective.  Dane Affidavid Para. 14, DX U, Tab A-1, at
6.

   /7/ A study of the St. Louis, Missouri Title I program reached
similar results.  J.A. 68;  C.A. App. A64.

   /8/ Because of the requirement of "comparability" (see page 5,
supra), the cuts in service would affect public school children as
well as those attending private schools.

   /9/ This conclusion was based on an investigation of the Title I
program is Missouri, undertaken by the Office of Education after
receipt of a written complaint by parents of nonpublic school
students.  The Missouri nonpublic school students received remedial
instruction only outside of regular school hours.  The investigation
concluded that services were not comparable, because the conditions
under which the services were provided led to poor attendance and poor
quality services.

   Attendance was poor because:  students were "too fatigued" after a
full day of school to attend the remedial program, the remedial
classes conflicted with other activites, including after-school jobs
or family chores, parents were reluctant to have their children remain
at school until late in the day, the children often viewed the
additional instructional time as "punishment," parents were concerned
for the safety of their small children, and children were unwilling to
attend on Saturdays.  J.A. 35-36;  C.A. App. A29-A30.  Even those
students who attended did not receive comparable services because of
fatigue or lack of concentration and the difficulty of hiring a
sufficient number of qualified specialists to provide services after
the regular school day.  J.A. 36;  C.A. App. A30.  Logistical
difficulties, lack of coordination with regular teachers, and
increased costs exacerbated these problems.  J.A. 36-37;  C.A. App.
A30-A31.

   As a result, the Office of Education found the Missouri program not
"comparable," as the statute demands (Wheeler v. Barrera, 417 U.S. at
407-408, 415, 420-421), and the federal government undertook to
provide services directly to Missouri nonpublic school students under
the "by-pass" provisions of 20 U.S.C. 2740(b).

   /10/ The final evaluation reports of these programs did not
translate pupil improvements into percentile figures, but test scores
indicated significant improvements relative to the nationally
representative group.

   /11/ The appeal in PEARL was not timely filed.  See C.A. Br. for
Plaintiffs-Appellants at 3 n.*.

   /12/ In Part A, infra, we set forward the reasons why this Court
has appellate jurisdiction under 28 U.S.C. 1252.  Appellees have not
challenged the jurisdictional basis for this appeal.

   /13/ In their complaint (J.A. 12;  C.A. App. A7), appellees sought
a declaration that "insofar as Title I of the Act authorizes the
expenditure of federal funds to finance educational services in
religious schools during school hours it is unconstitutional as
violative of the Establishment Clause of the First Amendment."

   /14/ Even under the existing program, resources are too scarce to
serve all eligible children.  In the 1981-1982 school year, only
21,000 out of 40,120 eligible nonpublic school students received Title
I remedial services.  J.A. 38, 81;  C.A. App. A32, A80.

   /15/ Title I "evinces a clear intention that state constitutional
spending proscriptions not be pre-empted as a condition of accepting
federal funds." Wheeler v. Barrera, 417 U.S. at 417 (footnote
omitted).  New York authorities concluded that to permit nonpublic
school students to participate with public school students at the
public schools during the school day might violate Article XI, Section
3 of the New York Constitution.  J.S. App. 8a.  Accordingly, if the
judgment below is sustained, the most likely result would be to return
to a system of transporting nonpublic schoolchildren to public schools
for remedial instruction after school hours -- a system New York and
other school districts have concluded is a "total failure" (J.A. 63;
C.A. App. A60).

   /16/ This Court has recognized that the welfare of all residents is
enhanced by government programs that endeavor to provide for all
schoolchildren, including those attending nonpublic schools, "a
fertile educational environment" (Wolman v. Walter, 433 U.S. 229, 236
(1977), and an "ample opportunity to develop to the fullest their
intellectual capacities" (Meek v. Pittenger, 421 U.S. at 352 n.2).
See also Mueller v. Allen, No. 82-195 (June 29, 1983), slip op. 5-6;
Committee for Public Education & Religious Liberty v. Regan, 444 U.S.
646, 654 (1980).

   /17/ Thus, even if there are benefits to religion stemming from the
Title I program (and none can be found from the record), such effects
are surely "'indirect,' 'remote,' or 'incidential'" in comparison with
the successful and well-documented record of secular achievement under
this program for disadvantaged students.  See Lynch v. Donnelly, No.
82-1256 (Mar. 5, 1984), slip op. 13, quoting Committee for Public
Education & Religious Liberty v. Nyquist, 413 U.S. 756, 771 (1973).

   /18/ See also, e.g., Widmar v. Vincent, supra;  Walz v. Tax
Commission, 397 U.S. 664 (1970);  Board of Education v. Allen, supra;
Americans United for the Separation of Church & State v. Blanton, 433
F. Supp. 97 (M.D. Tenn.), aff'd, 434 U.S. 803 (1977) (statute
providing aid to all needy college students, including those in
religious colleges, does not have prohibited effect on religion).  Cf.
Sloan v. Lemon, 413 U.S. 825, 832 (1973) (state tuition grant scheme
only for benefit of parents with children in private schools found
unconstitutional for "singl(ing) out a class of its citizens for a
special economic benefit").

   /19/ Appellees contend (C.A. Br. 40-43;  Geller Affidavit, J.A.
232-233;  C.A. App. A256-A257;  see Motion to Affirm or Dismiss 17-18)
that the on-premises program has benefitted religion because the
public school professionals working in the program have been induced
by the religious atmosphere of the nonpublic schools, or by their
contacts with nonpublic school faculty, to introduce religion into
their teaching.  The simple answer to this is that there is not a
shred of evidence in the record to support this supposition, and the
district courts to consider the argument flatly rejected it on the
record.  As the PEARL court found (J.S. App. 97a (emphasis in
original)), the Title I instructors here "have not fostered religious
views or advanced religious activities." See also J.S. App. 56a-57a,
91a-92a.

   /20/ In this respect, the program at issue here is less problematic
than the payments for administration of state-controlled tests upheld
in Committee for Public Education & Religious Liberty v. Regan, supra,
or the textbooks upheld in Mueller v. Allen, supra, and Meek v.
Pittenger, 421 U.S. at 359-362.  The governing principle is that
resources that are inherently capable of diversion to use for
religious instruction must not be placed under the control of
religious school authorities.  Wolman v. Walter, 433 U.S. at 250-251.
Resources that are inherently secular may be used by nonpublic school
authorities so long as nonpublic school students are treated on an
equal basis with public school students.  Id. at 240-241;  Regan, 444
U.S. at 654-657;  Meek, 421 U.S. at 359-362.  Here, the problem of
distinguishing between inherently secular materials and materials that
can be diverted to religious use does not arise;  all Title I
resources are selected by and remain under the exclusive control and
direction of public school authorities.

   /21/ The sole exception is Committee for Public Education &
Religious Liberty v. Nyquist, supra, which involved a government aid
program designed exclusively for the benefit of parents with children
in nonpublic schools, rather than a general program for the benefit of
all schoolchildren.

   /22/ Of course, even a direct governmental subsidy to a parochial
school is not necessarily unconstitutional.  See Committee for Public
Education & Religious Liberty v. Regan, supra;  Roemer v. Board of
Public Works, 426 U.S. 736 (1976);  Tilton v. Richardson, 403 U.S. 672
(1971).  Cf. Marsh v. Chambers, supra (upholding constitutionality of
direct government payments to legislative chaplains);  Bradfield v.
Roberts, 175 U.S. 291 (1899) (upholding constitutionality of federal
aid to hospitals operated by a religious order).

   /23/ Local educational agencies are also required to maintain
administrative control of all equipment and supplies acquired with
Title I funds.  34 C.F.R. 200.74(a);  see S. Rep. 89-146, supra, at
12.  Cf. Wolman v. Walter, 433 U.S. at 250 (loan of instructional
materials and equipment to nonpublic school students invalid because,
as a practical matter, the materials would be available for use by the
religious schools).  Title I teachers are instructed to keep such
equipment and supplies, when not being used, locked in separate
storage.  The record in this case shows that the New York Title I
program fully complies with these requirements.  J.A. 55-57;  C.A.
App. A51-A53.

   /24/ It should be noted that the "supplement, not supplant"
provisions apply to public schools as well.  See 20 U.S.C. 2736(c),
3807(b);  34 C.F.R. 200.62;  see also Bell v. Kentucky, cert. granted,
No. 83-1798 (Oct. 1, 1984).  Congress intended that Title I funds only
be used to improve the education of disadvantaged children, and not be
"diverted to meeting other needs of school systems, however, pressing
those other needs may be." S. Rep. 91-634, 91st Cong., 2d Sess. 10
(1970);  see also id. at 14.

   /25/ For example, the Board decided against providing speech
improvement and library services because of the danger that these
services might constitute aid to the institutions rather than to
children.  J.A. 44;  C.A. App. A39.

   /26/ The consultations between Title I teachers and regular private
school teachers "concerning the students' needs and progress" (J.S.
App. 12a) would occur even if Title I classes were taught solely on
the premises of public schools.  Any public welfare agency that deals
with a sectarian school student may have occasion to consult with his
teachers.  Such unexceptional professional contacts between public and
private authorities have never been thought to give rise, as a matter
of law, to unconstitutional government entanglements with religion.

   /27/ Interestingly, appellees agree with us that the degree of
surveillance occasioned by the New York Title I program is minimal.
See Appellees' Motion to Affirm or Dismiss 11, 33-34, 38.  If
appellees are correct, then the program cannot be invalidated on the
basis of excessive entanglement stemming from prophylactic
surveillance.  If there is a constitutional infirmity, it must be
found in the program's effects.  But see pages 28-33, supra.

   /28/ The court of appeals also mentioned surveillance of "the
institutions in which (the teachers) work" (J.S. App. 36a).  We do not
know what the court was referring to.  The record affirmatively
demonstrates (J.A. 62;  C.A. App. A58) that the public school
authorities did not interfere with "the nonpublic schools' control
over their internal affairs, curriculum, course content, or faculty
qualifications or performance." Perhaps the court was concerned by the
prospect of Title I supervisors "'prowling the halls'" of the
nonpublic schools (J.S. App. 43a, quoting Lemon v. Kurtzman, 403 U.S.
at 650 (Brennan, J., concurring)).  However, the record does not
reflect any friction caused by the occasional visits by Title I
supervisors to Title I classrooms, and the "entanglement" resulting
from these visits would seem to be less intrusive than that occasioned
by public regulatory programs, such as state accreditation
requirements or fire and building safety codes.  The nonpublic school
has no obligation under Title I other than to supply a classroom in
specified condition and to allow access to that classroom by teachers
and supervisors;  there is accordingly no reason for a Title I
supervisor to examine anything other than that classroom.

   /29/ In Wolman v. Walter, 433 U.S. at 246-248, this Court upheld
off-premises remedial instruction by public school teachers of classes
composed entirely of sectarian school students.  There is no reason to
presume that, merely because such a class is conducted on nonpublic
school premises the government's supervision of its own employees
would become an "excessive entanglement" with religion.

   /30/ A certain degree of surveillance or entanglement is tolerable
and may be inevitable.  "The test is inescapably one of degree." Walz
v. Tax Commission, 397 U.S. at 674;  see Lynch v. Donnelly, slip op.
14.  Only entanglements that are "excessive" are proscribed.  Mueller
v. Allen, slip op. 14;  Lemon v. Kurtzman, 403 U.S. at 613.  This
corresponds to the fact that "not every law that confers an
'indirect,' 'remote,' or 'incidental' benefit upon (religion) is, for
that reason alone, constitutionally invalid." Lynch v. Donnelly, slip
op 13 (brackets in original), quoting Committee for Public Education &
Religious Liberty v. Nyquist, 413 U.S. at 771.

   /31/ The court of appeals stated (J.S. App. 4a) that it also relied
on Public Funds for Public Schools v. Marburger, 358 F. Supp. 29
(D.N.J. 1973), aff'd, 417 U.S. 961 (1974), and Wolman v. Walter,
supra.  Neither precedent supports the result below.

   Marburger, a summary affirmance, involved a program in which public
funds were provided to nonpublic schools to enable them to hire public
school personnel for remedial services they deemed appropriate.  The
Marburger program, therefore, was essentially under the control of the
religious school authorities rather than the public schools.  The
program here is plainly distinguishable.  See pages 30-32, 33, supra.

   Wolman v. Walter, supra, does not support the court of appeals'
rigid interpretation of Meek.  Indeed, the Court in Wolman upheld the
provision of diagnostic educational services by public school
personnel on the premises of nonpublic schools, as well as the
provision of other remedial educational services to nonpublic school
students at ideologically neutral sites off the premises.  The Wolman
decision thus supports our view that questions of this sort must be
decided on the basis of a flexible and careful analysis of the
practical realities of the situation, rather than on inflexible per se
rules.  See pages 40-41, infra.

   /32/ If this Court were to conclude that Meek cannot be
distinguished from this case, we believe that it would be appropriate
for the Court to consider whether Meek, as so interpreted, should be
overruled.

   /33/ The three-judge court in PEARL found that the evidence showed
no political divisiveness over the inclusion of nonpublic school
students in the New York Title I program.  J.S. App. 101a-102a.  Nor
has there been such divisiveness at the federal level.  Equitable
participation by nonpublic school students was an intrinsic
erequirement of the program at its inception in 1965 (see S. Rep.
89-146, supra, at 11-12).  Since that time, Congress has twice amended
Title I to strengthen the nonpublic school participation requirement.
See pages 5-6, supra.  There has never been any serious effort in
Congress to depart from this aspect of the program.  It can be
expected that a forced exclusion of nonpublic school students from
equitable participation in the Title I program would engender far more
controversy than has their inclusion.

   If the potential for "political divisiveness" is considered to be a
"warning signal" for entanglement problems (see Nyquist, 413 U.S. at
797-798), then the rare and striking unanimity in the Congress that
nonpublic school students should be permitted to participate in the
Title I program on a comparable basis to public school students might
logically serve as a "reassuring signal" that this program steers
clear of the "evil(s) addressed by the Establishment Clause" (Lynch v.
Donnelly, slip op. 3 (O'Connor, J., concurring)).

   /34/ The court of appeals also objected that appellants' arguments
based on the record "too often invoke 'mosts' and averages" (J.S. App.
40a), pointing particularly to the figures (see page 11, supra) with
respect to the religious affiliations of Title I teachers and the
amount of time they spend in any particular religious school.  But the
fact that these figures fall short of perfection simply raises the
question whether the district court's findings of fact were clearly
erroneous;  they do not suggest that the evidence is to be wholly
disregarded.  Indeed, the most important circumstance shown by all
these figures is that religion is wholly extraneous and irrelevant to
the administration and operations of the Title I program.  The fact
that, on a random basis, occasionally a public school teacher who
happens to be Catholic will end up teaching Title I remedial math in a
Catholic school does not show that religion has been "established," in
the absence of at least some modicum of evidence that remedial math
has in fact been turned to a religious use.  The legal question is not
whether the record absolutely precludes entanglement, but whether it
shows with "sufficient clarity" that the program serves "legitimate
secular ends" without "appreciable risk of being used to transmit or
teach religious values." Committee for Public Education & Religious
Liberty v. Regan, 444 U.S. at 662.

   /35/ See also Board of Education v. Allen, 392 U.S. at 248
(emphasis added):

      Nothing in this record supports the proposition that all
      textbooks * * * are used by the parochial schools to teach
      religion.  No evidence has been offered about particular
      schools, particular courses, particular teachers, or particular
      books.  We are unable to hold, based solely on judicial notice,
      that this statute results in unconstitutional involvement of the
      State with religious instruction or that (the statute), for this
      or the other reasons urged, is a law respecting the
      establishment of religion within the meaning of the First
      Amendment.