Catholic Charities of Sacramento, Inc. v. Superior Court

BROWN, J., Dissenting.

This case presents questions on which reasonable minds can differ—especially in light of the whimsical and somewhat erratic path of free exercise jurisprudence after the Supreme Court’s decision in Employment Div., Ore. Dept, of Human Res. v. Smith (1990) 494 U.S. 872 [ 108 L.Ed.2d 876 , 110 S.Ct. 1595 ] (Smith). However, as a court pledged to defend constitutional limits, operating in the post-Smith environment, we ought to think very carefully about our role in defining the road ahead. *572 Instead of being dismissive of the very serious claims presented here, we should treat them with the highest respect.

After Smith , neutral, generally applicable laws do not have to survive compelling state interest review. Such laws require no justification no matter how severely they burden the individual religious claimant and no matter how inconsequential the government interest. (See Smith v. Fair Employment & Housing Com. (1996) 12 Cal.4th 1143, 1195 [ 51 Cal.Rptr.2d 700 , 913 P.2d 909 ] (cone. & dis. opn. of Kennard, J.) (Smith v. FEHC).) It is, however, far from self-evident if, or how, Smith applies to laws that directly contravene the religious conduct of religious organizations. The Women’s Contraceptive Equity Act (WCEA) attempts to circumvent this potentially substantial hurdle by creating a very narrow exemption for churches. But that begs an even more fundamental question: may the government determine what parts of bona fide religious organizations are religious and what parts are secular? And, in particular, may the government make such distinctions in order to infringe the religious freedom of that portion of the organization the government characterizes as secular? Because, unlike the majority, I do not think Smith provides obvious answers to these questions, I respectfully dissent.

I.

The proponents of the WCEA make an argument with which no one can disagree. Women in the workplace are entitled to be treated fairly and equitably and to be free from discrimination on the basis of gender. Government has not only the authority, but the obligation, to discourage invidious discrimination in the workplace, and this includes discrimination in the distribution of benefits. (See, e.g., Erickson v. Bartell Drug Company (W.D.Wash. 2001) 141 F.Supp.2d 1266, 1271 [title VII of the Civil Rights Act of 1964 (title VII) as amended by the Pregnancy Discrimination Act, prevented exclusion of contraception from prescription drug coverage offered by employer]; 42 U.S.C. § 2000e(k) [prohibiting, under title VII, discrimination on the basis of “pregnancy, childbirth, or related medical conditions”]; U.S. EEOC, Com. Dec. (Dec. 14, 2000) [coverage of contraception] <http://eeoc.gov/policy/docs/decision-contraception.html> [as of Mar. 1, 2004]; Conn. Gen. Stat., § 38a-503e (2001) [mandating insurance coverage of prescription contraception]; Mass. Gen. Laws, ch. 176B, § 4W(b) (2002) [same]; Vt. Stat. Ann., tit. 8, § 4099c (2000) [same].)

Neither the propriety, nor the wisdom of, nor the government’s authority to impose a prescription contraceptive mandate on California employers is at issue here. The question is a very narrow one. May the government impose a mandate on a religiously affiliated employer that requires the employer to pay for contraceptives—in violation of an acknowledged religious tenet—or to *573 redefine what constitutes religious conduct? 1 While antidiscrimination laws reflect a constitutional value, religious liberty occupies a commensurate level in the constitutional hierarchy. As often happens with First Amendment cases, this is “a collision between two interests of the highest order: the Government’s interest in eradicating discrimination in employment and the constitutional right of a church to manage its own affairs free from governmental interference.” (E.E.O.C. v. Catholic University, of America (D.C. Cir. 1996) 317 U.S. App. D.C. 343 [ 83 F.3d 455, 460 ] (Catholic University).) Thus, the desire to prevent discrimination cannot be the beginning and the end of the discussion.

A. Why Religious Liberty Is Important

A strong argument can be made that it was the primacy of religious liberty in the early history of this country, with its acknowledgment of the separate spheres of church and state, that gave rise to our notions of limited government and equal protection—the constitutional precursors of our antidiscrimination laws. (McConnell, Why Is Religious Liberty the “First Freedom”? (2000) 21 Cardozo L.Rev. 1243, 1244 [“the division between temporal and spiritual authority gave rise to the most fundamental features of liberal democratic order: the idea of limited government, the idea of individual conscience and hence of individual rights, and the idea of civil society, as apart from government, bearing primary responsibility for the formation and transmission of opinions and ideas”].)

Our ability to create a space for religious perspectives is both instrumental and regenerative for democracy. Religious institutions enhance individual autonomy “by challenging the sovereign power of the liberal state” (Noonan, The End of Free Exercise? (1992) 42 De Paul L.R. 567, 579-580) and by articulating alternative visions—“counter-cultural visions that challenge and push the larger community in . . . directions unimagined by prevailing beliefs.” (Brady, Religious Organizations and Mandatory Collective Bargaining Under Federal and State Labor Laws: Freedom From and Freedom For (2004) 49 Vill. L.Rev. 77, 156.) By protecting religious groups from gratuitous state interference, we convey broad benefits on individuals and society. By underestimating the transformative potential of religious organizations, we impoverish our political discourse and imperil the foundations of liberal democracy.

*574 B. Does Smith Apply to Religious Organizations?

Despite its surface simplicity, Smith is not an. easy case to understand or apply. The majority correctly quotes the critical passages from Smith: “ ‘[T]he right of free exercise does not relieve an individual of the obligation to comply with a “valid and neutral law of general applicability on the ground that the law proscribes (or prescribes) conduct that his religion prescribes (or proscribes).” ’ (Smith, [supra, 494 U.S.] at p. 879, quoting United States v. Lee [(1982)] 455 U.S. 252, 263, fn. 3 [ 71 L.Ed.2d 127 , 102 S.Ct. 1051 ] (cone. opn. of Stevens, I.).) To permit religious beliefs to excuse acts contrary to law, the Smith court reasoned, ' “would be to make the professed doctrines of religious belief superior to the law of the land, and in effect to permit every citizen to become a law unto himself.” ’ (Smith, at p. 879, quoting Reynolds v. United States (1879) 98 U.S. 145, 167 [ 25 L.Ed. 244 ].)” (Maj. opn., ante, at p. 548, italics added.)

Since Smith focused exclusively on the individual’s free exercise of religion, some courts have reasoned that religious institutions are exempted entirely from the Smith analysis. (Gellington v. Christian Methodist Episcopal Church (11th Cir. 2000) 203 F.3d 1299, 1303 ; see Kaplan, The Devil Is in the Details: Neutral, Generally Applicable Laws and Exceptions from Smith (2000) 75 N.Y.U. L.Rev. 1045, 1070.)

1. Individuals v. Institutions

This case involves a religious organization and not an individual. Perhaps more importantly, it does not deal with the denial of a benefit because of a violation of existing law. Rather, it attempts to assess the constitutional implications of a law that requires a religious organization to provide a benefit despite its theological objections. These fundamental differences are simply ignored in the majority’s analysis.

Under Smith, the right of free exercise does not relieve an individual of the obligation to comply with a valid and neutral law of general applicability even if the law requires conduct that contravenes a religious belief, but “[i]t does not follow . . . that Smith stands for the proposition that a church may never be relieved from such an obligation.” (Catholic University, supra, 83 F.3d at p. 462 .)

The majority may have made an abortive attempt to deal with this obvious distinction by citing, and dismissing, the so-called ministerial exception. It is true, as the majority notes, that the ministerial exception is not directly at issue here. (See, e.g., Alicea-Hernandez v. Catholic Bishop of Chicago (7th Cir. 2003) 320 F.3d 698 [ministerial exception to title VII]; E.E.O.C. v. *575 Roman Catholic Diocese of Raleigh, NC (4th Cir. 2000) 213 F.3d 795 [same]; Combs v. Cen. Tx. Ann. Conf. United Methodist Church (5th Cir. 1999) 173 F.3d 343 [same].) Likewise, it is certainly debatable whether the legislative action challenged here invades the narrow domain labeled church autonomy. (See, e.g., Serbian Orthodox Diocese v. Milivojevich (1976) 426 U.S. 696 [ 49 L.Ed.2d 151 , 96 S.Ct. 2372 ] [state court impermissibly encroached on church autonomy]; Kedroffv. St. Nicholas Cathedral (1952) 344 U.S. 94 [ 97 L.Ed. 120 , 73 S.Ct. 143 ] [state statute impermissibly encroached on church autonomy].) And yet, the logic of these cases suggests that the constitutionally protected space for religious organizations is actually broader than these obvious categories. In short, the ministerial exception and the church autonomy doctrine are ways of describing spheres of constitutionally required protection, but these categories are not exhaustive.

The court in Catholic University summarized the distinction it was making this way: “We conclude from our review of the Supreme Court’s First Amendment jurisprudence that whereas the Free Exercise Clause guarantees a church’s freedom to decide how it will govern itself, what it will teach, and to whom it will entrust its ministerial responsibilities, it does not guarantee the right of its members to practice what their church may preach if that practice is forbidden by a neutral law of general application.” (Catholic University, supra, 83 F.3d at p. 463 .) In fact, the Legislature apparently takes a similar view of the breadth of Smith because it provided an exemption from the WCEA for churches.

2. The Two Faces of Entanglement

Under venerable establishment clause precedent, however, the exemption itself is problematic. To put it bluntly, the government may generally separate the religious from the secular to decide how it will dispense its benefits, but it cannot parse a bona fide religious organization into “secular” and “religious” components solely to impose burdens on the secular portion.

As noted, ante, the constitutional basis for the distinction seems indisputable. The United States Supreme Court has recognized that government action may burden the free exercise of religion in two different ways: “by interfering with a believer’s ability to observe the commands or practices of his faith [citations], and by encroaching on the ability of a church to manage its internal affairs.” (Catholic University, supra, 83 F.3d at p. 460 ; see, e.g., Church of Lukumi Babalu Aye, Inc. v. City of Hialeah (1993) 508 U.S. 520, 531-533 [ 124 L.Ed.2d 472 , 113 S.Ct. 2217 ] (Lukumi); Kedroffv. St. Nicholas Cathedral, supra, 344 U.S. at p. 116 [free exercise clause protects power of religious organizations “to decide for themselves, free from state interference, matters of church government as well as those of faith and doctrine”].)

*576 If Catholic Charities were a “religious employer” it would be exempt from the WCEA’s requirement to include coverage for contraceptives in its group health care policy. Under the act, a religious employer must satisfy all of the following criteria: “(A) The inculcation of religious values is the purpose of the entity, [f] (B) The entity primarily employs persons who share the religious tenets of the entity. [j[] (C) The entity serves primarily persons who share the religious tenets of the entity. [|] (D) The entity is a nonprofit organization as described in Section 6033(a)(2)(A)i or iii, of the Internal Revenue Code of 1986, as amended.” (Health & Saf. Code, § 1367.25, subd. (b)(1).)

As the majority notes, “Catholic Charities does not qualify as a ‘religious employer’ under the WCEA because it does not meet any of the definition’s four criteria.” (Maj. opn., ante, at p. 539, italics added.) But Catholic Charities would be a religious employer if the Legislature had not designed the exemption narrowly enough to exclude it. 2 The plaintiffs contend the Legislature has “deliberately defined the Catholic Church in a manner entirely inconsistent with Catholic religious teaching, to exclude critical, constitutive elements of the Catholic Church—i.e., the Church’s health care, social service and educational ministries—from the definition of ‘religious employer’ included in the exemption provisions.”

The high court “ ‘has long recognized that the government may (and sometimes must) accommodate religious practices and that it may do so without violating the Establishment Clause.’ ” (Corporation of Presiding Bishop v. Amos (1987) 483 U.S. 327, 334 [ 97 L.Ed.2d 273 , 107 S.Ct. 2862 ] (Amos).) At the same time, acknowledging that churches often regard the community services provided by affiliated nonprofits as “a means of fulfilling religious duty and of providing an example of the way of life a church seeks to foster” (id. at p. 344 (cone. opn. of Brennan, J.)), the court concluded the case-by-case determination of whether an affiliated nonprofit is religious or secular is inappropriate under the free exercise clause. (Id. at pp. 341-342 (cone. opn. of Brennan, J.) [“Religion includes important communal elements for most believers. They exercise their religion through religious organizations and these organizations must be protected by. the [Free Exercise] Clause. . . . [R]eligious activity derives meaning in large measure from participation in a larger religious community”].)

Even after Smith, it seems quite clear the government may not discriminate among religions (Larson v. Valente (1982) 456 U.S. 228, 253 [ 72 L.Ed.2d 33 , *577 102 S.Ct. 1673 ]) or engender a risk of politicizing religion (id. at pp. 253-254) or purport to exempt “religious” but not “secular” activities (Cantwell v. Connecticut (1940) 310 U.S. 296, 301 [ 84 L.Ed. 1213 , 60 S.Ct. 900 ]; Espinosa v. Rusk (10th Cir. 1980) 634 F.2d 477 , 480-481, affd. (1982) 456 U.S. 951 [ 72 L.Ed.2d 411, 102 S.Ct. 2025 ]). In NLRB v. Catholic Bishop of Chicago (1979) 440 U.S. 490 [ 59 L.Ed.2d 533 , 99 S.Ct. 1313 ] (Catholic Bishop), the National Labor Relations Board (NLRB) certified unions as bargaining agents for lay teachers in church-affiliated schools. The NLRB asserted it was required to decline jurisdiction only when schools were“ ‘completely religious’ ” and not just “ ‘religiously associated.’ ” (Id. at p. 493.) The Seventh Circuit Court of Appeals rejected the NLRB’s standard as a “ ‘simplistic black or white, purported rule’ ” which offered no guide to discretion. (Id. at p. 495.) “ ‘The real difficulty is found in the chilling aspect that the requirement of bargaining will impose on the exercise of the bishops’ control of the religious mission of the schools.’ ” (Id. at p. 496.) The Supreme Court, after acknowledging that the NLRB’s attempt to distinguish between “ ‘completely religious’ ” and “ ‘religiously associated’ ” was a recognition of its intrusion into areas protected by the religion clauses, construed the National Labor Relations Act so as to avoid deciding whether jurisdiction “was constitutionally permissible under the Religion Clauses of the First Amendment.” (Catholic Bishop, at p. 499.) Nevertheless, the court expressed concern that NLRB jurisdiction would inevitably involve “inquiry into the good faith of the position asserted by clergy-administrators and its relationship to the school’s religious mission,” and the “very process of inquiry” would impinge on rights guaranteed by the religion clauses. (Catholic Bishop, at p. 502.)

In Universidad Cent, de Bayamón v. N.L.R.B. (1st Cir. 1985) 793 F.2d 383, 387 , the NLRB sought to avoid the problem by exempting “ ‘pervasively sectarian’ ” schools. The controlling opinion in Universidad Cent, de Bayamon found board jurisdiction posed just as great a risk as the Supreme Court envisioned in Catholic Bishop, supra, 440 U.S. 490 . “For the Board to exercise jurisdiction over an educational institution where ‘the inculcation of religious values is at least one purpose of the institution’ and ‘to promise that courts in the future will control the Board’s efforts to examine religious matters, is to tread the path that Catholic Bishop forecloses.’ ” (University of Great Falls v. N.L.R.B. (D.C. Cir. 2002) 349 U.S. App. D.C. 386 [ 278 F.3d 1335, 1342 ] (Great Falls), quoting Universidad Cent, de Bayamon v. N.L.R.B., at p. 402.)

In Great Falls, the Court of Appeals for the District of Columbia rejected the NLRB’s latest effort—the “substantial religious character” test—because the multifaceted analysis created the same concerns as the approach rejected in Catholic Bishop. Moreover, the court invoked a long line of precedents *578 which have made it clear that religious tests, inquiries into religious perspectives, or generally trolling through a person’s or institution’s religious beliefs is “ ‘not only unnecessary but also offensive.’ ” (Great Falls, supra, 278 F.3d at pp. 1341-1342, quoting Mitchell v. Helms (2000) 530 U.S. 793, 828 [ 147 L.Ed.2d 660 , 120 S.Ct. 2530 ]; Amos, supra, 483 U.S. 327, 340, 345 (cone, opn. of Brennan, J.).)

The court in Great Falls thus suggested a broad exemption which would avoid the pitfalls of having the government determine what is religious or how much religion is sufficient. The court would exempt any school which purports to provide a religious environment; is organized as a nonprofit; and affiliated with, or owned, or operated, or controlled directly or indirectly by a recognized religious organization or entity whose membership is determined at least in part with reference to religion. (Great Falls, supra, 278 F.3d at p. 1343 .) The point of this bright-line test was to avoid delving into religious doctrine or motive and to avoid coercing a religiously affiliated educational institution to alter its religious mission to meet regulatory demands. (Id. at p. 1345.) This approach responds to a long-standing concern that the religious liberty protected by the Constitution ought not to depend on a “determination by state authority as to what is a religious cause.” (Cantwell v. State of Connecticut, supra, 310 U.S. at p. 307 .)

Of course, the cited cases are distinguishable. The controversy here does not involve solicitation, or potential chilling effects, religious schools, administrative discretion, or ad hoc determinations. In reality, this case is worse. Here we are dealing with an intentional, purposeful intrusion into a religious organization’s expression of its religious tenets and sense of mission. The government is not accidentally or incidentally interfering with religious practice; it is doing so willfully by making a judgment about what is or is not religious. This is precisely the sort of behavior that has been condemned in every other context. The conduct is hardly less offensive because it is codified. Definition may be just as pernicious as ongoing monitoring if its purpose is to suppress or burden religious conduct. (Espinosa v. Rusk, supra, 634 F.2d at p. 481 [“The conception of religion entertained by the City . . . was that it had to be purely spiritual or evangelical. Thus, the charitable activity of the church having to do with the feeding of the hungry or the offer of clothing and shelter to the poor was deemed subject to regulation. This broad definition of secular is part of the problem”].)

3. The Meaning of Neutrality

In theory, when religious liberties are at stake, the state is only neutral when it does not choose sides. (Laycock, Religious Liberty as Liberty (1996) 7 J. Contemp. Legal Issues 313 [“[T]he core point of religious liberty is that *579 the government does not take positions on religious questions—not in its daily administration, not in its laws, and not in its Constitution either”].) This would mean that the state may not prefer or seek to impose a particular normative view by squelching a competing religious perspective. Genuine neutrality would “allow many different and contending voices to be represented in public discourse.” (McConnell, Why Is Religious Liberty the “First Freedom’’?, supra, 21 Cardozo L.Rev. at p. 1262.)

In the present controversy, one side posits that sex is an aspect of autonomy, a vital human function in which men and women should be able to engage, enjoying their sexuality “free from anxiety.” (Hayden, Gender Discrimination Within the Reproductive Health Care System: Viagra v. Birth Control (1999) 13 J.L. & Health 171 , 181.) This may in fact be the view of a majority of American adults. The Catholic Church’s view, in contrast, deems all forms of nonmarital sex immoral, and views sex within marriage as a unitive, procreative, and sacred reflection of a spiritual, emotional, and biological reality that comes complete with reproductive anxiety. (See George & Bradley, Marriage and the Liberal Imagination (1995) 84 Geo. L.J. 301 -320.) This is a perspective many people would disparage as archaic. Several of the legislators debating the WCEA seemed to think so. 3

The Catholic Church purports to be one of those different and contending voices, a church which “has never envisioned a sharp divide between the Church and the world, the spiritual and the temporal, or religion and politics. For the Church, the internal spiritual life of its members and institutions must always move outward as a sign and instrument for the transformation of the larger society.” (Brady, Religious Organizations and Mandatory Collective Bargaining Under Federal and State Labor Laws: Freedom From and Freedom For, supra, 49 Vill. L.Rev. at p. 157.)

Petitioner complains the narrow exemption was designed to lend the state’s “considerable weight to the dissenting side of a conflict within the church about the legitimacy of contraceptive practice—under the banner of protecting the ‘rights’ of those who disagree . . . and to deny the church exemption based on the allegedly unpopular nature of a church doctrine that diverges *580 from contemporary cultural mores.” In petitioner’s words, the state’s “action-has the effect of déclaring the Catholic hierarchy’s' stand ‘heresy’ in the eyes of secular culture.”

Of course, practice always diverges from theory. In contemporary American society, the government does take sides on policy issues. The First Amendment precludes the government from taking sides if the dispute involves internal church governance, but that leaves an area of overlap where the religiously dictated conduct of churches operating in the world comes into conflict with public policy. The question then is whether the coercive force of the law may be brought to bear to compel a religious organization that holds an alternative view, based on religious scruples, to support a hostile and competing vision of the good.

a. Religious bigotry

Smith could be read, as the majority apparently reads it, to suggest that religion is not entitled to constitutional protection unless the government action expressly and specifically targets religious expression. Under this interpretation, protection for religious liberty requires proof of religious bigotry, i.e., proof that government oficiáis acted out of antireligious motives. Thus, Smith—even as modified by Lukumi—would prohibit infringements of religious liberties only if a statute has the “object or purpose of . . . suppress[ing] religion or religious conduct” or involves “[o]fficial action that targets religious conduct for distinctive treatment.” ( Lukumi, supra, 508 U.S. at pp. 533-534.) Since this statute imposes a mandate on all employers that provide prescription coverage, it arguably does not target religious conduct. On one level, religious interests and secular interests are treated with equal dignity, and since the mandate provides an escape hatch, Catholic Charities’ attempt to claim specifically unequal treatment faces formidable obstacles. Consequently, the majority finds Catholic Charities has failed to prove an antireligious motive and the statute is neutral.

b. Objects and effect

There is, however, more than one way to look at neutrality. As Lukumi explains it, “[fjacial neutrality is not determinative. . . . The [free exercise] clause ‘forbids subtle departures from neutrality’ [citation] and ‘covert suppression of particular religious beliefs.’ ” (Lukumi, supra, 508 U.S. at p. 543 .) “Apart from the text, the effect of a law in its real operation is strong evidence of its object.” (Id. at p. 535.) “[I]f the object of the law- is to infringe upon or restrict practices because of their religious motivation, the law is not neutral [citation]; and it is invalid unless it is justified by a compelling interest and is narrowly tailored to advance that interest.” (Id. at p. 533.) “The *581 Free Exercise Clause ‘protect[s] religious observers against unequal treatment.’ ” (Lukumi, at p. 542.) But equality in the context of religious liberty must be broadly defined. In effect, the general applicability requirement is. needed to ensure neutrality across broad categories of regulation. Pursuant to Lukumi, if other activities which cause comparable harm to the same governmental interests are not regulated, the law is not generally applicable. Thus, Lukumi makes it clear that strict scrutiny is. required if a law is not neutral—and it considers the question of neutrality broadly.

In this case, for instance, defendants argue that Catholic Charities’ ability to opt out, i.e., to choose not to provide any prescription coverage, obviates any concern about infringement. Catholic Charities insists it should not be forced to relinquish its vision of appropriate employee relations to preserve its right to object to the use of contraceptives. From the Church’s perspective, to demand that contraception be funded, despite bona fide religious objections, is to take sides, to abandon the commitment to public neutrality. In this sense, the WCEA, with its grudging religious exemption, may not be neutral. The majority’s response that the WCEA’s narrow exemption is an accommodation and not an imposition seems entirely unresponsive.

In the whole scheme of things, the risk associated with allowing government to impose a stifling orthodoxy in pursuit of the good society may greatly outweigh the small harm of tolerating heterodoxy in this circumstance. 4 At oral argument, counsel indicated the Catholic Church, including Catholic Charities, employs fewer than 60,000 of California’s millions of employees. 5 Some of the Church’s employees belong to religious orders and are presumably fully in agreement with the church’s position. Some are men, *582 some are women no longer capable of childbearing, and some are spouses of people employed by other companies who are covered by their spouses’ health plans. Of the women of childbearing age who remain, and to whom contraceptive coverage is a critical concern, none are faced with a pervasive practice which would prevent them from finding more congenial employment. 6 The existence of WCEA’s mandate—to which the vast majority of California employers apparently have no religious objection—enhances their employment options. In fact, the defection of talented female employees may cause Catholic Charities to reconsider its position. Such a result has no First Amendment implications.

A substantial amount of federal case law supports Catholic Charities’ claim that the Legislature’s attempt to draw distinctions between the religious and secular activities of a single religious entity is an impermissible government entanglement in religion. I am inclined to agree. Such an action is constitutionally invalid, and that ends the discussion. If, however, the existence of the narrow exemption simply shows the statutory scheme is not neutral in operation or effect, it is invalid only if it fails strict scrutiny.

*583 C. Strict Scrutiny

Strict scrutiny is not what it once was. Described in the past as “strict in theory and fatal in fact” (Gunther, Foreword: In Search of Evolving Doctrine on a Changing Court: A Model for Newer Equal Protection (1972) 86 Harv. L.Rev. 1, 8), it has mellowed in recent decades (see, e.g., Grutter v. Bollinger (2003) 539 U.S. 306 [ 123 S.Ct. 2325, 2338 , 156 L.Ed.2d 304 ] [holding state law school’s race-based affirmative action program survived strict scrutiny and noting that “[s]trict scrutiny is not ‘strict in theory, but fatal in fact’ ”]).

If recent precedent is any guide, a state’s interest is compelling if the state says it is. Thus, consistent with federal precedent, compelling interest now seems more or less coextensive with the state’s asserted exercise of police power.

1. Compelling State Interest

Unquestionably, the desire to eradicate invidious discrimination is a compelling state interest. But is the desire to force conformity on a single employer that objects to contraception on religious grounds also a compelling state interest? In the latter case, the state is not dealing with invidious discrimination; it is trying to prevent a disparate impact. Catholic Charities does not discriminate because of an animus against women. It opposes all forms of birth control, except abstinence, whether for men or women, whether prescription or over-the-counter, whether surgical, oral, or mechanical.

2. Narrow Tailoring

The WCEA defines as religious only those organizations for which the inculcation of religious values is the sole purpose of the entity, that primarily employ only adherents of their own faith tradition, that primarily serve only people who share their religious tenets, and that qualify as nonprofit organizations described in section 6033(a)(2)(A)(i) or (iii) of the Internal Revenue Code of 1986.

This is such a crabbed and constricted view of religion that it would define the ministry of Jesus Christ as a secular activity. 7 The stinginess of the *584 exemption makes the structure of the act all the more baffling. The mandate applies only to employers that provide prescription coverage. Thus, Catholic Charities can avoid the mandate by dropping the coverage. The state wants to make sure that women are not burdened more than others. Where employers cooperate, the WCEA will reduce the inequitable financial burden of healthcare for women. If religiously affiliated employers are serious about their objections, however, women who work for those employers could actually be worse off.

The only reasons given for narrowing the exemption so drastically is the alleged concern that the exception could “swallow up” the rule because the numbers of employees who work for secular organizations affiliated with religious entities could easily approach several hundred thousand; the exemption might deprive thousands of employees of access to nondiscriminatory health and disability insurance; and a desire exists to extend coverage to as many people as possible. There are a few problems with this litany. First, the act, as its structure demonstrates and as the majority candidly admits, has nothing to do with access or extending coverage. “[T]he principal purpose of the WCEA ... is not to facilitate access to contraceptives but to eliminate a form of gender discrimination in the provision of health benefits.” (Maj. opn., ante, at p. 566.) Moreover, the record provides no support for the claim that the exemption potentially affects several hundred thousand employees.

Furthermore, employers have the option of self-insuring. The Employee Retirement Income Security Act preempts state regulation of self-insured companies and “prohibits states from mandating benefits or defining discrimination in self-insured employee benefit plans more broadly than federal law.” (Law, Sex Discrimination and Insurance for Contraception (1998) 73 Wash. L.Rev. 363, 395; 29 U.S.C. § 1001 et seq.) Such employers would not only not be subject to mandatory prescription coverage, they would not be subject to any of California’s more restrictive insurance regulations. Arguably, the existence of these secular exemptions supports a religiously-affiliated-employer exemption even under Smith. The state would also need to show its refusal to countenance a religious exception, in a regulatory arena rife with exceptions, is not “official action that targets religious conduct for distinctive treatment.” ( Lukumi, supra, 508 U.S. at pp. 533-534.)

II.

Thus, whether the WCEA would survive strict scrutiny—even under the relaxed federal standard—seems a much closer question than the majority *585 acknowledges. But there may be other good reasons to rely on independent state grounds. Changes in the interpretation of the federal charter are not only becoming more frequent, the balancing test, and the standards applied to them, are shifting. Instead of applying Smith, we might view it as effectively returning free exercise questions to the states.

A. A Document of Independent Force

“We may take it for granted that the meaning of California Constitution article I, section 4, ... is not dependent on the meaning of any provision of the federal Constitution. The state charter declares in so many words that ‘[flights guaranteed by this Constitution are not dependent on those guaranteed by the United States Constitution.’ (Cal. Const., art. I, § 24.)” (Smith v. FEHC, supra, 12 Cal.4th at p. 1177 .) “Respect for our Constitution as ‘a document of independent force’ [citation] forbids us to abandon settled applications of its terms every time changes are announced in the interpretation of the federal charter.” (People v. Pettingill (1978) 21 Cal.3d 231, 248 [ 145 Cal.Rptr. 861 , 578 P.2d 108 ], quoting People v. Brisendine (1975) 13 Cal.3d 528, 549-550 [ 119 Cal.Rptr. 315 , 531 P.2d 1099 ].)

This is true even when the language is identical to the federal Constitution, but is particularly true when the language differs. (See, e.g., Golden Gateway Center v. Golden Gateway Tenants Assn. (2001) 26 Cal.4th 1013, 1019 [ 111 Cal.Rptr.2d 336 , 29 P.3d 797 ] [“Unlike the United States Constitution, which couches the right to free speech as a limit on congressional power (see U.S. Const., 1st Amend.), the California Constitution gives ‘[e]very person’ an affirmative right to free speech. [Citation.] Accordingly, we have held that our free speech clause is ‘more definitive and inclusive than the First Amendment’ ” (fn. omitted)].)

Similarly, although we have said California’s establishment clause is coextensive with the federal provision (East Bay Asian Local Development Corp. v. State of California (2000) 24 Cal.4th 693, 718 [ 102 Cal.Rptr.2d 280 , 13 P.3d 1122 ]), California’s free exercise clause guarantees “free exercise and enjoyment of religion without discrimination or preference” and specifies that “liberty of conscience does not excuse acts that are licentious or inconsistent with the peace and safety of the state.” (Cal. Const., art. I, § 4.) We do not have to decide that this language literally embodies the strict scrutiny test. The drafting history of California’s free exercise clause is not clear enough to resolve the question definitively. Although the proponents of the licentious acts clause may simply have wanted to preserve the ability of the state to regulate specific practices they considered immoral or dangerous (Browne, Rep. of the Debates in Convention of Cal. on Formation of State Const. (1850) p. 39), that does not mean they thought the language was otherwise synonymous with the language of the federal Constitution.

*586 B. The Compelling State Interest Analysis

The majority carefully avoids deciding whether strict scrutiny would be required under the California Constitution. Other states with very similar constitutional liberty of conscience clauses have found that infringement requires strict scrutiny. (See, e.g., Humphrey v. Lane (2000) 89 OhioSt.3d 62 [ 2000 Ohio 435 , 728 N.E.2d 1039, 1043 ] [holding that under the Ohio Constitution, “the standard for reviewing a generally applicable, religion-neutral state regulation that allegedly violates a person’s right to free exercise of religion is whether the regulation serves a compelling state interest and is the least restrictive means of furthering that interest” and finding the regulation at bar not the least restrictive]; State v. Hershberger (Minn. 1990) 462 N.W.2d 393 [under the Minnesota Constitution, neutral motor vehicle statute, which burdened Amish religious exercise, failed compelling state interest test since state failed to show lack of reasonable alternative means]; First Covenant Church v. City of Seattle (1992) 120 Wn.2d 203 [ 840 P.2d 174, 187 ] [statute that burdened free exercise failed state compelling interest test under Washington Constitution since the state’s interest was not of sufficient magnitude to outweigh free exercise of religion].)

At the very least, the constitutional weight of the state’s interest must be affected by the size and severity of the problem the state is attempting to solve. To authorize the state to use a howitzer to smite a gnat should be no part of our constitutional jurisprudence. Where strict scrutiny applies, the state “may abridge religious practices only upon a demonstration that some compelling state interest outweighs the defendants’ interests in religious freedom.” (People v. Woody (1964) 61 Cal.2d 716, 718 [ 40 Cal.Rptr. 69 , 394 P.2d 813 ] (Woody).)

It may also be true that “[s]ection 4 has not played an independent role in free exercise claims” (Grodin et al., The Cal. State Constitution: A Reference Guide (1993) p. 44), but does that mean it should remain dormant? In Woody , the court relied on the First Amendment rather than the California provision, but in doing so, the court applied strict scrutiny and insisted on a searching inquiry. Under California law—at least up to now-—the compelling state interest test had bite and required the court to “weigh[] the competing values represented ... on the symbolic scale of constitutionality.” (Woody, supra, 61 Cal.2d at p. 727 .) Untested assertions of a possible deleterious effect on a statutory scheme were not sufficient. (Id. at p. 724.) In Woody , the court concluded that uniform enforcement of neutral criminal drug laws (similar to the laws at issue in Smith) was not a compelling reason to intrude upon sincere religious practices. In explaining why the interest in drug enforcement—while undeniably important—was not compelling enough, the court said: “In a mass society, which presses at every point toward conformity, the *587 protection of self-expression, however unique, of the individual and the group becomes ever more important. The varying currents of the subcultures that flow into the mainstream of our national life give it depth and beauty.” (Woody, at p. 727.) These concerns should be heightened when the government seeks to redefine the core theology of religious organizations.

Under the standard enunciated in Woody , the state has actually failed to meet its burden. The whole debate ensues because the state found that “approximately 10 percent of commercially insured Californians did not have coverage for prescription contraceptives.” (Maj. opn., ante, at p. 537.) Presumably that 10 percent includes both men and women. Still, it means that 90 percént of Californians who are commercially insured do have such coverage! The insurance gap itself is not large, and Catholic Church employers can constitute only a small percentage of that small percentage.

Moreover, even if we assume the interests at issue here are both compelling and of equal weight, the Legislature’s refusal to grant a broader exemption—one which would not embroil the government in the unseemly task of deciding what is “religious”—is inexplicable. The state has produced no substantial evidence that the exemption of Catholic Charities from this particular mandate would render the whole scheme ineffective or would be so administratively burdensome as to preclude enforcement. As petitioner poses the question: “[I]f closing the Catholic gap [was] not the problem,” how can “ ‘granting an exemption to Catholic employers’ . . . ‘defeat the purpose of the bill’ ”? There has been no showing that the interests served by the WCEA—which focuses on a modest 10 percent gap in coverage—cannot be achieved by less restrictive means.

Conclusion

Equality is one of those words, like justice, like freedom, which no one is against. But the invocation of the word “equality” often reduces analysis to empty platitudes. It is important to remember that in America we seek equality because it is a concomitant of freedom. When it is possible to accommodate both, that is what we should do.

The question has to be stated in the alternative because the California enactment has some peculiarities. Despite the state’s argument that if has a compelling interest in ensuring that all working women who desire prescription contraceptive coverage have that option available, the mandate is imposed only on employers that provide prescription coverage. Thus, Catholic Charities of Sacramento, Inc. (Catholic Charities), can choose either to provide contraceptives or not to provide prescription coverage to employees at all. This option would arguably make everyone worse off, but, in theory at least, equally so.

Earlier versions of the WCEA contained a broader concience clause—which Catholic Charities deemed acceptable—exempting bona fide religious employers and allowing religiously affilated hospitals, universities, and social service agencies to opt out. The current version of the act exempts churches, synagogues, mosques, temples, missions, parochial schools, seminaries and convents from the requirements to provede contraceptive coverage.

(See, e.g., Remarks of Sen. Speier, Sen. Floor Debate on Sen. Bill No. 41 (1999-2000 Reg. Sess.) Apr. 12, 1999, pp. 7-8 [floor statement of Senator Speier asserting that since 75 percent of all California Catholic hospitals already provide contraception coverage, the “issue has already been resolved . . . and its time has come”]; Remarks of Sen. Speier, Sen. Floor Debate on Assem. Bill No. 39 (1999-2000 Reg. Sess.) Sept. 7, 1999, p. 7 [floor statement of Senator Speier arguing that “59 percent of all Catholic women of childbearing age practice contraception [and] 88 percent of Catholics believe . . . that someone who practices artificial birth control can still be a good Catholic,” and commenting, “I agree with that. I think it’s time to do the right thing” (italics added)].)

This does not mean that the government may never limit what religious organizations can do. There are truly neutral laws which may be applied; there are aggressive interventions which are necessary to prevent harm. (See, e.g., Walker v. Superior Court (1988) 47 Cal.3d 112, 139 [ 253 Cal.Rptr. 1 , 763 P.2d 852 ] [finding Christian Scientist who did not seek medical treatment for her child liable for child’s death, notwithstanding the “religious infringement of significant dimensions,” since state’s interest is compelling and child endangerment statute is narrowly tailored]; Brady, Religious Organizations and Mandatory Collective Bargaining Under Federal and State Labor Laws: Freedom From and Freedom For, supra, 49 Vill. L.Rev. at p. 161 [“In rare cases, limitations on the freedom of religious organizations may be necessary. For example, if a religious group experiments with practices that endanger the lives of its employees or threaten them with serious bodily injury, interference may be justified”]; Laycock, Towards a General Theory of the Religion Clauses: The Case of Church Labor Relations and the Right to Church Autonomy (1981) 81 Colum. L.Rev. 1373, 1406 [“Courts have intervened to protect church members from serious bodily harm even when they voluntarily submitted”].) In contrast, what this case presents is essentially a clash of ideas.

These numbers are approximate. At oral argument, Catholic Charities counsel asserted that the Catholic Church employs fewer than 50,000 people, including those in holy orders. Proponents claim there are 52,000 employees in Catholic-affiliated hospitals alone. Using 60,000 as a point of reference, it appears all Catholic Church employees in California represent less than .5 percent of the California workforce, and female employees of the Catholic Church *582 represent about the same percentage of the number of working women of childbearing age in California. According to recent Bureau of Labor Statistics publications, the current number of California adults employed in nonfarm jobs is approximately 14.4 million. (Bur. of Lab. Statistics, U.S. Dept, of Lab. News Release No. 04-81 (Jan. 27, 2004) Employees on nonfarm payrolls by state and selected industry sector, table 5 <http://www.bls.gov/news.release/pdf/laus.-pdf> [as of Mar. 1, 2004].) A little less than half are women. Extrapolating from national statistics, around 5 million of that total will be women between 16 and 45. (Bur. of Lab. Statistics, U.S. Dept, of Lab. News Release No. 04-120 (Feb. 6, 2004) Selected employment indicators, table A-6 <http://www.bls.gov/newsrelease/pdf/empsit.pdf> [as of Mar. 1, 2004].) Even assuming these numbers need to be adjusted upward or downward for accuracy, an exemption for Catholic Charities would seem to have a negligible effect.

The majority cites language from United States v. Lee, supra, 455 U.S. 252 , for the proposition that allowing an employer to be exempt from a neutral law “operates to impose the employer’s religious faith on the employees.” (Id. at p. 261.) This is a curious statement. In Lee , both the employer and the employee were members of the Old Order Amish and all agreed they should be exempt from Social Security and unemployment insurance taxes. Even if that were not the case, it is not clear how an employer is in a position to impose anything on its employees to which they object. (U.S. Const., 13th Amend.. [prohibiting slavery or involuntary servitude].) Only the state, which holds the monopoly on coercive force, can compel adults to remain where they do not choose to be and do what they do not wish to do.

In Smith v. FEHC, supra, 12 Cal.4th 1143 , this court considered whether a state law prohibiting discrimination against unmarried cohabitating couples burdened the free exercise of a landlady who objected to renting to the couple on religious grounds. A majority of the court concluded Smith’s rights were not substantially burdened because she could simply abandon the rental business and redeploy her capital. If we reject the challenges of some religious claimants because they have other options, what logic compels us to assume that employees have no choice?

Even churches that do not operate schools, hospitals, or social service agencies would have trouble with the WCEA’s religious test. Not all religions proselytize. Those that do necessarily reach out to people who do not share their beliefs. Christian denominations, for example, are commanded to seek and save the lost. “Go ye into all the world and teach the gospel to every creature.” (Mark 15:15.) Catholic Charities suggest that some Catholic congregations might be “ineligible for the exemption depending . . . upon the demographics of a particular diocese, the' *584 fortuitous nature of hiring patterns, and the particular application of the theological criteria . . . .”