Opinions of the United 2004 Decisions States Court of Appeals for the Third Circuit 11-29-2004 Forum Academic v. Secretary Defense Precedential or Non-Precedential: Precedential Docket No. 03-4433P Follow this and additional works at: http://digitalcommons.law.villanova.edu/thirdcircuit_2004 Recommended Citation "Forum Academic v. Secretary Defense" (2004). 2004 Decisions. Paper 104. http://digitalcommons.law.villanova.edu/thirdcircuit_2004/104 This decision is brought to you for free and open access by the Opinions of the United States Court of Appeals for the Third Circuit at Villanova University School of Law Digital Repository. It has been accepted for inclusion in 2004 Decisions by an authorized administrator of Villanova University School of Law Digital Repository. For more information, please contact Benjamin.Carlson@law.villanova.edu. PRECEDENTIAL UNITED STATES COURT OF APPEALS FOR THE THIRD CIRCUIT No. 03-4433 FORUM FOR ACADEMIC AND INSTITUTIONAL RIGHTS, a New Jersey membership corporation; SOCIETY OF AM ERICAN LAW TEACHERS, INC., a New York corporation; COALITION FOR EQUALITY, a Massachusetts association; RUTGERS GAY AND LESBIAN CAUCUS, a New Jersey association; PAM NICKISHER, a New Jersey resident; LESLIE FISCHER, a Pennsylvania resident; MICHAEL BLAUSCHILD, a New Jersey resident; ERW IN CHEMERINSKY, a California resident; SYLVIA LAW , a New York resident, Appellants v. DONALD H. RUMSFELD, in his capacity as U.S. Secretary of Defense; ROD PAIGE, in his capacity as U. S. Secretary of Education; ELAINE CHAO, in her capacity as U.S. Secretary of Labor; TOMMY THOMPSON, in his capacity as U.S. Secretary of Health and Human Services; NORMAN Y. MINETA, in his capacity as U.S. Secretary of Transportation; TOM RIDGE, in his capacity as U.S. Secretary of Homeland Security Appeal from the United States District Court for the District of New Jersey (D.C. Civil Action No. 03-cv-04433) District Judge: Honorable John C. Lifland Argued June 30, 2004 Before: AMBRO, ALDISERT and STAPLETON, Circuit Judges (Opinion filed November 29, 2004 ) E. Joshua Rosenkranz, Esquire (Argued) Timothy P. Wei, Esquire Sharon E. Frase, Esquire Heller, Ehrman, White & McAuliffe LLP 120 West 45th Street, 20th Floor New York, NY 10036-4041 Warrington S. Parker, III, Esquire Aaron M. Armstrong, Esquire Benjamin D. Hauser, Esquire Heller, Ehrman, White & McAuliffe LLP 333 Bush Street San Francisco, CA 94104-2878 Attorneys for Appellants 2 Peter D. Keisler Assistant Attorney General Christopher J. Christie United State Attorney Gregory G. Katsas (Argued) Deputy Assistant Attorney General Douglas N. Letter, Esquire Scott R. McIntosh, Esquire United States Department of Justice Civil Division, Appellate Staff 601 D Street, N.W., Room 9550 Washington, DC 20530 George S. Leone, Esquire Office of the United States Attorney 970 Broad Street, Room 700 Newark, NJ 07102 Attorneys for Appellees Paul M. Smith, Esquire (Argued) William M . Hohengarten, Esquire Daniel Mach, Esquire Jenner & Block, Suite 1200 South 601 13th Street, N.W., 12th Floor Washington, DC 20005 Philip G. Gallagher, Esquire Lawrence S. Lustberg, Esquire Jonathan L. Hafetz, Esquire Gibbons, Del Deo, Dolan, Griffinger & Vecchione 3 One Pennsylvania Plaza, 37th Floor New York, NY 10119 Stuart D. Rosen, Esquire Bingham McCutchen One State Street Hartford, CT 06103 Jonathan A. Kenter, Esquire Bingham McCutchen LLP 399 Park Avenue New York, NY 10022 Tyler M. Paetkau, Esquire Melissa J. Goldberg, Esquire Bingham McCutchen LLP Three Embarcadero Center San Francisco, CA 94111 Walter E. Dellinger, III, Esquire (Argued) Pamela Harris, Esquire O’Melveny & M yers 1625 Eye Street, N.W. Washington, DC 20006 Hilary E. Ball, Esquire Sam Heldman, Esquire Gardner, Middlebrooks, Gibbons, Kittrell & Olsen 2805 31st Street, N.W. Washington, DC 20008 4 David M. Rabban, Esquire University of Texas School of Law 727 East Dean Keeton Street Austin, TX 78705 Ann D. Springer, Esquire Donna R. Euben, Esquire American Association of University Professors 1012 Fourteenth Street, N.W., Suite 500 Washington, DC 20005 John L. Moore, Jr., Esquire Louis J. Rouleau, Esquire Piper Rudnick LLP 1200 Nineteenth Street, N.W. Washington, DC 20036 E. O’Brien Kelley, Esquire Darren G. Gibson, Esquire Piper Rudnick LLP 1251 Avenue of the Americas New York, NY 10020 Attorneys for Amicus-Appellants Howard J. Bashman, Esquire (Argued) 1250 Virginia Drive Suite 1000 Fort Washington, PA 19034 5 Steven W . Fitschen, Esquire The National Legal Foundation 2224 Virginia Beach Boulevard Suite 204 Virginia Beach, VA 23454 Attorneys for Amicus-Appellees OPINION OF THE COURT Ambro, Circuit Judge The Solomon Amendment, 10 U.S.C. § 983, requires the United States Department of Defense (“DOD”) to deny federal funding to institutions of higher education that prohibit military representatives access to and assistance for recruiting purposes. Last fall, the Forum for Academic and Institutional Rights, Inc. (“FAIR”), 1 an association of law 1 Joining FAIR in its preliminary injunction motion and in this appeal are: the Society for Law Teachers, Inc.; the Coalition for Equality; Rutgers Gay and Lesbian Caucus; law professors Erwin Chemerinsky and Sylvia Law; and law students Pam Nickisher, Leslie Fischer, Ph.D., and Michael Blauschild. For convenience, we refer to all plaintiff-appellants collectively as “FAIR.” 6 schools and law faculty, asked the United States District Court for the District of New Jersey to enjoin enforcement of the Solomon Amendment. The District Court denied FAIR’s motion. Forum for Academic & Institutional Rights, Inc. v. Rumsfeld, 291 F. Supp. 2d 269 (D.N.J. 2003) (“FAIR”). On appeal, we hold that FAIR has demonstrated a likelihood of success on the merits of its First Amendment claims and that it is entitled to preliminary injunctive relief. Accordingly, we reverse. I. Background Facts 2 and Procedural Posture A. Law Schools’ Nondiscrimination Policies Law schools have long maintained formal policies of nondiscrimination that withhold career placement services from employers who exclude employees and applicants based on such factors as race, gender, and religion. In the 1970s law schools began expanding these policies to prohibit discrimination based on sexual orientation as well. In response to this trend the American Association of Law Schools (“AALS”) voted unanimously in 1990 to include sexual orientation as a protected category. As a result, 2 The facts on appeal are not in dispute. As the District Court noted, the Government did not challenge or supplement the factual assertions presented by FAIR in its motion for injunctive relief. FAIR, 291 F. Supp. 2d at 277. 7 virtually every law school now has a comprehensive policy like the following: [The] School of Law is committed to a policy of equal opportunity for all students and graduates. The Career Services facilities of this school shall not be available to those employers who discriminate on the grounds of race, color, religion, national origin, sex, handicap or disability, age, or sexual orientation . . . . Before using any of the Career Services interviewing facilities of this school, an employer shall be required to submit a signed statement certifying that its practices conform to this policy. B. Congress Passes the Solomon Amendment The United States military excludes servicemembers based on evidence of homosexual conduct and/or orientation. See 10 U.S.C. § 654.3 Citing their nondiscrimination policies, 3 While the current statutory version of the military’s exclusionary policy has existed since 1993, National Defense Authorization Act for Fiscal Year 1994, Pub. L. No. 103-160, § 571(a)(1), 107 Stat. 1547, 1670 (Nov. 30, 1993), the military has had formal regulatory policies excluding gays and lesbians since World War I and a practice of such exclusion since the 8 some law schools began in the 1980s refusing to provide access and assistance to military recruiters. This caught the attention of members of Congress. In 1994, Representative Gerald Solomon of New York sponsored an amendment to the annual defense appropriation bill that proposed to withhold DOD funding from any educational institution with a policy of denying or effectively preventing the military from obtaining entry to campuses (or access to students on campuses) for recruiting purposes. National Defense Authorization Act for Fiscal Year 1995, Pub. L. No. 103-337 Revolutionary War. See, e.g., Articles of War of 1916, Pub. L. No. 242, art. 93, 39 Stat. 619, 664 (assault with intent to commit sodomy punishable by court martial); see generally Randy Shilts, Conduct Unbecoming: Gays & Lesbians in the U.S. Military 11–17 (1994). Under the current statute, a servicemember is separated from the military if it is found that he or she “engaged in . . . a homosexual act” or “stated that he or she is a homosexual” or “married or attempted to marry a person known to be of the same biological sex.” 10 U.S.C. § 654(b). It defines “homosexual” and “homosexual act” to include evidence demonstrating “a propensity or intent to engage in homosexual acts.” Id. It also allows servicemembers to rebut findings of proscribed conduct with evidence of the lack of a propensity to engage in homosexual conduct, i.e., evidence of a heterosexual orientation. Id. Law schools interpret the ban as conflicting with their policies against discrimination on the basis of sexual orientation. 9 § 558, 108 Stat. 2663, 2776 (1994). During debate in the House of Representatives, Representative Solomon urged the passage of his amendment “on behalf of military preparedness” because “recruiting is the key to an all-volunteer military.” 140 Cong. Rec. H3861 (daily ed. May 23, 1994). He argued that it was hypocritical for schools to receive federal money while at the same time denying the military access to their campuses: “[T]ell[] recipients of Federal money at colleges and universities that if you do not like the Armed Forces, if you do not like its policies, that is fine. That is your [F]irst [A]mendment right[]. But do not expect Federal dollars to support your interference with our military recruiters.” Id. The amendment’s co-sponsor, Representative Richard Pombo of California, said Congress needed to target “policies of ambivalence or hostility to our Nation’s armed services” that are “nothing less than a backhanded slap at the honor and dignity of service in our Nation’s Armed Forces.” Id. at H3863. He urged his colleagues to “send a message over the wall of the ivory tower of higher education” that colleges’ and universities’ “starry-eyed idealism comes with a price. If they are too good—or too righteous—to treat our Nation’s military with the respect it deserves[,] then they may also be too good to receive the generous level of taxpayer dollars presently enjoyed by many institutions of higher education in America.” Id. 10 Other Representatives opposed the amendment, alleging violations of academic freedom and civil rights. See, e.g., id. at H3862 (Rep. Dellums) (“We should not . . . chill or abridge privacy, speech, or conscience by threatening a college with a Federal funds termination because it chose for whatever reason to deny access to military recruiters . . . . We should not browbeat them . . . into becoming involuntary agents of Federal policy.”). In light of Vietnam War-era legislation, rarely invoked, that already granted the DOD discretion to withhold funding from colleges and universities that barred military recruiters, see Pub. L. No. 92-436, § 606, 86 Stat. 734, 740 (1972), the DOD itself objected to the proposed amendment as “unnecessary” and “duplicative.” 140 Cong. Rec. H3864 (Rep. Schroeder) (explaining the DOD’s position). The DOD also feared that withholding funds from universities could be potentially harmful to defense research initiatives. Id. But the House voted for the amendment by a vote of 271 to 126. Id. at H3865. Several months later the Senate approved the defense spending appropriations bill, including Representative Solomon’s amendment, and the “Solomon Amendment” ultimately became law. C. Subsequent Amendments and Regulatory Interpretations In 1997 Congress amended the Solomon Amendment by expanding its penalty to include, in addition to DOD funds, funds administered by other federal agencies, including the 11 Departments of Transportation,4 Labor, Health and Human Services, and Education.5 Omnibus Consolidated Appropriations Act, 1997, Pub. L. No. 104–208, § 514(b), 110 Stat. 3009–270 (1996). This amendment was recodified in another amendment in 1999. National Defense Authorization Act for Fiscal Year 2000, Pub. L. No. 106–65, § 549, 113 Stat. 512, 609–11 (1999). DOD regulations have clarified this expansion, penalizing an offending “subelement” of a college or university (i.e., a law school) that prohibits or effectively prevents military recruiting with the loss of federal funding from all of the federal agencies identified in the statute, while withholding from the offending subelement’s parent institution only DOD funds. 32 C.F.R. § 216.3(b)(1). The 1999 amendment also codified exceptions to the Solomon Amendment’s penalties for schools that (1) have ceased an offending policy or practice, or (2) have a longstanding religious-based policy of pacifism. § 549, 113 4 Department of Homeland Security funds later replaced Department of Transportation funds. Pub. L. No. 107-296, § 1704(b)(1), 116 Stat. 2314 (2002). 5 A separate amendment cancelled the application of the Solomon Amendment to direct student aid. Department of Defense Appropriations Act of 2000, § 8120, Pub. L. No. 106- 79, 113 Stat. 1212, 1260 (1999). 12 Stat. at 610(c) (codified at 10 U.S.C. § 983(c)). DOD regulations subsequently added a third exception for schools that provide military recruiters a degree of access equal to that provided to other recruiters. 32 C.F.R. § 216.4(c). Following the 1999 amendment, the DOD enforced the Solomon Amendment consistent with its terms. Only schools whose policies or practices “prohibit[ed], or in effect prevent[ed],” military representatives “from gaining entry to campuses, or access to students . . . on campuses for purposes of military recruiting,” were penalized. Thus, by merely allowing military recruiters to gain access to campuses, many law schools avoided the Solomon Amendment’s penalty while reaffirming their opposition to the military’s exclusionary employment policy by not providing them affirmative assistance in the manner provided to other recruiters. Harvard Law School, for example, allowed military recruiters on campus to recruit at the offices of its Veterans Association but did not volunteer its placement personnel to arrange interviews. Boston College Law School allowed military recruiters to conduct on-campus interviews, but kept their literature in the library rather than in the career services office. Until the fall of 2001, the DOD did not consider these and other similar “ameliorative measures” to violate the Solomon Amendment and expressed enthusiasm for the law schools’ cooperation with what it described as successful recruiting efforts. See FAIR, 291 F. Supp. 2d at 282 (citing record evidence). 13 But following the terrorist attacks in the United States in September 2001, the DOD began applying an informal policy of requiring not only access to campuses, but treatment equal to that accorded other recruiters. As evidence of this informal policy, a letter from the DOD’s Acting Deputy Undersecretary William J. Carr to Richard Levin, the President of Yale University, stated that universities are required “to provide military recruiters access to students equal in quality and scope to that provided to other recruiters.” 6 The same letter stated that the “DOD requires that there not be a substantial disparity in the treatment of military recruiters as compared to other potential employers.” This changed context meant that Yale’s willingness to let military recruiters use a room in Yale Law School’s building for interviews would not pass muster unless it also provided military recruiters with the same level of assistance from its career development office (arranging interviews, posting notices, etc.) provided to other recruiters. Furthermore, the DOD intimated that failure to comply would result in a loss to 6 In wording the new informal policy’s substantive requirement, the DOD borrowed language from the existing policy’s regulatory exception—32 C.F.R. 216.4(c) (exempting from Solomon Act compliance a law school that “presents evidence that the degree of access by military recruiters is at least equal in quality and scope to that afforded to other employers”). 14 Yale University not only of DOD funds, but of all federal funds (a penalty that is not consistent with the DOD’s existing regulations, under which the offending subelement’s parent institution is penalized with the loss of only DOD funds, see 32 C.F.R. § 216.3(b)(1)). In another example, the DOD advised the University of Southern California Law School in 2002 that its past practice of accommodating military recruiters—providing them with standard employer information, referring them to the campus ROTC office for scheduling of interview office space, posting notices in the weekly newsletter for students, and making military recruitment materials available to students—would violate the Solomon Amendment unless its career services office invited military recruiters to participate in an off- campus job fair open to other employers. According to the DOD, anything less than equal treatment for military recruiters “sends the message that employment in the Armed Forces is less honorable or desirable than employment with other organizations”—a dangerous message to be sending “in today’s military climate.” In light of the millions of dollars at stake, every law school that receives federal funds had, by the 2003 recruiting season, suspended its nondiscrimination policy as applied to military recruiters. This past summer Congress amended the Solomon Amendment to codify the DOD’s informal policy. Ronald W. Reagan National Defense Authorization Act for Fiscal Year 15 2005, Pub. L. No. 108–375, § 552, 118 Stat. 1811, 1911 (2004). Now, under the terms of the statute itself, law schools and their parent institutions are penalized for preventing military representatives from gaining entry to campuses for the purpose of military recruiting “in a manner that is at least equal in quality and scope to the [degree of] access to campuses and to students that is provided to any other employer.” 10 U.S.C. § 983(b). D. Current Litigation In September 2003, FAIR sued the DOD and the other federal departments whose funds are restricted under the Solomon Amendment, seeking on constitutional grounds a preliminary injunction enjoining enforcement of the statute and the then-existing (now codified) informal policy. The Government defendants moved to dismiss for lack of standing. The District Court denied both the motion to dismiss and FAIR’s motion for preliminary injunction. See FAIR, 291 F. Supp. 2d at 296, 322. This appeal followed. II. Jurisdiction Under 28 U.S.C. § 1331, a federal district court has original subject matter jurisdiction over an action for injunctive relief based on constitutional claims. Tenafly Eruv Ass’n v. Borough of Tenafly, 309 F.3d 144, 156 n.12 (3d Cir. 16 2002), cert. denied, 539 U.S. 942 (2003).7 Our appellate jurisdiction exists under 28 U.S.C. § 1292(a)(1). III. Analysis To obtain a preliminary injunction FAIR must establish (1) a reasonable likelihood of success on the merits, (2) irreparable harm absent the injunction, (3) that the harm to FAIR absent the injunction outweighs the harm to the Government of granting it, and (4) that the injunction serves 7 Standing must also be proper for subject matter jurisdiction to exist. See, e.g., Storino v. Borough of Point Pleasant Beach, 322 F.3d 293, 296 (3d Cir. 2003); Charles Alan Wright & Arthur R. Miller et al., Federal Practice & Procedure § 3531 (2d ed. 1984). The District Court held that FAIR had standing to seek a preliminary injunction against the Solomon Amendment, and the Government has conceded this issue on appeal. Acknowledging our continuing obligation to verify subject matter jurisdiction when it is in question, see, e.g., Exxon Mobil Corp. v. Saudi Basic Indus. Corp., 364 F.3d 102, 104 (3d Cir.), cert. granted on other grounds, No. 03-1696, 125 S. Ct. 310 (2004), we affirm the District Court’s holding that FAIR’s standing was proper for the reasons it provided. FAIR, 291 F. Supp. 2d at 285–91. While the Government does not concede that the non- FAIR plaintiffs had standing, the presence of one plaintiff with standing is sufficient to satisfy that requirement. Bowsher v. Synar, 478 U.S. 714, 721 (1986). 17 the public interest. Tenafly Eruv Ass’n, 309 F.3d at 157. While we review a district court’s balancing of the preliminary injunction factors for abuse of discretion, we review “any determination that is a prerequisite to the issuance of an injunction . . . according to the standard applicable to that particular determination.” Id. at 156 (citations omitted). Thus, because the District Court’s ruling was based on its application of the First Amendment and other constitutional principles to the Solomon Amendment—issues of law to which a plenary standard of review applies— our review is plenary. Id. A. Unconstitutional Conditions Doctrine FAIR argues that the Solomon Amendment is an unconstitutional condition.8 Under the unconstitutional 8 Our dissenting colleague urges us to begin our analysis with the presumption that congressional statutes are constitutional. It is a fundamental canon of statutory construction that, when there are “‘two possible interpretations of a statute, by one of which it would unconstitutional and by the other valid, our plain duty is to adopt that which will save the Act.’” Rust v. Sullivan, 500 U.S. 173, 190 (1991) (quoting Blodgett v. Holden, 275 U.S. 142, 148 (1927)). But in this case it is not argued that there are two possible constructions of the Solomon Amendment. The canons of statutory construction therefore do not apply. Moreover, “although a duly enacted statute normally carries with it a presumption of constitutionality, when a [statute] 18 conditions doctrine, the Government “may not deny a benefit to a person on a basis that infringes his constitutionally protected interests—especially, his interest in freedom of speech.” Perry v. Sindermann, 408 U.S. 593, 597 (1972). If Congress “could deny a benefit to a person because of his constitutionally protected speech or associations, his exercise of those freedoms would in effect be penalized and inhibited.” Id. Put another way, the Government may not propose a penalty to “produce a result which [it] could not command directly.” Speiser v. Randall, 357 U.S. 513, 526 (1958) (state could not condition property tax exemption on loyalty oath); see also Rosenberger v. Rectors & Visitors of the Univ. of Va., 515 U.S. 819 (1995) (public university could not condition funds for student publications on their secular perspective); FCC v. League of Women Voters, 468 U.S. 364 (1984) (FCC could not condition federal funds to radio stations on editorial content). Thus, if the law schools’ compliance with the Solomon Amendment compromises their First Amendment rights, the statute is an unconstitutional condition.9 allegedly infringes on the exercise of [F]irst [A]mendment rights, the statute’s proponent bears the burden of establishing [its] constitutionality.” ACORN v. City of Frontenac, 714 F.2d 813, 817 (8th Cir. 1983) (citing Org. for a Better Austin v. Keefe, 402 U.S. 415, 419 (1971)). 9 As the District Court noted, the Supreme Court’s exception to the unconstitutional conditions doctrine for selective spending programs does not apply here. FAIR, 291 F. Supp. 2d at 19 B. First Amendment Analysis The First Amendment provides that “Congress shall make no law . . . abridging the freedom of speech.” U.S. Const. amend. I. This simple commandment plays out differently depending on the avenue of analysis. Two avenues applicable here are: (1) whether the law schools are “expressive associations” whose First Amendment right to disseminate their chosen message is impaired by the inclusion of military recruiters on their campuses; and (2) whether the law schools are insulated by free speech protections from being compelled to assist military recruiters in the expressive 299–300. When the Government appropriates for a particular spending program, it may endorse one viewpoint over another by conditioning its spending on certain criteria. United States v. Am. Library Ass’n, 539 U.S. 194, 211 (2003) (providing library assistance funds to only those libraries who agree to block obscene Internet sites); Rust, 500 U.S. at 192–93 (funding family planning services that eschew abortion counseling). In those cases, “the Government [was] not denying a benefit to anyone, but [was] instead simply insisting that public funds be spent for the purposes for which they were authorized.” Rust, 500 U.S. at 196; see also Am. Library Ass’n, 539 U.S. at 211. That exception does not apply in our case because the Solomon Amendment does not create a spending program; it merely imposes a penalty—the loss of general funds. 20 act of recruiting.10 A violation of freedom of speech under either analytical approach draws down the curtain on Solomon Amendment enforcement unless the Government can establish that the statute withstands strict scrutiny. The levels of scrutiny applicable in the First Amendment context are crucial. A regulation that disrupts an expressive association or compels speech must be narrowly tailored to serve a compelling governmental interest, and must use the least restrictive means of promoting the Government’s asserted interest (here, recruiting talented lawyers). See infra Parts 10 FAIR also argues that the Solomon Amendment and the then-existing informal policy are void under the First Amendment’s vagueness doctrine because they provide insufficient notice as to what activities will trigger funding penalties. But the statutory amendment enacted during FAIR’s pending appeal, see supra Part I.C, has rendered moot both the challenge to the Solomon Amendment, see Black United Fund of N.J., Inc. v. Kean, 763 F.2d 156, 160 (3d Cir. 1985), and the challenge to the regulatory policy, see Prometheus Radio Project, Inc. v. FCC, 373 F.3d 372, 396 (3d Cir. 2004). The recent amendment to the Solomon Amendment does not, however, moot FAIR’s other challenges to it. See Northeastern Fla. Chapter of the Associated Gen. Contractors of Am. v. City of Jacksonville, 508 U.S. 656, 662 (1993) (stating that a challenge to a statute is not moot when the new version of it “disadvantages [appellants] in the same fundamental way”). 21 III.B.1(c), 2(e). Needless to say, this is an imposing barrier. The District Court, by contrast, emphasized a third potential theory of this case that invokes only intermediate scrutiny, i.e., whether the government action at issue furthers an important government interest that would be achieved less effectively without that action. The Court asked whether the law schools’ resistance to the Solomon Amendment is sufficiently communicative to bring it within the ambit of the First Amendment’s protection for “expressive conduct,” the suppression of which receives intermediate scrutiny under United States v. O’Brien, 391 U.S. 367 (1968). See infra Part III.B.3(b). We emphasize at the outset that we need not decide this issue because we conclude that the Solomon Amendment violates the First Amendment by impeding the law schools’ rights of expressive association and by compelling them to assist in the expressive act of recruiting. Nonetheless, we explain briefly our conclusion that FAIR would prevail even under O’Brien’s less strict framework. 1. Expressive Association FAIR argues that the Solomon Amendment impairs law schools’ First Amendment rights under the doctrine of expressive association. The Supreme Court most recently addressed this doctrine in Boy Scouts of America v. Dale, 530 U.S. 640 (2000). There the Court held that a state public accommodations law that prohibited discrimination based on 22 sexual orientation could not constitutionally be invoked to force the Boy Scouts to accept openly gay James Dale as an assistant scoutmaster. Id. at 659. Central to its analysis was the deference it gave to the Boy Scouts’ “view of what would impair its expression,” which compelled the Court’s conclusion that Dale’s presence would “significantly burden the Boy Scouts’ desire to not ‘promote homosexual conduct as a legitimate form of behavior.’” Id. at 653 (citation omitted). Under Dale, the elements of an expressive association claim are (1) whether the group is an “expressive association,” (2) whether the state action at issue significantly affects the group’s ability to advocate its viewpoint, and (3) whether the state’s interest justifies the burden it imposes on the group’s expressive association. Id. at 648–58; accord The Circle School v. Pappert, 381 F.3d 172, 181-82 (3d Cir. 2004) (applying the Dale framework); Pi Lambda Phi Fraternity, Inc. v. Univ. of Pittsburgh, 229 F.3d 435, 442 (3d Cir. 2000) (same). We apply each in turn to analyze FAIR’s expressive association claim. (a) The law schools are expressive associations. A group that engages in some form of public or private expression above a de minimis threshold is an “expressive association.” Pi Lambda Phi, 229 F.3d at 443. The group 23 need not be an advocacy group or exist primarily for the purpose of expression. Dale, 530 U.S. at 648. The Supreme Court held that the Boy Scouts, which “seeks to transmit . . . a system of values, engages in expressive activity.” Id. at 650. “By nature, educational institutions are highly expressive organizations, as their philosophy and values are directly inculcated in their students.” The Circle School, 381 F.3d at 182. Because FAIR has shown that the law schools “possess[] clear educational philosophies, missions and goals,” id., we agree with the District Court’s conclusion that they qualify as expressive associations. FAIR, 291 F. Supp. 2d at 303–04. Therefore, FAIR satisfies the first element of the Dale analysis. (b) The Solomon Amendment significantly affects the law schools’ ability to express their viewpoint. FAIR argues that the Solomon Amendment significantly affects law schools’ ability to express their viewpoint, reflected in their policies, that discrimination on the basis of sexual orientation is wrong. The Solomon Amendment compels them, they contend, to disseminate the opposite message. The schools believe that, by coordinating interviews and posting and publishing recruiting notices of an employer who discriminates on the basis of sexual orientation, they impair their ability to teach an inclusive message by 24 example. Put another way, FAIR maintains that the Solomon Amendment suppresses the law schools’ chosen speech by interfering with their prerogative to shape the way they educate (including, of course, the manner in which they communicate their message). In Dale, the Supreme Court recognized that “[t]he forced inclusion of an unwanted person in a group” could significantly affect the group’s ability to advocate its public or private viewpoint. 530 U.S. at 648. The viewpoint at issue in Dale was the Boy Scouts’ long-held belief that “homosexual conduct is inconsistent with . . . the Scout Oath” and that “homosexuals [do not] provide a role model consistent with the[] expectations [of Scouting families].” Id. at 652. Because the Boy Scouts’ expressive purpose was to “inculcate [youth] with the Boy Scouts’ values—both expressively and by example,” id. at 649–50, the organization believed that the presence of an openly gay assistant scoutmaster could be perceived as “promot[ing] homosexual conduct as a legitimate form of behavior,” a message inconsistent with the expression it wished to convey and the example it wished to set. Id. at 651. The Supreme Court agreed. Because James Dale was openly gay, his “presence in the Boy Scouts would, at the very least, force the organization to send a message, both to youth members and the world, that the Boy Scouts accepts homosexual conduct as a legitimate form of behavior.” Id. at 25 653. Just as the Boy Scouts believed that “homosexual conduct is inconsistent with the Scout Oath,” id. at 652, the law schools believe that employment discrimination is inconsistent with their commitment to justice and fairness. Just as the Boy Scouts maintained that “homosexuals do not provide a role model consistent with the expectations of Scouting families,” id., the law schools maintain that military recruiters engaging in exclusionary hiring “do not provide a role model consistent with the expectations of,” id., their students and the legal community. Just as the Boy Scouts endeavored to “inculcate [youth] with the Boy Scouts’ values—both expressively and by example,” id. at 649–50, the law schools endeavor to “inculcate” their students with their chosen values by expression and example in the promulgation and enforcement of their nondiscrimination policies. FAIR Br. at 22–25. And just as “Dale’s presence in the Boy Scouts would, at the very least, force the organization to send a message, both to youth members and the world, that the Boy Scouts accepts homosexual conduct as a legitimate form of behavior,” Dale, 530 U.S. at 653, the presence of military recruiters “would, at the very least, force the law schools to send a message,” both to students and the legal community, that the law schools “accept” employment discrimination “as a legitimate form of behavior.” Id. Notwithstanding this compelling analogy, the District 26 Court distinguished our case from Dale by suggesting there was a critical difference between the forced inclusion of a gay assistant scoutmaster and the forced presence of an “unwanted periodic visitor,” the military recruiter, in the context of a larger recruiting effort. FAIR, 291 F. Supp. 2d at 304, 305. While there was “no question” that the gay scoutmaster would “undermine the Boy Scouts’ ability to . . . inculcate its values in younger members,” the District Court wrote, the Solomon Amendment does not compel the law schools to accept the military recruiters as a “member” and does not “bestow upon them any semblance of authority.” Id. at 305. But our Court has recently held that compulsory accommodation of a government-prescribed message may violate schools’ First Amendment expressive association rights, even when that message involves our most revered affirmations of American patriotism—the Pledge of Allegiance and our National Anthem, is only minimally intrusive and lacks the schools’ imprimatur. The Circle School, 381 F.3d at 182 (holding that a statute requiring private schools to lead the Pledge of Allegiance and National Anthem violates their rights under the expressive association doctrine—“Certainly, the temporal duration of a burden on First Amendment rights is not determinative of whether there is a constitutional violation . . . . Similarly, the fact that the schools can issue a general disclaimer does not erase the First Amendment infringement at issue here, for the schools are 27 still compelled to speak the [Government’s] message.”). If the Pledge and Anthem “only take[] a very short period of time each day,” and may be preceded by “a general disclaimer regarding the recitation,” yet do not “erase the First Amendment infringement at issue here,” id., then focusing on the periodic nature of the military recruiter’s visits 11 is similarly unavailing. Moreover, the District Court’s scrutiny of the law schools’ belief that the presence of military recruiters will undermine their expressive message about fairness and social justice violates the Dale Court’s instruction to “give deference to an association’s view of what would impair its expression.” 530 U.S. at 653.12 In Dale, the Court did more 11 Furthermore, the Solomon Amendment requires law schools to do more than passively accept the presence of an “unwanted periodic visitor.” They must actively assist military recruiters in a manner equal in quality and scope to the assistance they provide other recruiters. 10 U.S.C. § 983(b)(1). 12 Dale may appear to depart from prior Supreme Court jurisprudence in this area. In two expressive association cases from the 1980s, the Court considered the claims of civic associations that state statutes forcing them to accept women as members violated their expressive association rights. Bd. of Dirs. of Rotary Int’l v. Rotary Club of Duarte, 481 U.S. 537 (1987); Roberts v. United States Jaycees, 468 U.S. 609 (1984). Closer review explains the distinction from Dale. In both cases 28 than pay lip service to deference notions. Deference distinguished the Supreme Court’s conclusion on the impairment question from that of the New Jersey Supreme Court, which had decided the case previously. The state court had ruled in Dale’s favor, holding that because the Boy Scouts have a policy of “discourag[ing] its leaders from disseminating any views on sexual issues,” Dale’s presence the Court examined the organizations’ expressive charitable and humanitarian purposes and determined that they would not be impaired by the forced inclusion of women members. Duarte, 481 U.S. at 548–49; Roberts, 468 U.S. at 626–27. The difference in outcome between these cases and Dale—the civic associations had to admit women, but the Boy Scouts did not have to admit Dale—underscores the significance of the Court’s decision to extend “deference to an association’s view of what would impair its expression.” 530 U.S. at 653. Moreover, we note that the Supreme Court had previously extended deference to what an expressive association said would impair its expression. E.g., Meyer v. Grant, 486 U.S. 414, 424 (1988) (“The First Amendment protects appellees’ right not only to advocate their cause but also to select what they believe to be the most effective means for so doing.”); Democratic Party v. Wisconsin ex rel. La Follette, 450 U.S. 107, 123–24 (1981) (“[A] court[] may not constitutionally substitute its own judgment for that of the Party. A political party’s choice among the various ways of determining the makeup of a State’s delegation to the party’s national convention is protected by the Constitution.”). 29 would not significantly affect its ability to disseminate its message. 530 U.S. at 654 (citing Dale v. Boy Scouts of America, 734 A.2d 1196, 1223 (N.J. 1999) (emphasis in original)). But faced with competing views— the Boy Scouts’ view that Dale’s presence impaired their message and the state court’s view that it could not— the Supreme Court deferred to the Boy Scouts’ view. In other words, the reason why there was “no question” (in the District Court’s words in our case, 291 F. Supp. 2d at 305) that a gay scoutmaster would undermine the Boy Scouts’ message was because the Boy Scouts said it would. Dale, 530 U.S. at 653. In our case, FAIR has supplied written evidence of its belief that the Solomon Amendment’s forcible inclusion of and assistance to military recruiters undermines their efforts to disseminate their chosen message of nondiscrimination. Accordingly, we must give Dale deference to this belief,13 and conclude that 13 Furthermore, the law schools are entitled to at least as much deference as the Boy Scouts, as the Supreme Court has recognized in other contexts that universities and law schools “occupy a special niche in our constitutional tradition,” Grutter v. Bollinger, 539 U.S. 306, 329 (2003), because of their “vital role in . . . democracy,” Sweezy v. New Hampshire, 354 U.S. 234, 250 (1957). The Court has acknowledged the importance of “autonomous decisionmaking by the academy.” Regents of the Univ. of Mich. v. Ewing, 474 U.S. 214, 226 n.12 (1985); Sweezy, 354 U.S. at 263 (Frankfurter, J., concurring) (recognizing “four essential freedoms” of a university “to determine for itself on academic grounds who may teach, what 30 FAIR likely satisfies the second element of an expressive association claim. (c) Balancing of interests The third step in evaluating an expressive association claim is “balancing the First Amendment interests implicated by the Solomon Amendment with competing societal interests to determine whether the statute transgresses constitutional boundaries.” FAIR, 291 F. Supp. 2d at 310.14 We need not linger on this analysis. Rarely has government action been deemed so integral to the advancement of a compelling may be taught, how it shall be taught, and who may be admitted to study”). The Supreme Court’s academic freedom jurisprudence thus underscores the importance of Dale deference in our case. 14 The District Court rejected FAIR’s argument that strict scrutiny applies because it did not believe that the Solomon Amendment directly burdens expressive association rights. FAIR, 291 F. Supp. 2d at 310–311. But because we concluded at step two that the Solomon Amendment impairs law schools’ expression, strict scrutiny will apply. Dale, 530 U.S. at 659 (rejecting the argument that only intermediate scrutiny should apply); The Circle School, 381 F.3d at 182 (applying strict scrutiny to statute impairing schools’ expressive association rights by requiring them to lead the Pledge of Allegiance and National Anthem). 31 purpose as to justify the suppression or compulsion of speech. We presume that the Government has a compelling interest in attracting talented military lawyers.15 But “[i]t is not enough 15 Our colleague in dissent states that “[w]e do not write on a clean slate regarding the importance Congress places in access to college and university facilities by the military” and that “[w]e have already decided that issue contrary to the argument pressed by Appellants.” In United States v. City of Philadelphia, 798 F.2d 81 (3d Cir. 1986), our Court acknowledged that “Congress considers access to college and university employment facilities by military recruiters to be a matter of paramount importance.” Id. at 86. City of Philadelphia, however, is distinguishable from this case in two important respects. First, in that case the university invited the military recruiters on campus; the recruiters’ presence was not effectively dictated by a statute, as is the case here. Id. at 83. Second, City of Philadelphia engaged in a conflict preemption analysis and held that, because it was not possible for the university to comply with both a Philadelphia anti- discrimination ordinance and the clear congressional policy concerning military recruitment on campus, the ordinance was preempted. Id. at 88–89. Our Court did not reach a balancing- of-interests inquiry. Therefore, neither this Court’s prior acknowledgment of the importance Congress places on military recruiting on college and university campuses, nor our presumption in this case that there is an important governmental interest in attracting talented lawyers to the military, ends our analysis. Rather, we must go on to reach an issue that was not present in City of Philadelphia—whether the Solomon 32 to show that the Government’s ends are compelling; the means must be carefully tailored to achieve those ends.” Sable Communications of Cal., Inc. v. FCC, 492 U.S. 115, 126 (1989). As we explain in the final section of our opinion, infra Part III.B.3(b), the Solomon Amendment could barely be tailored more broadly. Unlike a typical employer, the military has ample resources to recruit through alternative means. For example, it may generate student interest by means of loan repayment programs. And it may use sophisticated recruitment devices that are generally too expensive for use by civilian recruiters, such as television and radio advertisements. These methods do not require the assistance of law school space or personnel. And while they may be more costly, the Government has given us no reason to suspect that they are less effective than on-campus recruiting. The availability of alternative, less speech-restrictive means of effective recruitment is sufficient to render the Solomon Amendment unconstitutional under strict scrutiny analysis. Sable, 492 U.S. at 126; The Circle School, 381 F.3d at 182. But our path in this case is even clearer. The Government has failed to proffer a shred of evidence that the Amendment is narrowly tailored to achieve the Government’s ends. 33 Solomon Amendment materially enhances its stated goal. And not only might other methods of recruitment yield acceptable results, they might actually fare better than the current system. In fact, it may plausibly be the case that the Solomon Amendment, which has generated much ill will toward the military on law school campuses,16 actually impedes recruitment. 17 16 See, e.g., FAIR, 291 F. Supp. 2d at 282 (describing record evidence of student protests over military recruiting). 17 The dissent, applying the balancing-of-interests test from Roberts, 468 U.S. at 620, comes to the opposite conclusion—“that the law schools’ interests here fall at the remote extreme of Justice Brennan’s spectrum–‘where that relationship’s objective characteristics locate it . . . [near] the most attenuated of personal attachments.’” This balancing test, however, comes not from the portion of Roberts dealing with freedom of expressive association, but from the portion dealing with freedom of intimate association. The law schools are clearly not intimate associations, and where they may fall on the spectrum articulated by Justice Brennan for determining whether particular relationships merit protection under that doctrine is irrelevant to our analysis here. In Roberts, the Court went on to engage in a strict scrutiny expressive association analysis and applied the balancing test we apply here, determining that the Government had a compelling interest in eliminating discrimination and that the statute at issue was the least restrictive means of achieving that end. Roberts, 468 U.S. at 620. 34 * * * * * FAIR likely satisfies the three elements of an expressive association claim. The law schools are expressive associations, they believe the message they choose to express is impaired by the Solomon Amendment, and no compelling governmental interest exists in the record to justify this impairment. Therefore, FAIR has a reasonable likelihood of success on the merits of its expressive association claim against the Solomon Amendment. 2. Compelled Speech The Supreme Court has long recognized that, in addition to restricting suppression of speech, “the First Amendment may prevent the government from . . . compelling individuals to express certain views.” United States v. United Foods, Inc., 533 U.S. 405, 410 (2001) (citing, inter alia, W. Va. State Bd. of Educ. v. Barnette, 319 U.S. 624 (1943)). “At the heart of the First Amendment lies the principle that each person should decide for himself or herself the ideas and beliefs deserving of expression, consideration, and adherence.” Turner Broad. Sys, Inc. v. FCC, 512 U.S. 622, 641 (1994). Consistent with this principle, the Supreme Court has found impermissible compelled speech in three categories of government action. The first is government action that forces 35 a private speaker to propagate a particular message chosen by a government. See Barnette, 319 U.S. at 642 (state could not enforce compulsory flag salute statute); Wooley v. Maynard, 430 U.S. 705, 717 (1977) (state could not require drivers to display state motto on their license plates). The second is government action that forces a private speaker to accommodate or include another private speaker’s message. See Hurley v. Irish-American Gay, Lesbian, & Bisexual Group of Boston, 515 U.S. 557, 581 (1995) (state nondiscrimination statute could not be constitutionally applied to require parade organizers to include a contingent of gay marchers behind their own banner); Pacific Gas & Elec. Co. v. Pub. Utils. Comm’n, 475 U.S. 1, 12–16 (1986) (state regulatory commission could not require public utility to distribute ratepayer-group’s message in the extra space of the utility’s billing statements); Miami Herald Publ’g Co. v. Tornillo, 418 U.S. 241, 258 (1974) (state could not force newspaper to provide equal editorial-page space to candidates it opposes). The third category is government action that forces an individual to subsidize or contribute to an organization that engages in speech that the individual opposes. See United Foods, 533 U.S. at 413 (Congress could not require mushroom growers to pay assessments to fund advertisements to promote mushroom sales); Abood v. Detroit Bd. of Educ., 431 U.S. 209, 235 (1977) (state could not compel non-union employees to pay union dues to promote 36 union causes). 18 FAIR argues that the Solomon Amendment forces law schools to propagate, accommodate, and subsidize the military’s recruiting, and therefore implicates each of the three varieties of compelled speech cases. The District Court rejected FAIR’s argument and held that the law schools are not compelled to express a particular ideological message by admitting and actively assisting the military recruiters. We disagree. As we explain in the analysis that follows, the military’s recruiting is expressive of a message with which the law schools disagree. To comply with the Solomon Amendment, the law schools must affirmatively assist military recruiters in the same manner they assist other recruiters, which means they must propagate, accommodate, and subsidize the military’s message. In so doing, the Solomon Amendment conditions funding on a basis that violates the law schools’ First Amendment rights under the compelled speech doctrine. (a) Recruiting is expression. The expressive nature of recruiting is evident by the oral and written communication that recruiting entails: published and posted announcements of the recruiter’s visit, 18 We note that the subsidization line of compelled speech case law is the only one of these three categories addressed by the dissent. 37 published and oral descriptions of the employer and the jobs it is trying to fill, 19 and the oral communication of an employer’s recruiting reception and one-on-one interviews. The expressive nature of recruiting is also evident in its purpose—to convince prospective employees that an employer is worth working for. So understood, recruiting necessarily involves “communication of information, the dissemination and propagation of views and ideas, and the advocacy of causes”—the hallmarks of First Amendment expression. Village of Schaumburg v. Citizens for a Better Env’t, 444 U.S. 620, 632 (1980) (soliciting for charitable cause is expression entitled to First Amendment protection); see also Thomas v. Collins, 323 U.S. 516, 538 (1945) (recognizing First Amendment protection for the solicitation of union members). The District Court held that recruiting is not expressive activity because it “differs dramatically” from other forms of expressive activity, such as soliciting contributions and 19 For example, most recruiters submit a National Association for Law Placement (“NALP”) form that, as NALP puts it, “offers employers a thorough yet succinct way to tell their story to candidates” and includes a “narrative” section to “discuss the special characteristics” of the employer. NALP compiles these forms into a directory, which is distributed and/or made available by both law schools and employers to prospective employees. 38 proselytizing. While soliciting and proselytizing cannot be separated from the “concomitant advocacy of a particular case or viewpoint,” the District Court reasoned, recruiting does not advocate any particular cause but only has “an economic or functional motive.” FAIR, 291 F. Supp. 2d at 307–08. We agree with the District Court that soliciting and proselytizing are obvious forms of expressive activity. We part, however, on the notion that efforts to raise a legal staff are “economic or functional” while efforts to raise funds and membership are not. Recruiting, soliciting and proselytizing are similarly economic and functional and, at the same time, similarly expressive. Recruiting conveys the message that “our organization is worth working for,” while soliciting and proselytizing convey the similar functional message that “our charity is worth giving to” or “our cause is worth joining.” Having determined that recruiting is expressive, we now turn to the law schools’ disagreement with that expression. (b) The law schools’ disagreement with the speech of military recruiters. Military recruiters visiting law school campuses undoubtedly speak to students about the benefits of a career in the military, and the Solomon Amendment requires law schools to accept this speech. The law schools do not seem to 39 take issue with most of the “expressions of value, opinion, or endorsement,” Hurley, 515 U.S. at 573, made by military recruiters on campus (to the extent recruiters suggest that military careers are honorable and rewarding experiences). Nor, for the most part, do military recruiters describing careers in the military make “statements of fact the [law schools] would rather avoid.” Id. The law schools’ lack of objection to most of the speech they are forced to accept within their fora raises a key question under the compelled speech doctrine: to what extent must they disagree with the Government’s message in order for strict scrutiny to apply? Justice Souter’s dissent in Glickman v. Wileman Bros. & Elliott, Inc., 521 U.S. 457 (1997), summarized the Court’s jurisprudence to that time in suggesting that it is not necessary to show disagreement in order to sustain a compelled speech challenge. [T]he requirement of disagreement finds no legal warrant in our compelled-speech cases. In Riley [v. Fed’n of the Blind of North Carolina, Inc., 487 U.S. 781 (1988)], for example, we held that the free-speech rights of charitable solicitors were infringed by a law compelling statements of fact with which the objectors could not, and did not profess to, disagree. See 487 U.S., at 497-98, 108 S.Ct., at 2677-2678. See also Hurley, 515 U.S., at 573, 115 S.Ct., at 40 2347 (“[The] general rule, that the speaker has the right to tailor the speech, applies not only to expressions of value, opinion, or endorsement, but equally to statements of fact the speaker would rather avoid . . . [.]”); Barnette, 319 U.S., at 635, 63 S.Ct., at 1183-1184 (if the Free Speech Clause bars the government from making the flag salute a legal duty, nonconformist beliefs are not required to exempt one from saluting). Indeed, the Abood cases themselves protect objecting employees from being forced to subsidize ideological union activities unrelated to collective bargaining, without any requirement that the objectors declare that they disagree with the positions espoused by the union. See, e.g., [Chicago Teachers Union v. Hudson, 475 U.S. 292, 301-02 (1986)]; Abood, 431 U.S., at 234, 97 S.Ct., at 1799. Requiring a profession of disagreement is likewise at odds with our holding two Terms ago that no articulable message is necessary for expression to be protected, Hurley, supra, at 569, 115 S.Ct., at 2345; protection of speech is not limited to clear-cut propositions subject to assent or contradiction, but covers a broader sphere of expressive preference. . . . One need not “disagree” with an abstractionist when buying a 41 canvas from a representational painter; one merely wishes to support a different act of expression. Glickman, 521 U.S. at 488–89 (Souter, J., dissenting, joined by Rehnquist, C.J., Scalia, J., and Thomas, J.). Despite the numerous precedents to the contrary discussed by Justice Souter, it is possible to read the Glickman majority as implicitly endorsing a disagreement requirement in the compelled speech context. Glickman involved a First Amendment challenge to regulations requiring fruit growers, handlers, and processors to finance generic advertising of California nectarines, plums, and peaches. Id. at 460. The majority “presume[d]” that the fruit growers, handlers, and processors “agree[d] with the central message of the speech that is generated by the generic [government] program [at issue],” and stated that “compelled speech case law” was “inapplicable” because the scheme at issue did not, inter alia, “require them to use their own property to convey an antagonistic ideological message,” or “force them to respond to a hostile message when they would prefer to remain silent,” id. at 470–71 (citations and internal quotation marks omitted) (emphases added). However, because the degree of disagreement that may be required is minimal and in any event is present in this case, we need not determine whether such a requirement exists nor, if so, decipher its precise bounds. 42 As our dissenting colleague recently explained, the “individual’s disagreement [in a compelled speech case] can be minor, as ‘[t]he general rule is that the speaker and the audience, not the government, assess the value of the information presented.’” Cochran v. Veneman, 359 F.3d 263, 275 (3d Cir. 2004) (quoting United Foods, 533 U.S. at 411). In Cochran, we held unconstitutional a law requiring dairy producers to pay small assessments in support of “generic advertising that promotes milk.” Id. Although the aggrieved dairy producers did not disapprove of the pro-milk message at issue, the ads featured milk “produced by methods they view[ed] as wasteful and harmful to the environment,” and did not promote milk produced by their own favored methods. Id. The ads, in effect, served to promote milk produced by efforts with which the plaintiff dairy producers disagreed. Here the law schools similarly object to conveying the message that all employers are equal, and instead would rather only open their fora and use their resources to support employers who, in their eyes, do not discriminate against gays. This objection constitutes as much of a protected First Amendment interest as the objection of the dairy farmers in Cochran. Moreover, there is at least one important sense in which the law schools strenuously disagree with the very words spoken by military recruiters that the Solomon Amendment compels them to accept and to which they have been forced to respond. 10 U.S.C. § 654(b) prohibits open, practicing gays from serving in the armed forces. Military 43 recruiters undisputedly are bound by § 654(b), and do not recruit gay persons for service. Unsurprisingly, in light of § 654(b), the record demonstrates that openly gay persons who meet with military recruiters are told by the recruiters that they may not pursue military careers.20 Such speech by military recruiters is perhaps the most discordant speech the Solomon Amendment compels the law schools to accept. Yet, as we have indicated, the act of being forced to accept speech promoting an employer whose discriminatory policies the law schools disagree with is sufficient “disagreement” to bring the Solomon Amendment within the Supreme Court’s compelled speech jurisprudence. Thus, unlike the regulatory scheme at issue in Glickman, the Solomon Amendment, by requiring law schools to open their fora to military recruiters when they would prefer to do so only for non-discriminating employers, “require[s] them to use their own property to convey an antagonistic ideological message.” Glickman, 521 U.S. at 471. Likewise, by directly providing “access” to campuses for speech by military recruiters where law students are told that openly gay applicants may not serve, the Solomon Amendment requires the law schools to allow an 20 See JA107 (former ROTC student who had “wanted to be an officer in the JAG Corps since high school” interviewed with military recruiter, admitted his homosexuality, and was told that he was “ineligible due to his sexual orientation”). 44 objectionable message counter to their beliefs. In addition, both forms of speech with which the law schools disagree have resulted in, according to the record, hundreds (if not thousands) of instances of responsive speech by members of the law school communities (administrators, faculty, and students), including various broadcast e-mails by law school administrators to their communities, posters in protest of military recruiter visits, and open fora held to “ameliorate” the effects of forced on-campus speech by military recruiters. All of these represent instances in which the schools were “force[d] . . . to respond to a hostile message when they would prefer to remain silent.” Id. (internal quotation marks omitted). Therefore, the degree of the law schools’ disagreement with the military recruiters’ expression is sufficient to warrant First Amendment protection. We now determine whether the Solomon Amendment compels the law schools to engage in that expression. (c) The law schools must propagate, accommodate, and subsidize the military’s expressive message. Reasoning that the Solomon Amendment was not “an outright regulation on speech,” the District Court held that the Supreme Court’s compelled speech doctrine did not apply. FAIR, 291 F. Supp. 2d at 309. Put another way, the District Court concluded that the statute does not “directly requir[e] a private speaker to participate in the dissemination of a 45 particular message.” Id. We disagree. Having concluded above that recruiting is expression, we believe that the Solomon Amendment compels the law schools to engage in that expression in all three proscribed ways: propagation, accommodation, and subsidy. The statute insists not only on access to campus for military recruiters, but the active and equal assistance of law schools’ career services offices. For example, Harvard Law School’s career services staff offers to assist employers to “get [their] message out to students in an effective manner.” Like many law schools, the assistance Harvard provides includes coordinating interviews with students, counseling employers on effective recruiting, stuffing students’ mailboxes with employers’ information, scheduling social receptions for students, and printing employers’ announcements in the School’s newsletter. Under the express terms of the Solomon Amendment, law schools like Harvard must do the same for the military recruiters. By requiring law schools to help military recruiters “get [their] message out to students” by distributing newsletters and posting notices, the Solomon Amendment compels law schools to propagate the military’s message. Like the forced display of an unwanted motto on one’s license plate, or the compulsory recitation of a pledge, this is compelled speech. Wooley, 430 U.S. at 717; Barnette, 319 U.S. at 642. By requiring schools to include military 46 recruiters in the interviews and recruiting receptions the schools arrange, the Solomon Amendment compels the schools to accommodate the military’s message in the recruiting-assistance programs they provide for other employers. Like the forced inclusion of a parade contingent, a statement in the extra space of a utility’s billing statement, or a response in a newspaper’s editorial page, this is compelled speech. See Hurley, 515 U.S. at 569–81; Pacific Gas, 475 U.S. at 12–16; Miami Herald, 418 U.S. at 255–58. And by putting demands on the law schools’ employees and resources,21 the schools are compelled to subsidize the military’s recruiting message. Like mandatory assessments to support advertisements or political funds, this is compelled speech. See United Foods, 533 U.S. at 411–17; Abood, 431 U.S. at 235. (d) The Solomon Amendment prohibits disclaimers and, even if it did not, risk of misattribution is not an element of a compelled speech violation. 21 While we recognize that the relative cost of providing these services to one particular employer is marginal, the Supreme Court has never required that compelled subsidies be substantial to present a constitutional concern. See, e.g., United Foods, 533 U.S. at 408 (mushroom assessment at issue was one cent per pound and only some of it was going toward the objectionable advertising). 47 The District Court suggested that assisting military recruiters is not “obvious endorsement” by the law schools of the military’s point of view because “law schools can effectively disclaim any recruiting message and can easily distance themselves ideologically from the military recruiters.” FAIR, 291 F. Supp. 2d at 308, 310. But the Solomon Amendment, as recently amended, does not appear to permit law schools to disclaim the military’s message. Its express terms require them to provide treatment to the military recruiters “equal in quality and scope” to that provided to other employers. As the law schools do not disclaim the messages of those employers, similarly they may not disclaim the message of the military. Furthermore, it was in apparent response to the law schools’ ameliorative measures—their efforts to “distance themselves” (in the District Court’s words) from the military’s position—that the DOD and eventually Congress insisted on equal treatment for military recruiters. But even if the Solomon Amendment allowed for disclaimers, the Supreme Court has never held that compelled speech concerns evaporate if a speaker can ameliorate the risk of misattribution by disclaiming the message it is being compelled to propagate. To the contrary, “the presence of a disclaimer . . . does not suffice to eliminate the impermissible pressure . . . to respond to [compelled] speech.” Pacific Gas, 475 U.S. at 15 n.11 (plurality opinion). While a disclaimer reduces the risk that readers will misattribute the message, it 48 “does nothing to reduce the risk that [the compelled speaker] will be forced to respond when there is strong disagreement with the substance of [the] message.” Id. Thus, in Pacific Gas, the Supreme Court invalidated as compelled speech a requirement that a utility share the extra space in its billing statements with an organization that opposed its viewpoint. The utility’s ability to include a disclaimer did not change the analysis. In fact, a “forced reply” may add to the injury of compelled speech, not its cure. Id. at 15–16 (noting that the “pressure to respond” to compelled speech is “antithetical to the free discussion that the First Amendment seeks to foster”). In Miami Herald, the Supreme Court also invalidated a state law compelling newspapers to provide editorial page space to any political candidates that the newspaper assailed in an editorial. 418 U.S. at 255–58. It did not suggest that a newspaper could alleviate compelled speech by running a disclaimer above the candidate’s message.22 22 While the newspapers could avoid triggering the penalty of having to provide editorial page space to assailed candidates by not criticizing any candidates at all, the Court noted that this self-censorship was a form of speech suppression, itself a First Amendment injury. 418 U.S. at 257. Our case presents this self-censorship concern as well, as the law schools could avoid triggering—or at least minimize—the quality and scope of active assistance they must provide to military recruiters by limiting the quality and scope of their assistance to other 49 Similarly, in Wooley the Court held that the state motto on the Maynards’ license plate was compelled speech even though the state supreme court had expressly found in another case that “nothing in the state law . . . precludes appellees from displaying their disagreement with the state motto as long as the methods used do not obscure the license plates.” 430 U.S. at 722 (Rehnquist, J., dissenting) (citing State v. Hoskin, 295 A.2d 454 (1972)). 23 On the facts of Wooley, recruiters. 23 The Supreme Court has expressed concerns about misattribution and ability to disclaim in several of its compelled speech cases. See Hurley, 515 U.S. at 556–57 (noting that parade organizers do not customarily “disavow ‘any identity of viewpoint’ between themselves and the selected participants” and that “such disclaimers would be quite curious in a moving parade”); Turner Broadcast System v. FCC, 512 U.S. 622 (1994) (“TBS”) (noting that regulations requiring cable operators to carry broadcast signals posed little risk of misattribution because broadcasters are required by federal regulation to identify themselves at least once every hour); PruneYard Shopping Center v. Robins, 447 U.S. 74, 87 (1984) (suggesting that there was no risk that the message of students distributing political pamphlets and conducting a petition drive at a shopping mall would be attributed incorrectly to the mall owner and noting that the mall owner could disavow any connection with the message by posting signs near the petition table). But in none of these cases did the Court hold that the risk of misattribution and the speaker’s ability to disclaim the 50 there was virtually no risk that the compelled speech would be attributed to anyone other than the state. message were dispositive elements of the compelled speech doctrine. In Hurley, the Court noted that it was not “deciding on the precise significance of the likelihood of misattribution” and did not rest its holding on the parade organizer’s presumed difficulty in disclaiming the gay marchers’ message. 515 U.S. at 517. And in both PruneYard and TBS the absence of a risk of misattribution was only one of a number of factors distinguishing them from prior cases in which compelled speech had been found. PruneYard, 447 U.S. at 87; TBS, 512 U.S. at 654–55. The Court also considered the content-neutral nature of the law causing the challenged “compelled” speech, the nonexistent risk of self-censorship, and the unique characteristics of the forum (the Court later described the shopping mall in PruneYard as a “peculiarly public” forum, see Pacific Gas, 475 U.S. at 13 n.8; the TBS Court noted cable’s monopoly status and exclusive control over the “essential pathway” for disseminating a particular type of communication). TBS, 512 U.S. at 654–56; PruneYard, 447 U.S. at 87–88. And while PruneYard comes closest to holding that a speaker’s ability to disclaim a message may be relevant to the compelled speech analysis, it is notable that PruneYard predated Pacific Gas, the most express rejection of the ability to disclaim as an antidote for compelled speech. Pacific Gas, 475 U.S. at 15 n.11 (plurality opinion) (“The presence of a disclaimer . . . does not suffice to eliminate the impermissible pressure on the appellant to respond to [the unwanted] speech . . . .”). 51 In sum, law schools are expressly precluded from disclaiming or retorting the military’s recruiting message by the Solomon Amendment’s new requirement that their treatment of military recruiters be “equal in quality and scope” to the treatment of other recruiters. And while the Court has mentioned the danger of misattribution and the speaker’s ability to disclaim in several of its compelled speech cases, it has not held to date that the presence of either factor eliminated compelled speech concerns. Therefore, the District Court was wrong to reject FAIR’s compelled speech claims on the basis of its conclusion that the Solomon Amendment’s requirements posed little risk of misattribution to the law schools who in any event could effectively disclaim the military’s message. (e) The Solomon Amendment would not likely survive strict scrutiny. Although the Solomon Amendment impairs the law schools’ First Amendment rights by compelling them to propagate, accommodate, and subsidize the military’s recruiting message against their will, the statute “could still be valid if it were a narrowly tailored means of serving a compelling state interest”—i.e., if it passed strict First Amendment scrutiny. Pacific Gas, 475 U.S. at 19; see also Riley, 487 U.S. at 798 (regulation impairing speakers’ First Amendment rights under the compelled speech doctrine was subject to “exacting First Amendment scrutiny” that it did not 52 survive). We thus inquire (1) whether the Government’s interest in recruiting military lawyers is compelling, and (2) whether the Solomon Amendment is narrowly tailored to advance that goal. But as discussed above in the context of FAIR’s expressive association claim, see supra Part III.B.1(c), the Solomon Amendment does not survive strict scrutiny because the Government has not demonstrated (or even argued) that it cannot recruit effectively by less speech- restrictive means. Therefore, the balance of interests likely tips in the law schools’ favor. * * * * * To summarize, the Solomon Amendment conditions funding on the law schools’ propagation, accommodation, and subsidy of the military’s recruiting, which is expression. The Government has not shown that the assistance from law schools that the Solomon Amendment requires is narrowly tailored to advance its interest in recruiting. FAIR has thus established a reasonable likelihood of establishing that the Solomon Amendment unconstitutionally conditions funding on a basis that infringes law schools’ constitutionally protected interests under the First Amendment doctrine of compelled speech. 3. Consideration of O’Brien We turn finally to an argument that is ancillary to our 53 holding. Although the Solomon Amendment fits within the categories of First Amendment cases described in the previous sections, the District Court placed it instead into a mold it does not fit: the doctrine of expressive conduct. In so doing, it applied the intermediate scrutiny test set out by the Supreme Court in United States v. O’Brien, 391 U.S. 367 (1968), discussed at length below, for review of governmental regulations with only an incidental effect on expression. For the sake of completeness, we close by considering whether the law schools’ resistance to the military’s recruitment policy, motivated by their ideological opposition to exclusion based on sexual orientation, is expressive conduct protected by the First Amendment. 24 (a) O’Brien is inapplicable when First 24 While the expressive content of the law schools’ message is relevant also to the law schools’ expressive association claim under Dale, the analysis is different in that context. Under the rubric of expressive association, we consider whether the Solomon Amendment interferes with the law schools’ extant message of nondiscrimination, and thus impinges their associational freedom, by compelling them to assist in the military’s recruitment efforts. But with expressive conduct we ask whether resistance to the statute, i.e., exclusion of the recruiters in contravention of the statute (or its flip side, “the conduct of law schools in permitting or assisting a recruiting activity,” FAIR, 291 F. Supp. 2d at 309), is itself expressive conduct warranting First Amendment protection. 54 Amendment activity is protected on other grounds. Before exploring the contours of the O’Brien test, we explain briefly why expressive conduct fails as a descriptive model of the First Amendment issues at stake in this case. Activity simultaneously may give rise to an expressive conduct claim and to claims based on alternative theories. The premise of the category “expressive conduct” is that some activity, though it is not speech proper and is not protected under other First Amendment grounds, is crucial to public debate and warrants protection. See Texas v. Johnson, 491 U.S. 397, 404 (1989) (explaining that the Court has “long recognized that [the First Amendment’s] protection does not end at the spoken or written word” and that conduct may be “sufficiently imbued with elements of communication” to merit First Amendment protection) (citations omitted). Expressive conduct is, loosely stated, an overflow category; it is broad.25 It is therefore unsurprising that much expression 25 As noted in Johnson, id., the Supreme Court has recognized the expressive nature of students' wearing of black armbands to protest the war in Vietnam, Tinker v. Des Moines Independent Community School Dist., 393 U.S. 503, 505 (1969); of a sit-in by black citizens in a segregated area, Brown v. Louisiana, 383 U.S. 131, 141-142 (1966); of the wearing of American military uniforms in a dramatic presentation criticizing American involvement in Vietnam, Schacht v. United States, 398 U.S. 58 55 that falls squarely within the doctrines discussed in the first sections may also be cast as expressive conduct. In those cases, application of the O’Brien test is inappropriate. We need only look at the seminal expressive association and compelled speech cases to see that this is so. In Dale, for example, the Supreme Court expressly declined to rely on O’Brien, explaining: “New Jersey’s public accommodations law directly and immediately affects associational rights, in this case associational rights that enjoy First Amendment protection. Thus, O’Brien is inapplicable.” 530 U.S. at 659. Likewise in Wooley v. Maynard, the Supreme Court elected not to consider O’Brien because it considered compelled speech to be a “more appropriate First Amendment ground[].” 430 U.S. at 713. In short, the Court has not applied O’Brien where alternative First Amendment grounds were available. Taking our cue from the Supreme Court, because the Solomon Amendment is subject to strict scrutiny under the doctrines of expressive association and compelled speech, we need not engage in an O’Brien analysis. Because O’Brien scrutiny is intermediate rather than strict, demonstrating a constitutional violation under a theory of expressive conduct (1970); and of picketing in support of a wide variety of causes, see, e.g., Food Employees v. Logan Valley Plaza, Inc., 391 U.S. 308, 313-314 (1968). 56 is significantly more burdensome than under the models we have discussed. And the law schools need establish only one constitutional infirmity to justify an injunction. See, e.g., Sys. Operations, Inc. v. Scientific Games Dev. Corp., 555 F.2d 1131, 1144 (3d Cir. 1977). (b) Even under O’Brien, the Solomon Amendment is likely to impair expressive conduct unconstitutionally. Even if O’Brien applied, we would reverse the District Court’s decision because we disagree with its application of intermediate scrutiny. Notwithstanding that the District Court’s opinion featured a consistent theme—that the Solomon Amendment “targets conduct, not speech”—the Court acknowledged a communicative or expressive element in the law schools’ policies against offering the schools’ resources, support, or endorsement to any employer who does not conform to their antidiscrimination policies. FAIR, 291 F. Supp. 2d at 311. Thus, to the extent we focus on the law schools’ conduct, it is nonetheless expressive. The First Amendment protects the right to engage in expressive conduct. See, e.g., NAACP v. Claiborne Hardware Co., 458 U.S. 886, 907 (1982) (recognizing the right of boycotters to “band[] together and collectively express[] their dissatisfaction with a social structure that had denied them rights to equal treatment and respect”); Spence v. Washington, 57 418 U.S. 405, 411 (1974) (acknowledging First Amendment protection for conduct that “convey[s] a particularized message” that is understood as expression in the context of surrounding circumstances). A government regulation impairing expressive conduct is only justified “[1] if it is within the constitutional power of the Government; [2] if it furthers an important or substantial governmental interest; [3] if the governmental interest is unrelated to the suppression of free expression; and [4] if the incidental restriction on alleged First Amendment freedoms is no greater than is essential to the furtherance of that interest.” O’Brien, 391 U.S. at 377. We take no issue with the District Court’s conclusion that the Solomon Amendment is within the constitutional power of the Government, as the Constitution authorizes Congress to raise and support a military. FAIR, 291 F. Supp. 2d at 312 (citing U.S. Const. art. 1 § 8). We assume arguendo that the District Court was correct in determining that the Solomon Amendment is unrelated to the suppression of ideas. Id. at 314. And we of course presume that the United States has a vital interest in having a system for acquiring talented military lawyers. But as we noted above, the Government has chosen to submit no evidence that would support the necessity of requiring law schools to provide the military with a forum for, and assistance in, recruiting. Instead, the Government argues that “the impact of the wholesale exclusion of military recruiters [from law school campuses] is self-evident, and the government is not obligated” during 58 preliminary injunction proceedings “to assemble and present a factual record that merely confirms the dictates of common sense.” The Government fails to offer even an affidavit indicating that enforcement of the Solomon Amendment has enhanced military recruiting efforts. It suggests simply that the scope of the remedy sought by the plaintiffs relieves the Government of its obligation, pursuant to the First Amendment, to justify its curtailment of expression. How this is so we cannot conjure. We are unaware of any case so holding.26 And while the Government emphasizes that the Nation’s military is at stake, invoking the importance of a well-trained military is not a substitute for demonstrating that there is an important governmental interest in opening the law schools to military recruiting. See Rostker v. Goldberg, 453 U.S. 57, 89 (1981) (“‘[T]he phrase “war power” cannot be invoked as a talismanic incantation to . . . remove constitutional limitations safeguarding essential liberties.’” (quoting United States v. Robel, 389 U.S. 258, 263-64 26 The Government quotes Nixon v. Shrink Missouri Gov’t PAC, 528 U.S. 377, 378 (2000), for the proposition that “[t]he quantum of empirical evidence needed to satisfy heightened judicial scrutiny of legislative judgments will vary up or down with the novelty and plausibility of the justification raised.” But this is not a case where the Government has presented less evidence than might otherwise be required; here the Government has presented no evidence. 59 (1967))). 27 It may be the case, as the Government argues, that on- campus recruitment is an employer’s principal tool for attracting talented students. But it does not thereby follow that recruiting by means of the Solomon Amendment is effective. On the contrary, it seems to us equally plausible that the Solomon Amendment has in fact hampered recruitment by subjecting the military’s exclusionary policy to public scrutiny. The record is replete with references to student protests and public condemnation. In this context, it is hardly “common sense,” as the military alleges, that its presence on campus amidst such commotion and opposition has aided its recruitment efforts. In closing, we emphasize again that we need not enter the thicket of O’Brien analysis in this case. We rely on the doctrines of expressive association and compelled speech to conclude that FAIR has made the requisite showing of a likelihood of success on the merits in support of its motion for 27 We note that this is not a case involving military discretion to determine whether internal policies are necessary and appropriate. Cf. Parker v. Levy, 417 U.S. 733, 743 (1974) (“[T]he military is, by necessity, a specialized society separate from civilian society” (citation omitted)). On the contrary, this case involves the military’s compelled presence on the campuses of civilian institutions. 60 a preliminary injunction. And even under the intermediate scrutiny test of O’Brien the Solomon Amendment falters thus far, for the Government has chosen not to produce any evidence that it is no more than necessary to further the Government’s interest. Perhaps this explains why the DOD initially objected to the Amendment as “unnecessary” and “duplicative.” 140 Cong. Rec. H3864 (daily ed. May 23, 1994). C. Other preliminary injunction factors By establishing a likelihood of success on the merits of its unconstitutional condition claim based on a First Amendment violation, FAIR has necessarily satisfied the second element: irreparable harm. Elrod v. Burns, 427 U.S. 347, 373 (1976) (“The loss of First Amendment freedoms, for even minimal periods of time, unquestionably constitutes irreparable injury.”); see also Beal v. Stern, 184 F.3d 117, 123 (2d Cir. 1999) (“[T]he irreparable injury issue and the likelihood of success issue overlap almost entirely” in the First Amendment context). On the third element, we conclude that the balance of interest tips in FAIR’s favor. Without an injunction, the law schools’ First Amendment rights under the expressive association doctrine and the compelled speech doctrine will be impaired during on-campus recruiting seasons. The Government, on the other hand, does not lose the opportunity, in a proceeding on the merits, to “shoulder its full constitutional burden of proof” of showing 61 that a less restrictive alternative would not be as effective. Ashcroft v. ACLU, 124 S. Ct. 2783, 2794 (2004). As for the final element, we believe the public is best served by enjoining a statute that unconstitutionally impairs First Amendment rights. IV. Conclusion The Solomon Amendment requires law schools to express a message that is incompatible with their educational objectives, and no compelling governmental interest has been shown to deny this freedom. While no doubt military lawyers are critical to the efficient operation of the armed forces, mere incantation of the need for legal talent cannot override a clear First Amendment impairment. Even were the test less rigorous than a compelling governmental riposte to the schools’ rights under the First Amendment, failure nonetheless is foreordained at this stage, for the military fails to provide any evidence that its restrictions on speech are no more than required to further its interest in attracting good legal counsel. In this context, the Solomon Amendment cannot condition federal funding on law schools’ compliance with it. FAIR has a reasonable likelihood of success on the merits and satisfies the other injunctive elements as well. We reverse and remand for the District Court to enter a preliminary injunction against enforcement of the Solomon Amendment. 62 FAIR et al. v. Rumsfeld et al., No. 03-4433 ALDISERT, Circuit Judge, Dissenting. I would affirm the judgment of the district court. Although I have myriad problems with the fundamental contentions presented by the Appellants and the host of supporting amicus curiae briefs, essentially my disagreement is with the all-pervasive approach that this is a case of First Amendment protection in the nude. It is not. Rather, the issues before us are threefold. First, we must inquire whether Appellants have met the high burden of overcoming the presumption of constitutionality of a congressional statute that is not only bottomed on the Spending Clause, but on a number of other specific provisions in the Constitution that deal with Congress’ obligation to support the military. This is especially relevant because, in the entire history of the United States, no court heretofore has ever declared unconstitutional on First Amendment grounds any congressional statute specifically designed to support the military. Second, we must determine, using canons of logic, 63 whether a permissible factual inference—let alone a compellable one—may be properly drawn that the law schools’ anti-discrimination policies are violated from the sole evidentiary datum that a military recruiter appears on campus for a short time. Third, only if a proper inference may be drawn do we meet First Amendment considerations. The First Amendment is implicated if and only if, after applying the “balance-of- interests” test originally articulated by Justice Brennan in Roberts v. United States Jaycees, 468 U.S. 609 (1984), it can be concluded that the operation of the First Amendment trumps the several clauses of Articles I and II relating to the spending power and support of the military. Upon analysis, the argument of the Appellants and many of the amici curiae, including but not limited to the Association of American Law Schools, is rather complex. Its point of beginning takes the following tripartite form: (1) most, but not all, accredited American law schools have adopted policies that indicate they will not discriminate based on age, race, color, national origin, disability, religion, gender or sexual orientation; (2) the law schools have committed themselves to “admit students, grant scholarships, grade exams, recruit and promote faculty, and hire staff in light of these principles” (J.A. at 509); (3) in conjunction with their own commitment not to discriminate, the law schools have adopted policies stating that they will not assist employers 64 who discriminate. Their intermediate statement is that the United States military excludes service members based on evidence of homosexual conduct or orientation. See 10 U.S.C. § 654 (2004). From this, the law schools conclude that permitting the military to recruit on campus for military lawyers and military judges creates a compellable inference that the law schools are violating their own policies prohibiting discrimination on the basis of sexual orientation. They then move to the Solomon Amendment which provides that certain federal grants will not be made to “an institution of higher education ... if the Secretary of Defense determines that that institution ... has a policy or practice .. that either prohibits, or in effect prevents – (1) the Secretary of a military department or [the Department of Homeland Security] from gaining entry to campuses, or access to students (who are 17 years of age or older) on campuses, for purposes of military recruiting . . .” 28 10 U.S.C. § 983. 28 Congress has clarified that the funding restriction does not apply to the following: (1) federal grants of funds “to be available solely for student financial assistance or related administrative costs,” Pub. L. No. 106-79, § 8120, 113 Stat. 1260 (Oct. 25, 1999); (2) an institution that ceased its prior policy or practice of prohibiting or effectively preventing entry to campus or access to students on campus for military 65 This year, Congress amended the Solomon Amendment to require military recruiting access “in a manner that is at least equal in quality and scope to the [degree of] access to campuses and to students that is provided to any other employer.” National Defense Authorization Bill for Fiscal Year 2005, Pub. L. No. 108-287 (2004). From the foregoing premises Appellants’ Second Amended Complaint alleges that the Solomon Amendment and regulations promulgated thereunder violate the First Amendment as applied to law schools by: (1) imposing unconstitutional conditions on the receipt of federal funding; (2) effecting viewpoint discrimination; (3) forcing the plaintiffs to endorse messages repugnant to them and suppressing their expression of dissent; and (4) imposing vague and overbroad restrictions on speech. I would hold that Congress’ use of the spending power and fulfillment of the requirements to maintain the military under Articles I and II do not unreasonably burden speech and, therefore, do not offend the First Amendment. I apply the balance-of-interests test and decide that the interest of protecting the national security of the United States outweighs recruiting, 10 U.S.C. § 983(c)(1); and (3) an institution that “has a longstanding policy of pacifism based on historical religious affiliation,” 10 U.S.C. § 983(c)(2). 66 the indirect and attenuated interest in the law schools’ speech, expressive association and academic freedom rights. The Solomon Amendment survives the constitutional attack because its provisions, the 2004 amendments thereto and related regulations, govern conduct while only incidentally affecting speech. In serving its compelling interest in recruiting military lawyers, the statute does not require the government to engage in unconstitutional conduct. Accordingly, with respect, I dissent. I agree with the thoughtful statement of reasons of the district court and would affirm its judgment. I. The starting point for analysis must be fealty to the precept that congressional statutes are presumed to be constitutional. See, e.g., Edward J. DeBartolo Corp. v. Florida Gulf Coast Bldg. & Constr. Trades Council, 485 U.S. 568, 575 (1988) (“‘[t]he elementary rule is that every reasonable construction must be resorted to, in order to save a statute from unconstitutionality.’”) (quoting Hooper v. California, 155 U.S. 648, 657 (1895)); NLRB v. Catholic Bishop of Chicago, 440 U.S. 490, 500 (1979) ( “[A]n act of Congress ought not be construed to violate the Constitution if any other possible construction remains available.”). Thus in Rust v. Sullivan, 500 U.S. 173 (1991), the Court teaches: The principle enunciated in Hooper v. California, supra 67 and subsequent cases, is a categorical one: As between two possible interpretations of a statute, by one of which it would be unconstitutional and by the other valid, our plain duty is to adopt that which would save the Act. Blodgett v. Holden, 275 U.S. 142, 148 (1927) (opinion of Holmes, J.). This principle is based at least in part on the fact that a decision to declare an Act of Congress unconstitutional “is the gravest and most delicate duty that this Court is called on to perform.” Ibid. Following Hooper, supra, cases such as United States ex rel. Attorney General v. Delaware & Hudson Co., 213 U.S. 366, 408 (1909), and United States v. Jin Fuey and Moy, 241 U.S. 394, 401 (1916), developed the corollary doctrine that “[a] statute must be construed, if fairly possible, so as to avoid not only the conclusion that it is unconstitutional but also grave doubts upon that score.” This canon is followed out of respect for Congress, which we assume legislates in the light of constitutional limitations. FTC v. American Tobacco Co., 260 U.S. 298, 305-307 (1924). It is qualified by the proposition that “avoidance of a difficulty will not be pressed to the point of disingenuous evasion.” George Moore Ice Cream Co. v. Rose, 298 U.S. 373, 379 (1933). Id. at 190-191. 68 It is noted that although the Supreme Court considers this principle “a categorical one,” it is not included in the majority’s analysis. II. A second disagreement with the approach of my distinguished brothers of the majority is that they have not identified by name or discussed the several important provisions of the Constitution that provide for the support of the military and that antedate the promulgation of the amendments contained in the Bill of Rights. Among the powers granted to Congress is the spending power: “The Congress shall have Power To lay and collect Taxes, Duties, Imposts and Excises, to pay the Debts and provide for the common Defence and general Welfare of the United States. . .” U.S. Const. art. I, § 8, cl. 1. Furthermore, Congress is specifically given several powers related to the military: (1) “[t]o declare War, grant Letters of Marque and Reprisal, and make Rules concerning Captures on Land and Water,” id. cl. 11; (2) “[t]o raise and support Armies, but no appropriation of Money to that Use shall be for a longer Term than two Years,” id. cl. 12; (3) “[t]o provide and maintain a Navy,” id. cl. 13; and (4) “[t]o make Rules for the Government and Regulation of the land and naval Forces,” id. cl. 14. 69 The Constitution also authorizes Congress “[t]o make all Laws which shall be necessary and proper for carrying into Execution the foregoing Powers, and all other Powers vested by this Constitution in the Government of the United States, or in any Department or Officer thereof.” Id. cl. 18. The Constitution further states: “[t]he President shall be Commander in Chief of the Army and Navy of the United States. . . .” Const. art. II, § 2, cl. 1. The President also “shall take Care that the Laws be faithfully executed . . .” Id. § 3, cl. 1. Indeed, the only oblique reference to these countervailing provisions of the Constitution appears in the majority’s discussion of the unconstitutional conditions doctrine, citing Rosenberger v. Rector & Visitors of Univ. of Va., 515 U.S. 819 (1995) (public university could not condition funds for student publications on their secular perspective); FCC v. League of Women Voters, 468 U.S. 364 (1984) (FCC could not condition federal funds to radio stations on editorial content); and Perry v. Sindermann, 408 U.S. 593, 597 (1972) (relating to non-renewal of a contract and citing cases relating to denials of tax exemptions and welfare payments, but emphasizing that “most often, we have applied the principle to denials of public employment”). Significantly, my research has not discovered any reported case where an act of Congress exclusively predicated 70 on supporting the military has been declared unconstitutional by application of the seminal doctrine that “[the government] may not deny a benefit to a person on a basis that infringes his constitutionally protected interests–especially, his interest in freedom of speech.” Speiser v. Randall, 357 U.S. 513, 526- 529 (1958); see also Perry, 408 U.S. at 597. By reversing the district court’s judgment, the majority has created new law, totally unsupported by binding precedent. In doing so the majority selects analogues to cases where state public accommodation statutes were involved and not a single case where an act of Congress was not only authorized by various Clauses in Articles I and II, but commanded by them. In the posture of this case, Appellants do not urge that the Solomon Amendment is facially unconstitutional, but only that it is unconstitutional as applied to the law schools because it offends their stated policies of anti-discrimination. To succeed in their burden of overcoming the presumption of constitutionality of the Solomon Amendment, they must first demonstrate that the mere presence of recruiting officers on campus constitutes a compellable inference that the law schools will be objectively and reasonably viewed as violating their anti-discrimination policies. If they succeed at that stage, then they must demonstrate that the bite of the First Amendment under the facts of this case is so strong as to outweigh Congress’ interests to “provide for the common Defense . . .,” U.S. Const. art. I, § 8, cl. 1.; “declare War, grant Letters of Marque and Reprisal, and make Rules 71 concerning Captures on Land and Water,” id. cl. 11; “raise and support Armies, id. cl. 12; “provide and maintain a Navy,” id. cl. 13; “make Rules for the Government and Regulation of the land and naval Forces,” id. cl. 14; and for the President to “be Commander in Chief of the Army and Navy of the United States. . . .,” U.S. Const. art. II, § 2, cl. 1; and to “take Care that the Laws be faithfully executed . . .,” id. § 3, cl. 1. Before proceeding into this analysis, it bears note that the military’s policy against homosexual activity, codified at 10 U.S.C. § 654, previously has been adjudged by a number of our sister courts of appeals not to violate the Constitution. See, e.g., Richenberg v. Perry, 97 F.3d 256, 261 (8th Cir. 1996) (“We join six other circuits in concluding that the military may exclude those who engage in homosexual acts as defined in [10 U.S.C.] § 654(f)(3)(A).”). Moreover, in United States v. City of Phil., 798 F.2d 81 (3d Cir. 1986) this court has discussed the very subject of this appeal. In that case, the Temple School of Law’s placement office invited the Judge Advocate General Corps of the Army, Navy and Marine Corps to participate in a job recruiting program on its campus. he Philadelphia Commission on Human Relations issued an order restraining the law school from doing so on the ground that the military services did not accept homosexuals. We affirmed a district court order prohibiting the Commission from taking any 72 adverse action. After reviewing Congressional legislation implementing what we described as “the long standing Congressional policy of encouraging colleges and universities to cooperate with, and open their campuses to, military recruiters,” we stated: We believe that only one reasonable conclusion can be drawn from this legislation: Congress considers access to college and university employment facilities by military recruiters to be a matter of paramount importance. In other words, we think that Congress views such access an integral part of the military’s effort to conduct “intensive recruiting campaigns to obtain enlistments.” This conclusion is buttressed by the legislative history of these provisions. For example, a committee report accompanying the DDA Act of 1973 states, in pertinent part, that “the Committee believes that [the] national interest is best served by colleges and universities which provide for the full spectrum of opportunity for various career fields, including the military field through the Reserve Officers Training Corps program, and by the opportunity for students to talk to all recruiting sources, including military recruiters.” H.R.Rep No. 92-1149, 92d Cong., 2d Sess. 79 (1972). . . . We conclude, therefore, that the Order conflicts with a clearly discernible Congressional policy concerning military recruitment on the campuses of this nation’s colleges and universities. 73 Id. at 86, 88. We do not write on a clean slate regarding the importance Congress places in access to college and university facilities by the military. We already have decided that issue contrary to the argument pressed by the Appellants. And we made this determination almost twenty years ago. III. Before we address the application of First Amendment precepts, I am unwilling to accept that there is a permissible inference, let alone a compellable one, that a military presence on campus to recruit, in and of itself, conjures up an immediate impression of a discriminatory institution. Throughout our history, especially in times of war, like the present conflicts in Afghanistan and Iraq, and the military campaign against the Al Qaeda, a completely different impression is evoked. The men and women in uniform are almost universally considered as heroes, sacrificing not only their lives and well-being, but living separate from all the comforts of stateside living. Again in the current era, almost every day, a candidate for President emphasized his four months as a swift boat commander in the Vietnam conflict. As masters of public opinion, the political apparatus on both sides of the aisle certainly would not put a premium on military service if the inference of the discrimination advanced by Appellants here was attached thereto. Indeed, the respect to the man and woman in uniform is so profound that in the same Presidential campaign, the other candidate was 74 criticized for serving at home in a National Guard unit during the Vietnam conflict instead of going overseas. This view of service in the armed forces is at the farthest polar extreme from the Appellants’ position that the mere presence of military recruiters conjures up the image of an institution that discriminates. That the military does so in fact, does not, in and of itself, generate the direct and universal feeling of loathing and abomination to the extent that their presence on campus a few days a year deprives law school institutions of rights inferred from the First Amendment. What is involved here in the first instance is not operation of legal principles but precepts of logic that determine what can be properly inferred from stated circumstances. An inference is a process in which one proposition (a factual conclusion) is arrived at and affirmed on the basis of one or more other propositions, which were accepted as the starting point of the process. Professor Stebbing observes that an inference “may be defined as a mental process in which a thinker passes from the apprehension of something given, the datum, to something, the conclusion, related in a certain way to the datum, and accepted only because the datum has been accepted.” L.S. Stebbing, A M odern Introduction to Logic 211-212 (1948). Inference is a process where the thinker passes from 75 one proposition to another that is connected with the former in some way. But for the passage to be valid, it must be made according to the laws of logic that permit a reasonable movement from one proposition to another. Inference, then is “any passing from knowledge to new knowledge.” Joseph Gerard Brennan, A Handbook of Logic 1 (1957). The passage cannot be mere speculation, intuition or guessing. The key to a logical inference is the reasonable probability that the conclusion flows from the evidentiary datum because of past experiences in human affairs. A moment is necessary to discuss the difference between inference and implication. These terms are obverse sides of the same coin. We infer a conclusion from the data; the data imply a conclusion. Professor Cooley explains: [w]hen a series of statements is an instance of a valid form of inference, the conclusion will be said to follow from the premises, and the premises to imply the conclusion. If a set of premises implies a conclusion, then, whenever the premises are accepted as true, the conclusion must be accepted as true also . . . . John C. Cooley, A Primer of Formal Logic 13 (1942). As Professor Brennan put it: “In ordinary discourse, [implication] may mean ‘to give a hint,’ and [inference], ‘to take a hint.’” Brennan, A Handbook of Logic at 2-3. Drawing a proper inference is critical in this case, and this court has 76 heretofore suggested some broad guidelines: The line between a reasonable inference that may permissibly be drawn by a jury from basic facts in evidence and an impermissible speculation is not drawn by judicial idiosyncracies. The line is drawn by the laws of logic. If there is an experience of logical probability that an ultimate fact will follow a stated narrative or historical fact, then the jury is given the opportunity to draw a conclusion because there is a reasonable probability that the conclusion flows from the proven facts. As the Supreme Court has stated, “The essential requirement is that mere speculation be not allowed to do duty for probative facts after making due allowance for all reasonably possible inferences favoring the party whose case is attacked.” Tose v. First. Pa. Bank, N.A., 648 F.2d 879, 895 (3d Cir. 1981) (quoting Galloway v. United States, 319 U.S. 372, 395 (1943)). From these basic precepts of logic we cannot conclude that the mere presence of a uniformed military recruiter 77 permits or compels the inference that a law school’s anti- discrimination policy is violated. It bears repetition that the passage from datum to conclusion cannot be mere speculation, intuition, or guessing, or by “judicial idiosyncracies.” The subjective idiosyncratic impressions of some law students, some professors, or some anti-war protesters are not the test. What we know as men and women we cannot forget as judges. And this we know from elementary canons of logical processes—the validity vel non of a logical inference is the reasonable probability that the conclusion flows from the evidentiary datum because of past experiences in human affairs. A participant in a military operation cannot be ipso facto denigrated as a member of a discriminatory institution. And conjuring up such an image is the cornerstone of Appellant’s First Amendment argument. In my view it is not necessary to meet any First Amendment argument because given the evidentiary datum of a military recruiter on campus for a few days, a proper inference may not be drawn that this, in and of itself, supports a factual inference that the law school is violating its anti- discrimination policy. I think that this alone is sufficient to affirm the judgment of the district court. Nevertheless, I go further and assume that Appellants’ suggested inference may properly be drawn as a fact, and now 78 turn to a discussion of whether First Amendment concerns trump the demands placed on Congress and the President under Articles I and II to support the military. IV. Our beginning point in approaching a First Amendment analysis is the balancing-of-interests test set forth in Justice Brennan’s important opinion in Roberts: Determining the limits of state authorities over an individual’s freedom to enter into a particular association therefore unavoidably entails a careful assessment of where that relationship’s objective characteristics locate it on a spectrum from the most intimate to the most attenuated of personal attachments. . . . We need not mark the potentially significant points on this terrain with any precision. 468 U.S. at 620 (emphasis added). Moreover, important for our immediate purposes is the recognition that “[t]he right to associate for expressive purposes is not, however, absolute. Infringements on that right may be justified by regulations adopted to serve compelling state interests, unrelated to the suppression of ideas, that cannot be achieved through means 79 significantly less restrictive of associational freedoms.” Id. at 623. Although dealing with distinctions between abortions and other procedures, Justice Blackmun emphasized that in constitutional matters we do not deal with absolutes. “The constitutionality of such distinction will depend on its degree and the justification for it.” Bellotti v. Baird, 428 U.S. 132, 149-150 (1976). For other cases discussing the necessity to weigh or balance conflicting interests, see also New York State Club Ass’n. Inc. v. City of New York, 487 U.S. 1 (1988); Bd. of Dirs. of Rotary Int’l v. Rotary Club, 481 U.S. 537 (1987); Dun & Bradstreet, Inc. v. Greenmoss Builders Inc, 472 U.S. 749, 758 (1985) (“We have long recognized that not all speech is of equal First Amendment importance.”); Ohralik v. Ohio State Bar Ass’n, 436 U.S. 447 (1978); Maher v. Roe, 432 U.S. 464, 473 (1977); Whalen v. Roe, 429 U.S. 589, 599, 600 and nn.24 and 26 (1977); Zacchini v. Scripps- Howard Broad. Co., 433 U.S. 562 (1977) (emphasizing that a line has to be drawn between media reports that are protected and those that are not). A. I now turn to identify and then weigh competing interests involved in this case. I have written elsewhere that “[a]n interest is a social fact, factor or phenomenon reflected by a claim or demand or desire which human beings, either 80 individually or as groups or associations or relations, seek to satisfy and which has been recognized as socially valid by authoritative decision makers in society.” Ruggero J. Aldisert, The Judicial Process: Text, Materials and Cases 489 (2d ed. 1996) (citing authorities). Two important interests conflict here. Using the formulation of Dean Roscoe Pound, they are: (1) “an interest in general safety, long recognized in the legal order in the maxim that the safety of the people is the highest law;” and (2) the social interest in political progress and individual mental self-assertion, taking form in “the [p]olicy in favor of free speech and free belief and opinion[.]” Roscoe Pound, “A Survey of Social Interests,” 57 Harv. L. Rev. 1, 17, 34 (1943). The interest in public safety is expressed in the clauses of Articles I and II of the Constitution relating to support of the military; the interest in free speech is found in the First Amendment. I now proceed to weigh these interests. B. What is perceived to be the flash point of controversy here is whether the general interest in public safety has been trumped by the interests embodied in the First Amendment. Supporting the government’s position are the line of cases emphasizing the Supreme Court’s deference to Congress’ 81 support of the military. Arrayed against this is Appellant’s insistence that the national defense interest is trumped by the teachings of Boy Scouts of Amer. v. Dale, 530 U.S. 640 (2000), and Hurley v. Irish-American Gay, Lesbian & Bisexual Group of Boston, 515 U.S. 557 (1995). The Court has consistently deferred to congressional decisions relating to the military. “The case arises in the context of Congress’s authority over national defense and military affairs, and perhaps in no other area has the [Supreme] Court accorded Congress greater deference.” Rostker v. Goldberg, 453 U.S. 57, 64-65 (1981); see also Weiss v. United States, 510 U.S. 163, 177 (1994) (“Judicial deference . . . ‘is at its apogee’ when reviewing congressional decision making...” in the realm of military affairs). As the Supreme Court has explained, “[n]ot only is the scope of Congress’ constitutional power in this area broad, but the lack of competence on the part of the courts is marked.” Rostker, 453 U.S. at 65; see also Gilligan v. Morgan, 413 U.S. 1, 10 (1973) (stating “it is difficult to conceive of an area of governmental activity in which the courts have less competence...”). For example, in Goldman v. Weinberger, 475 U.S. 503 (1986), the Court rejected a Free Exercise challenge to a military dress regulation notwithstanding the plaintiff’s claim that the military’s assessment of the need for the regulation “is mere ipse dixit, with no support from actual experience or 82 a scientific study in the record, and is contradicted by expert testimony . . .” Id. at 509. As the Court explained, “whether or not expert witnesses may feel that religious exceptions . . . are desirable is quite beside the point[;] [t]he desirability of dress regulations in the military is decided by the appropriate military officials, and they are under no constitutional mandate to abandon their considered professional judgment.” Id. Appellants suggest that even if the military requires physical access to campuses, there is no need for military recruiters to be given the same degree of access provided to other employers. It must be emphasized that even bare physical access is more than the Appellants are willing to tolerate; they are asserting a constitutional right to exclude the military from campuses altogether. Second, it is hardly credible for the Appellants to suggest that physical access alone is sufficient for effective military recruiting, particularly when other employers are being granted far more extensive and meaningful access. It is fair to assume that all of the facilities and services provided to prospective employers by law schools are intended to facilitate the hiring process. If military recruiters are denied the ability to reach students on the same terms as other employers, damage to military recruiting is not simply probable but inevitable. The Solomon Amendment reflects Congress’ judgment about the requirements of military recruiting, and “[t]he validity of such regulations does not turn on a judge’s agreement with the 83 responsible decision maker concerning the most appropriate method for promoting significant government interests.” United States v. Albertini, 472 U.S. 675, 689 (1985). What disturbs me personally and as a judge is that the law schools seem to approach this question as an academic exercise, a question on a constitutional law examination or a moot court topic, with no thought of the effect of their action on the supply of military lawyers and military judges in the operation of the Uniform Code of Military Justice. They make it perfectly clear that they are not opposed to military institutions as such; they only want to curtail recruitment of military lawyers and judges. It is important for private employers to appear on campus to recruit law school graduates for positions with law school-sponsored “On Campus Recruiting Days” or “On Campus Interviewing” replete with interviews, followed by dinners and parties, but somehow the military will recruit its lawyers without appearing on campus. Somehow, Appellants urge, better law graduates will be attracted to the military legal branches with its lower pay and fewer benefits by some other recruiting method, for example, from the ranks of undergraduate ROTC programs. 29 Much of Appellants’ brief takes the form of 29 The following colloquy took place at the oral argument: THE COURT: What else could the government do as a less restrictive alternative? 84 conclusory statements that the military is able to attract top of the line or high quality students without stepping foot on campus. There is no explanation, however, why the law schools consider it important to have private national law firms come to campus and boast about first year associates’ salaries and signing bonuses and emphasize that if the students want to clerk for a federal judge for a year, the firm will add another bonus. This is not only OK for the private sector, but also it’s good for the law school. But we don’t want military recruiters to pollute our students. No, say the law schools, what’s sauce for the private sector goose is not sauce for the military gander. No, say the law schools, we don’t need a level playing field; let the military shift for themselves. In its demand for total exclusion of military recruiters from their campuses, “fair play” is not a phrase in the law schools’ lexicon. They obviously do not desire that our men and women in the armed services, all members of a closed society, obtain optimum justice in military courts with the best-trained lawyers and judges. It scarcely can be an exaggeration to suggest that in many respects the need for MR. ROSENKRANTZ: [A]ny number of things. Number one, ROTC, the single most effective recruiting device the military has, by their own admission. (Tr. at 25.) 85 specially competent lawyers and exceptionally qualified judges may be more important in a settled environment dominated by the strictures of discipline than in the open society of civilian life. V. I turn now to Appellants’ compelled speech argument. They argue that the Solomon Amendment trenches on their freedom of speech by compelling them to convey a message other than their own. In making this argument, the Appellants place principal reliance on the teachings of Hurley v. Irish- American Gay, Lesbian & Bisexual Group of Boston. The district court recognized, however, that nothing in Hurley suggests that the Solomon Amendment crosses the line into unconstitutionality. I agree completely and accept the government’s analysis of this issue. A. In Hurley, the Court held that a state public accommodation law could not constitutionally be applied to compel organizers of a St. Patrick’s Day parade to allow a group of gay, lesbian, and bisexual individuals to march in the parade for the purpose of conveying a public message about homosexual pride and solidarity. 515 U.S. at 572-581. The organizers did not object to the participation of the group’s members in the parade; the only question was whether the 86 group could participate in the parade “as its own parade unit carrying its own banner.” Id. at 572. The Court concluded that the law’s “apparent object is simply to require speakers to modify the content of their expression to whatever extent beneficiaries of the law choose to alter it with messages of their own,” and that in so doing, the law “violates the fundamental rule of protection under the First Amendment[] that a speaker has the autonomy to choose the content of his own message.” Id. at 573, 578. Hurley involved an effort by the government to dictate the content of a quintessential form of expressive activity—a public parade. The Court emphasized that parades “are . . . a form of expression, not just motion,” and “the inherent expressiveness of marching to make a point,” id. at 568, formed the predicate for its opinion. In contrast, there is nothing remotely so expressive about the activity of recruiting. The military engages in recruiting on college campuses for precisely the same reason as do other employers: to hire employees. Recruiting is undertaken solely for instrumental reasons, not expressive ones. To be sure, recruiting involves speaking, but the recruiter speaks purely as part of an economic transaction, and the expression is entirely subordinate to the transaction itself. It bears no resemblance to the activities of the would-be marchers in Hurley, who formed their group “for the very purpose of marching” in the parade, and who sought to march 87 “as a way to express pride in their Irish heritage as openly gay, lesbian, and bisexual individuals, to demonstrate that there are such men and women among those so descended, and to express their solidarity with like individuals who sought to march in New York’s St. Patrick’s Day parade.” Id. at 560, 570. In Hurley, unlike here, expression was not a subsidiary part of an instrumental activity; expression was the activity. The role of the parade organizers in Hurley consisted of choosing the messages that would comprise the parade, and the vice of the challenged statute was that the homosexual group’s protest message would be attributed to the organizers themselves. The Court reasoned that the group’s participation in the parade “would likely be perceived as having resulted from the Council’s customary determination about a unit admitted to the parade, that its message was worthy of presentation and quite possibly of support as well.” Id. at 575. Here, in contrast, the likelihood that members of a law school community will perceive a military recruiter’s on- campus activities as reflecting the school’s “customary determination” that the recruiter’s message is “worthy of presentation and quite possibly of support” is vanishingly small. Unlike bystanders watching a passing parade, law school students, and to be sure, their professors, are an extraordinarily sophisticated and well-informed group, who understand perfectly well that their schools admit military 88 recruiters not because they endorse any “message” that may be conveyed by the recruiters’ brief and transitory appearance on campus, but because the economic consequences of the Solomon Amendment have induced them to do so. The likelihood that the military’s recruiting will be seen as part of a law school’s own message is particularly small when schools can take— and have taken— ameliorative steps to publicize their continuing disagreement with the military’s policies and the reasons for their acquiescence in military recruiting. There is nothing to prevent the law school communities from making speeches discouraging military recruiting, posting signs and erecting huge billboards on campus or public approaches announcing their opposition and stating their reasons. That this is an important consideration has been emphasized by the Supreme Court in PruneYard Shopping Ctr. v. Robbins, 447 U.S. 74 (1980): [f]inally, as far as appears here appellants can expressly disavow any connection with the message by simply posting signs in the area where the speakers or handbillers stand. Such signs, for example, could disclaim any sponsorship of the message and could explain that the persons are communicating their own messages by virtue of state law. 89 Id. at 87. Clearly, the interests expressed in Hurley lack the power to dilute the judiciary’s traditional deference to Congress in the interest of national defense. In addition to arguing that the Solomon Amendment trenches on freedom of speech simpliciter, the Appellants also contend that the statute infringes on the law schools’ interests in expressive association. Although the First Amendment provides a measure of protection to expressive association, “the Supreme Court has required a close relationship between the [government] action and the affected expressive activity to find a constitutional violation.” Pi Lambda Phi Fraternity, Inc. v. Univ. of Pittsburgh, 229 F.3d 435, 438 (3d Cir. 2000). In the case at bar, the impact of the Solomon Amendment on the law schools’ interests in expressive association is far too remote to violate the First Amendment. In applying the balancing-of-interests test of Roberts, I am persuaded that the law schools’ interests here fall at the remote extreme of Justice Brennan’s spectrum – “where that relationship’s objective characteristics locate it . . . [near] the most attenuated of personal attachments.” 468 U.S. at 620. It is important to say again that “[t]he right to associate for expressive purposes is not, however, absolute.” Id. at 623. First Amendment claims based on expressive association are subject to a three-step constitutional inquiry. 90 See Pi Lambda Phi, 229 F.3d at 442. The first question is “whether the group making the claim [is] engaged in expressive association.” Id. If so, the next question is whether the government action at issue “significantly affect[s] the group’s ability to advocate its viewpoints.” Id. If it does, the final question is whether the governmental interests served by the law outweigh the burden imposed on the group's associational interests. Id.; see also The Circle School v. Pappert, 381 F.3d 172, 178 (3d Cir. 2004). In the case at bar the district court found as a threshold matter that law schools are engaged in expressive association, but went on to determine that the Solomon Amendment does not place a significant burden on their associational interests and that, in any event, the governmental interests served by the Solomon Amendment outweigh whatever associational burden the law may impose. (J.A. at 54-75.) C. The majority invokes cases like Glickman v. Wileman Bros. & Elliott, Inc., 521 U.S. 457, 469-470 (1997), United States v. United Foods. Inc., 533 U.S. 405 (2001), and Cochran v. Veneman, 359 F. 3d 263, (3d Cir. 2004), for the proposition that the Solomon Amendment impermissibly obligates them to "subsidize" military recruiting. In all these cases the challenged statutes obligated individuals to make direct payments of money to finance private speech with which they disagreed. Here, in contrast, the recruiting 91 activities of military recruiters are paid for exclusively with federal tax revenues; the Solomon Amendment does not obligate educational institutions to pay one red cent to the government or to a private organization. Although Appellants complain of having to provide "scarce interview space" and "make appointments," (Appellant br. at 31), this kind of physical accommodation simply does not present the constitutional concern underlying cases like Abood , Glickman, United Foods and Cochran—the concern that compelling an individual to pay for someone else's speech impinges on his right to "believe as he will" and to have his beliefs "shaped by his mind and his conscience rather than coerced by the State." Abood, 431 U.S. at 235. Unlike Abood, this case does not involve the right to make or not make “contributions for political purposes.” 431 U.S. at 234. Unlike Glickman, there was no mandatory assessments similar to those to be paid by growers of nectarines, plums and peaches under regulations 7 C.F.R. sections 916.31©), 917.35(f) promulgated under the Agricultural Marketing Agreement Act, 7 U.S.C. § 601 et seq. Unlike United Foods, there were no mandatory assessments similar to those imposed on mushroom producers for the purpose of funding generic mushroom advertisements under the Mushroom Act, 7 U.S.C. § 6101. Unlike Cochran, there were no mandatory assessments similar to those imposed on milk producers under the Dairy Promotion Stabilization Act of 1983, 7 U.S.C. § 4501 et seq. The teachings of United 92 Foods and Cochran are not applicable because, unlike the compelled advertising scheme in those cases, the principal object of the Solomon Amendment is not communication of expression but rather a furtherance of the government’s compelling interest in raising and maintaining a military force as mandated by the Constitution. Unlike a regulatory scheme requiring subsidization of generic advertising for fruit, mushrooms or milk, the Solomon Amendment “impose[s] no restraint on the freedom of any [law school] to communicate any message to any audience . . . do[es] not compel any person to engage in any actual or symbolic speech . . . [and] do[es] not compel the [law schools] to endorse or to finance any political or ideological views.” Glickman v. Wileman Bros. & Elliott, Inc., 521 U.S. 457, 469-470 (1997). Moreover, even if law schools were being required to provide direct financial payments to the government to support military recruiting, which they manifestly are not, the First Amendment provides far more latitude for compelled financial support of governmental speech than it does for compelled support of private speech. See Abood, 431 U.S. at 259 n. 13 (Powell, J., concurring in the judgment) ("Compelled [financial] support of a private association is fundamentally different from compelled support of government"); United States v. Frame, 885 F.2d 1119, 1130- 1133 (3rd Cir. 1989). Finally, what we said in Frame is relevant here: 93 Both the right to be free from compelled expressive association and the right to be free from compelled affirmation of belief presuppose a coerced nexus between the individual and the specific expressive activity. When the government allocates money from the general tax fund to controversial protects or expressive activities, the nexus between the message and the individual is attenuated. 885 F. 2d at 1132. It becomes necessary to say again that our task in this case is to identify and weigh competing interests and to emphasize again that in applying the balancing-of-interests test of Roberts v. United States Jaycees, 468 U.S. 609 (1984), the law schools’ interests here fall at the remote extreme of Justice Brennan’s spectrum— “where that relationship’s objective characteristics locate it . . . [near] the most attenuated of personal attachments.” 468 U.S. at 620. The attempt to analogize the First Amendment considerations in compelling an individual to pay for someone else's speech with a program of military recruiting fails completely because the extreme differences in the compared factual scenarios totally dominate over any purported 94 resemblances. What we explained in In re Linerboard Anti Trust Litig., 305 F. 3d 145 (3d Cir. 2002), is appropriate here: To draw an analogy between two entities is to indicate one or more respects in which they are similar and thus argue that the legal consequence attached to one set of particular facts may apply to a different set of particular facts because of the similarities in the two sets. Because a successful analogy is drawn by demonstrating the resemblances or similarities in the facts, the degree of similarity is always the crucial element. You cannot conclude that only a partial resemblance between two entities is equal to a substantial or exact correspondence. Id. at 147. VI. In challenging the district court’s reasoning, Appellants also seek to analogize this case to the teachings of Dale. As the district court recognized, (J.A. 68-70), a comparison of this case to Dale shows not why the Appellants should prevail in this case, as urged by the majority, but why they must lose, see id. at 648-650. In Dale, the Court was presented with a New Jersey 95 public accommodations law that compelled the Boy Scouts of America (“BSA”) to admit “an avowed homosexual and gay rights activist,” id. at 644, as an adult member and scoutmaster. The declared mission of the BSA was to “instill values in young people,” id. at 649, and disapproval of homosexual conduct was one of BSA’s values. BSA relied on its scoutmasters to “inculcate [Boy Scouts] with the Boy Scouts’ values—both expressly and by example.” Id. at 650. The Court reasoned that “[t]he forced inclusion of an unwanted person in a group infringes the group’s freedom of expressive association if the presence of that person affects in a significant way the group’s ability to advocate public or private viewpoints.” Id. at 648. The Court found that “the presence of Dale as an assistant scoutmaster would surely interfere with the Boy Scouts’ choice not to propound a point of view contrary to its beliefs,” because it would “force the organization to send a message, both to the youth members and the world, that the Boy Scouts accept[] homosexual conduct as a legitimate form of behavior.” Id. at 653-654. Let me now count the two ways the Solomon Amendment differs from the state statute in Dale, both of which are critical to the law’s impact vel non on associational interests. First, the Solomon Amendment simply does not impinge on the right of educational institutions to determine their membership. See 10 U.S.C. § 983. It does not purport to tell colleges and universities whom to admit as students or whom to hire as professors or administrators. It merely 96 requires them to allow the transient presence of recruiters, who are not a part of the law school and do not become members through their mere presence. In contrast to the scoutmaster in Dale, recruiters do not purport to speak “for”—and cannot reasonably be understood to be speaking “for”—the law schools that they are visiting. This case thus does not involve “[t]he forced inclusion of an unwanted person in a group.” Dale, 530 U.S. at 648. It cannot be denied that this was the genesis of the constitutional injury in Dale. Second, as noted in my discussion of Hurley, recruiting is an economic activity whose expressive content is strictly secondary to its instrumental goals. In contrast, the fundamental goal of the relationship between adult leaders and boys in the Boy Scout movement is “[t]o instill values in young people,” a goal that is pursued “by example” as well as by word. Id. at 649, 650. As a result, compelling the BSA to appoint an adult leader who was committed to “advocacy of homosexual teenagers’ need for gay role models,” id. at 645, struck at the heart of the organization’s goals. Military recruiting is not intended to “instill values” in anyone, nor is it meant to convey any message beyond the military’s interest in enlisting qualified men and women to serve as military lawyers and judges. As a result, the burden on the law schools’ associational interests is vastly less significant than the burden imposed on the BSA by the statute in Dale. 97 These profound distinctions demonstrate that the teachings of Dale lack the power to dilute the judiciary’s traditional deference to Congress in the interest of national defense. VII. I now turn to the proper measure by which to evaluate the weighing of competing interests implicated in this case. There should be no question that the teachings of United States v. O’Brien, 391 U.S. 367 (1968), control. In that case, the Court considered whether a 1965 amendment to the Universal Military Training and Service Act, which prohibited the knowing destruction or mutilation of a Selective Service Registration Certificate, was unconstitutional as applied to a man who burned his certificate as a symbolic expression of his antiwar beliefs. Id. at 369-370. The Court stated: We cannot accept the view that an apparently limitless variety of conduct can be labeled “speech” whenever the person engaging in the conduct intends thereby to express an idea. However, even on the assumption that the alleged communicative element in O’Brien’s conduct is sufficient to bring into play the First Amendment, it does not necessarily follow that the destruction of a registration certificate is 98 constitutionally protected activity. This Court has held that when “speech” and “nonspeech” elements are combined in the same course of conduct, a sufficiently important governmental interest in regulating the nonspeech element can justify incidental limitations on First Amendment freedoms. Id. at 376. In this case, the law schools portray their efforts to keep military recruiters off their campuses as “quintessential expression.” (Appellant br. at 20.) But when an institution excludes military recruiters from its campuses or otherwise restricts their access to students, it is engaging in something different from “quintessential expression.” It is engaging in a course of conduct which contains both nonspeech and speech elements. The acts which the law schools claim they are compelled to do by virtue of the military’s post-2001 “unwritten policy”—disseminating and posting military recruitment literature, making appointments for military recruiters to meet with students and providing military recruiters a place to meet with students—also contain both nonspeech and speech elements. The constitutional framework for evaluating such laws 99 is provided by O’Brien. Regulation of conduct that imposes incidental burdens on expression is constitutional if “it furthers an important or substantial governmental interest; if the governmental interest is unrelated to the suppression of free expression; and if the incidental restriction on alleged First Amendment freedoms is no greater than is essential to the furtherance of that interest.” 391 U.S. at 377. “[A]n incidental burden on speech is no greater than is essential, and therefore is permissible under O’Brien, so long as the neutral regulation promotes a substantial government interest that would be achieved less effectively absent the regulation.” Albertini, 472 U.S. at 689. Regulations of conduct that place incidental burdens on expression are not subject to a least- restrictive-alternative requirement “[s]o long as the means chosen are not substantially broader than necessary to achieve the government’s interest, . . . the regulation will not be invalid simply because a court concludes that the government's interest could be adequately served by some less-speech-restrictive alternative.” Ward v. Rock Against Racism, 491 U.S. 781, 800 (1989). The Solomon Amendment readily passes constitutional muster under these constitutional standards. The Appellants themselves do not dispute that the government has a substantial interest—indeed, a compelling one— in recruiting talented men and women for the nation’s armed forces. As the Court recognized in O’Brien, “the Nation has a vital interest in having a system for raising armies that functions with maximum efficiency . . .” 391 U.S. at 381. Effective military recruiting is the linchpin of that system. See City of Phil., 798 F.2d at 86 (“Congress considers access to college and university employment facilities by military recruiters to be a matter of paramount importance.”) The government’s interest in military recruiting, as embodied in the Solomon Amendment, is manifestly “unrelated to the suppression of free expression.” O’Brien, 391 U.S. at 377. The Solomon Amendment makes no effort to condition federal funding on the absence of campus criticism of military policies; a law school and its faculty and students are free to denounce military recruiting policies without jeopardizing federal funding in the slightest. The only thing that matters under the Solomon Amendment is whether the institution is denying access to military recruiters. And if the institution is denying access, it is irrelevant under the Solomon Amendment whether its reasons for doing so are communicative (to convey a message about its own principles or those of the military) or non-communicative (for example, to avoid participation in a recruiting process that it regards as unfair). What matters under the Solomon Amendment is “only the independent noncommunicative impact of [the] conduct,” id. at 382,—its impact on the ability of the military to reach students. The Appellants argue that because the Solomon Amendment is intended to facilitate military recruiting, and because recruiters speak to students, the governmental interest underlying the Solomon Amendment “is not unrelated to expression.” (Appellant br. at 26.) But the question posed by O’Brien is not whether the governmental interest is “unrelated to expression,” but instead whether the interest “is unrelated to the suppression of free expression.” 391 U.S. at 377 (emphasis added). The Appellants’ argument deliberately omits the touchstone of suppression from the constitutional test. Once it is recognized that suppression of expression is the focus of O’Brien, the Appellants’ argument falls apart, for the governmental interests served by the Solomon Amendment are manifestly unrelated to the suppression of anyone’s expression. It bears constant emphasis that the First Amendment test involves a balancing-of-interests as repeatedly emphasized above. The O’Brien measure is quintessentially correct because this case involves a weighing of the government’s interest in national defense and Appellants’ interest in First Amendment protections. In this posture it is difficult to conjure a case that is a more perfect fit for the exposition in O’Brien. For the foregoing reasons, I respectfully dissent. 102