Chaplaincy of Full Gospel Churches v. United States Navy, 407 U.S. App. D.C. 436

Background

Chaplains (non-liturgical Christians) and two endorsing agencies sue the Navy Chaplaincy alleging facially neutral promotion practices discriminate against non-liturgical denominations and violate equal protection and the Establishment Clause.
District court denied a preliminary injunction; this court previously vacated and remanded for clarity on likelihood of success on the merits.
The Navy uses seven-member chaplain selection boards with two chaplains and secret balloting; chairs may preside as board president, and votes are confidential.
Plaintiffs seek to enjoin (1) staffing boards with two chaplains, (2) secret ballots via sleeves, and (3) the Chief of Chaplains or deputy as board president, arguing these practices advantage liturgical denominations.
District court found no substantial evidence of discriminatory intent or discriminatory outcomes supporting likelihood of success; this court reviews de novo on questions of law.
This court holds the challenged policies are facially neutral and plaintiffs failed to show discriminatory intent or a lack of rational basis; no likelihood of success on the merits.

Issues

Issue	Plaintiff's Argument	Defendant's Argument	Held
Equal protection: intent vs. neutral rules	Chaplains claim policies discriminate in effect.	Policies are facially neutral and applied neutrally; no intentional discrimination shown.	No likelihood of success on equal protection; no discriminatory intent or irrational basis shown.
Establishment Clause: endorsement through statistics	Statistical data show denominational preference, implying endorsement.	Statistics do not demonstrate government endorsement; policies are facially neutral.	No likelihood of success under endorsement theory; no reasonable observer would perceive endorsement.
Use of statistics to prove discrimination	Statistical disparity supports discriminatory impact.	Statistical significance without controls is inconclusive; correlation not causation.	Statistical data fail to control for confounders; do not demonstrate likelihood of success on the merits.
Lemon test applicability to challenged policies	Lemon Endorsement approach should apply (strict scrutiny under Larson).	Policies are facially neutral; Larson not triggered; Lemon test applied only if necessary.	Chaplains fail to show violations under Lemon/endorsement framework; no likelihood of success.

Key Cases Cited

Village of Arlington Heights v. Metropolitan Housing Development Corp., 429 U.S. 252 (U.S. Supreme Court 1977) (impact alone not determinative; must show discriminatory intent)
Gomillion v. Lightfoot, 364 U.S. 339 (U.S. Supreme Court 1960) (requires stark discriminatory impact to infer intent)
Yick Wo v. Hopkins, 118 U.S. 356 (U.S. Supreme Court 1886) (history of discriminatory application invalidates neutral rules)
Larson v. Valente, 456 U.S. 228 (U.S. Supreme Court 1982) (if denominational preference claimed, facial differentiation triggers scrutiny)
Lemon v. Kurtzman, 403 U.S. 602 (U.S. Supreme Court 1971) (three-part test for Establishment Clause scrutiny)
Zelman v. Simmons-Harris, 536 U.S. 639 (U.S. Supreme Court 2002) (endorsement analysis under reasonable observer concept)
Bonham v. D.C. Library Admin., 989 F.2d 1242 (D.C. Cir. 1993) (no de minimis exception to Establishment Clause scrutiny)
Berger v. Iron Workers Reinforced Rodmen Local 201, 843 F.2d 1395 (D.C. Cir. 1988) (statistical disparities in Title VII contexts require controls for confounders)
Segar v. Smith, 738 F.2d 1249 (D.C. Cir. 1984) (statistical proof in discrimination cases requires confounding control)
Washington v. Davis, 426 U.S. 229 (U.S. Supreme Court 1976) (impact alone not enough without discriminatory intent or non-rational basis)
Chaplaincy of Full Gospel Churches v. England, 454 F.3d 290 (D.C. Cir. 2006) (establishment analysis for chaplaincy context in circuit court)
Katcoff v. Marsh, 755 F.2d 223 (2d Cir. 1985) (courts have treated chaplaincy program compatibility with Establishment Clause)