“The challenge posed ... is how to define the proper Establishment Clause limits on voluntary government efforts to facilitate the free exercise of religion.” Wallace v. Jaffree (1985) 472 U.S. 38, 82 [ 105 S.Ct. 2479, 2503 , 86 L.Ed.2d 29 ] (conc. opn. of O’Connor, J.).)

The freedom to exercise one’s religion without government prohibition is a cornerstone of our constitutional system. Motivated by the desire to give that freedom full scope in practice, legislatures sometimes grant religious adherents and institutions special relief from taxes and regulations. Legal efforts to assist religious institutions and promote their activities are also naturally popular with legislators because those organizations perform crucial charitable, educational, and community-building functions that enrich society generally. But however well motivated such laws may be, if they go too far they stand to run afoul of the simple, but fundamental, constitutional principle that government must remain neutral among religious groups and adherents, and between religious and nonreligious members of society—that legislation may not have the purpose or effect of advancing, promoting, or endorsing religion or any particular religion. (Board of Ed. of Kiryas Joel Village School Dist. v. Grumet (1994) 512 U.S. 687, 696 [ 114 S.Ct. 2481, 2487-2488 , 129 L.Ed.2d 546 ]; Lemon v. Kurtzman (1971) 403 U.S. 602, 612 [ 91 S.Ct. 2105, 2111 , 29 L.Ed.2d 745 ]; Walz v. Tax Commission (1970) 397 U.S. 664, 668-670 [ 90 S.Ct. 1409, 1411-1412 , 25 L.Ed.2d 697 ]; Everson v. Board of Education (1947) 330 U.S. 1, 15, 18 [ 67 S.Ct. 504, 511, 513 , 91 L.Ed. 711 , 168 A.L.R. 1392 ].)

Of course, not every law or government action that helps a religious organization violates the First Amendment’s bar on laws respecting the establishment of religion. The permissibility of two types of aid or relief is both logical and well established. First, government may provide assistance for organizations to conduct a nonreligious activity, offering help on an impartial basis to sectarian and secular groups alike, though religious organizations are thereby incidentally benefited. (Mitchell v. Helms (2000) 530 U.S. 793, 807-816, 857-858 [ 120 S.Ct. 2530, 2540-2544, 2567 , 147 L.Ed.2d 660 ]; California Educational Facilities Authority v. Priest (1974) 12 Cal.3d 593, 600-602 [ 116 Cal.Rptr. 361 , 526 P.2d 513 ].) Second, government may *728 target help to religious individuals and groups, in the form of relief from the burdens of generally applicable legislation, when application of the general law would significantly impair the person’s or group’s free exercise of religion. (Texas Monthly, Inc. v. Bullock (1989) 489 U.S. 1, 18-19, fn. 8 [ 109 S.Ct. 890, 901 , 103 L.Ed.2d 1 ] (plur. opn. of Brennan, J.) (Texas Monthly); Corporation of Presiding Bishop v. Amos (1987) 483 U.S. 327, 334-336 [ 107 S.Ct. 2862, 2867-2868 , 97 L.Ed.2d 273 ] (Corporation of Presiding Bishop); Wallace v. Jaffree, supra, 472 U.S. at pp. 82-83 [ 105 S.Ct. at p. 2503 ] (conc, opn. of O’Connor, J.).)

The laws at issue here, sections 25373, subdivision (d) and 37361, subdivision (c) of the Government Code (hereafter sections 25373(d) and 37361(c)), cannot be defended on either basis. Their assistance is expressly restricted to “religiously affiliated” organizations {ibid.); those entities are granted the power (unusual, to say the least) to exempt themselves from certain land use regulations, while no comparable relief is provided to nonprofit secular organizations whose noncommercial activities may be similarly affected by the same regulations. Clearly, therefore, these provisions cannot be justified as an impartial grant of nonreligious assistance.

Nor, in light of their unusual character and extraordinary breadth, can the laws fairly be characterized as a mere attempt to prevent government interference with religious practices. Under sections 25373(d) and 37361(c), a religiously affiliated organization may exempt itself from a landmark regulation not only when the regulation would interfere with its religious mission, but also when the regulation would have no significant effect on the exercise of religion but would, according to the organization’s own declaration, deprive it of “economic return” on, or the “reasonable use” of, its property. (§§ 25373(d)(2), 37361(c)(2).) Perhaps most extraordinarily, rather than merely permitting individual religious organizations to obtain exemptions from the landmark laws by application and with a showing that the group’s religious practices will otherwise be impaired, the statutes permit a religious group to exempt itself, without any actual showing of need, any assurance that the exempted property is or will remain in religious use, or any governmental review of the self-declared hardship exemption.

The majority responds to these obvious defects in the legislation with the mantra that sections 25373(d) and 37361(c) do not assist religious groups in any way, but merely accommodate their religious practices by removing a government-imposed burden. In the words of the Court of Appeal, “[t]he state has not assisted religious organizations but has merely stepped out of their way.” Repetition of a facile formula, however, cannot substitute for *729 constitutional scrutiny. When the state removes legal barriers from one landowner’s use and exploitation of its property and not from another’s, has not the state effectively assisted the former and advanced its mission, both absolutely and as compared to the latter? This type of preferential treatment would be constitutionally permissible if justified by a concrete need to avoid interference with religious practice; absent that justification, the state’s decision to “step out of the way” only of religious organizations must be seen by an objective observer as an endorsement of religious enterprises, violating the fundamental principle of government neutrality in matters of religion. (Texas Monthly, supra, 489 U.S. at p. 15 [109 S.Ct. at pp. 899-900].)

As Justice O’Connor has observed, the distinction between government’s advancing religion and government’s taking measures that allow religion to better advance itself “seems ... to obscure far more than to enlighten. Almost any government benefit to religion could be recharacterized as simply ‘allowing’ a religion to better advance itself, unless perhaps it involved actual proselytization by government agents.” (Corporation of Presiding Bishop, supra, 483 U.S. atp. 347 [ 107 S.Ct. at p. 2874 ] (cone. opn. of O’Connor, J.).) Particularly in the area of urban land use, where pervasive legal regulation forms the background for private action, a special exemption from regulation constitutes a benefit to the property owner, both absolutely and relative to others in the marketplace. To call such an exemption an “accommodation” of religion adds nothing to the analysis unless—as is not the case here—the exemption is granted to avoid a demonstrable interference with worship or other religious practices. A law that singles out religious organizations alone for exemption from regulation, without requiring any showing that, absent the exemption, religious freedom will be impaired, has, whatever label it wears, “crossed the Une from permissible accommodation to unconstitutional establishment.” (Lee v. Weisman (1992) 505 U.S. 577, 629 [ 112 S.Ct. 2649, 2677 , 120 L.Ed.2d 467 ] (conc. opn. of Souter, J.).)

The majority nonetheless regards as justified the blanket grant to religious landowners of self-exemption power, because “the Legislature could reasonably believe that the restrictions accompanying designation as a historical landmark . . . may burden the ability of a religious entity ... to carry out its religious mission.” (Maj. opn., ante, at pp. 712-713.) In fact, the case law does not support this kind of deference to presumed legislative beliefs. (See Texas Monthly, supra, 489 U.S. at p. 18 & fn. 8 [ 109 S.Ct. at p. 901 ] [blanket sales tax exemption for religious periodicals not justified as lifting burden on exercise of religion where “[n]o concrete need to accommodate religious activity has been shown” and exemption therefore “does not remove a *730 demonstrated and possibly grave imposition on religious activity”]; Corporation of Presiding Bishop, supra, 483 U.S. at p. 348 [ 107 S.Ct. at p. 2875 ] (conc. opn. of O’Connor, J.) [“Of course, in order to perceive the government action as a permissible accommodation of religion, there must in fact be an identifiable burden on the exercise of religion that can be said to be lifted by the government action”]; Duffy v. State Personnel Bd. (1991) 232 Cal.App.3d 1, 12 [ 283 Cal.Rptr. 622 ] [adopting and applying Justice O’Connor’s standard].)

The majority thus errs in its deference to the Legislature on this point, for “judicial deference to all legislation that purports to facilitate the free exercise of religion would completely vitiate the Establishment Clause.” (Wallace v. Jaffree, supra, 472 U.S. at p. 82 [ 105 S.Ct. at p. 2503 ] (cone, opn. of O’Connor, J.).) Perhaps even more fundamental a flaw in the majority’s approach, however, is the extraordinary and unjustifiable breadth of the presumed finding to which the majority defers. To reiterate, sections 25373(d) and 37361(c) do not require any neutral government body, such as a zoning or planning board, to determine the organization’s need for an exemption; rather, the laws allow the organization to grant itself the exemption upon making certain unreviewable “determinations.” Those determinations, moreover, need not even include a claim that the landmark regulation will impede the organization’s religious practice; instead, the organization may simply find its economic use of the property will be impaired. To hold, then, that these statutes, in the legislative view, serve primarily to prevent government interference with religious practice, we would need to ascribe to the Legislature the finding that all religious groups owning a landmarked noncommercial property will be impeded in their religious missions unless they are given the power to exempt themselves from landmark laws by making an unreviewable claim that the regulation would make their use of the property uneconomical, whether or not the property has been or will be used for purposes the group itself defines as religious. Even were we to ascribe such an absurd belief to the Legislature, I submit, it would not be entitled to any deference even under the majority’s standard of deference to “reasonable” legislative judgments.

I therefore believe the challenged provisions, granting to religious entities alone the power to exempt themselves from economically burdensome land use regulations, without requiring a showing in a given case that application of the regulations would impede any religious practice, go far beyond a reasonable accommodation of the exercise of religion and, as a practical matter, grant a significant, unjustified and preferential benefit to religious organizations. As will be demonstrated in more detail below, such legislation *731 cannot be enforced consistently with the establishment clause of the federal Constitution’s First 1

Discussion

Under the analysis articulated in Lemon v. Kurtzman, supra, 403 U.S. at pages 612-613 [ 91 S.Ct. at page 2111 ], a statute providing aid to religious individuals or organizations violates the establishment clause unless it has a secular legislative purpose, has as its primary effect neither the advancement nor the suppression of religion, and does not foster excessive entanglement of government with religion. The Lemon test, and especially its first two parts, embodies the core principle of the clause: government neutrality among religious creeds, and between religious and nonreligious viewpoints and organizations. (See, e.g., Texas Monthly, supra, 489 U.S. at pp. 8-9 [ 109 S.Ct. at p. 896 ].) Although this opinion does not follow Lemon's organization, which I do not take to be talismanic, it does attempt to determine whether sections 25373(d) and 37361(c) are neutral in design and consequence, and whether they create an excessive entanglement by improperly delegating governmental functions to private religious organizations.

Justice Souter recently observed that thp term “neutrality” has borne at least three distinct meanings in establishment clause debate. It has been used to denote the required position of government as neither promoting nor impeding religion; to describe government assistance that is not inherently religious in its content or character; and to describe the generality, of evenhandedness, with which some item of government service or assistance is given to secular and sectarian organizations alike. (Mitchell v. Helms, supra, 530 U.S. at pp. 878-882 [120 S.Ct. at pp. 2578-2580] (dis. opn. of Souter, J.).) We are concerned here with the first and third meanings.

In part I of the discussion, I will focus on whether the self-exemption provisions of sections 25373(d) and 37361(c) may be considered neutral in Justice Souter’s first sense—as merely accommodating, but not advancing or promoting, religion; I conclude they may not, because the assistance they provide goes far beyond any relief that might be needed to prevent landmark regulation from interfering with free exercise. Part II of the discussion examines the last sense, or dimension, of neutrality identified by Justice Souter and argues that, in their unnecessary lack of evenhandedness or *732 generality, sections 25373(d) and 37361(c) demonstrate nonneutrality in this third sense as well.

I. Accommodation, Advancement and Endorsement

A. The Role of Burden

“Whatever else may define the scope of accommodation permissible under the Establishment Clause, one requirement is clear: accommodation must lift a discernible burden on the free exercise of religion.” (Lee v. Weisman, supra, 505 U.S. at p. 629 [ 112 S.Ct. at p. 2677 ] (conc. opn. of Souter, J.).)

Laws “accommodating” religious freedom have been viewed as having a secular goal for purposes of the Lemon test. (See Corporation of Presiding Bishop, supra, 483 U.S. at p. 335 [ 107 S.Ct. at p. 2868 ].) With good reason, Justice O’Connor has declined to embrace this characterization, or the application of the Lemon test to accommodation cases generally. She regards such laws as having a religious purpose and effect but as being nonetheless permissible if they do not go so far as to convey a message of endorsement. (See 483 U.S. at pp. 346-348 [107 S.Ct. at pp 2874-2875] (cone. opn. of O’Connor, J.).) As we shall see below, under either view “accommodation” justifies exemptions targeted to religious organizations only when the law from which an exemption is made would otherwise interfere significantly with some religious activity.

Thus, in Corporation of Presiding Bishop, supra, 483 U.S. 327 , the majority approved an exemption for religious organizations from title VII’s prohibition on religious discrimination, on the ground it served to “alleviate significant governmental interference with the ability of religious organizations to define and carry out their religious missions.” (Id. at p. 335 [ 107 S.Ct. at p. 2868 ].) Justice O’Connor, concurring, stressed that “in order to perceive the government action as a permissible accommodation of religion, there must in fact be an identifiable burden on the exercise of religion that can be said to'be lifted by the government action.” (Id. at p. 348 [ 107 S.Ct. at p. 2875 ] (cone. opn. of O’Connor, J.).) She concurred in the judgment because the exemption, designed to lift from religious organizations the “burden of refraining from discriminating on the basis of religion,” would be perceived by an “objective observer ... as an accommodation of the exercise of religion rather than as a Government endorsement of religion.” (Id. at p. 349 [ 107 S.Ct. at p. 2875 ] (cone. opn. of O’Connor, J.).)

In Wallace v. Jaffree, supra, 472 U.S. 38 , conversely, the court struck down a state statute authorizing public schools to have a period of silence for *733 “ ‘meditation or voluntary prayer.’ ” (Id. at pp. 41, 61 [105 S.Ct. at pp. 2482, 2492].) Rejecting the state’s characterization of its law as “ ‘a means for accommodating the religious and meditative needs of students,’ ” the majority reasoned that at the time of the statute’s enactment “there was no governmental practice impeding students from silently praying for one minute at the beginning of each schoolday; thus, there was no need to ‘accommodate’ . . . .” (Id. at p. 58, fn. 45 [ 105 S.Ct. at p. 2490 ].) Justice O’Connor concurred here too, agreeing that the statute “lifts no state-imposed burden on the free exercise of religion, and accordingly cannot properly be viewed as an accommodation statute.” (Id. at p. 84 [ 105 S.Ct. at p. 2504 ] (cone. opn. of O’Connor, J.).)

Similarly, in Texas Monthly, the high court held that a sales tax exemption targeted at religious publications violated the establishment clause, in part because the exemption “does not remove a demonstrated and possibly grave imposition on religious activity.” (Texas Monthly, supra, 489 U.S. at p. 19, fn. 8 [ 109 S.Ct. at p. 901 ] (plur. opn. of Brennan, J.).) The Texas exemption, applicable to periodicals and books promulgating the teachings of a religious faith, and exclusively to such publications, “cannot reasonably be seen as removing a significant state-imposed deterrent to the free exercise of religion”; in that and other respects, the law conveyed a message of endorsement forbidden under the establishment clause. (Id. at p. 15 [ 109 S.Ct. at p. 899 ].)

The high court’s holdings, therefore, do not support the view of the majority that absence of a significant burden on free exercise does not matter when government chooses only to grant an exemption from, otherwise applicable taxes or regulations.

The majority seeks support for its view that exemptions are inherently immune from advancement challenges in Justice White’s opinion for the court in Corporation of Presiding Bishop, supra, 483 U.S. 327 . Justice White wrote that “[f]or a law to have forbidden ‘effects’ under Lemon , it must be fair to say that the government itself has advanced religion through its own activities and influence.” (Id. at p. 337 [ 107 S.Ct. at p. 2869 ].) In the case at bench, he continued, “we find no persuasive evidence . . . that the Church’s ability to propagate its religious doctrine ... is any greater now than it was prior to the passage of the Civil Rights Act in 1964.” (Ibid.)

Corporation of Presiding Bishop cannot reasonably be read as approving all exemptions favoring religious organizations regardless of whether they actually serve to alleviate substantial burdens on free exercise. There the *734 general law at issue, title VII’s prohibition on employment discrimination based on religion, did threaten “significant governmental interference with the ability of religious organizations to define and carry out their religious missions” (Corporation of Presiding Bishop, supra, 483 U.S. at p. 335 [ 107 S.Ct. at p. 2868 ]), and the majority opinion stands for nothing more than that government does not improperly advance religion in seeking to alleviate such interference. To th.e extent Justice White’s opinion nevertheless might be read as establishing a per se rule equating exemptions to permissible accommodations, the court implicitly repudiated it two terms later in Texas Monthly. In that case, religious organizations were no better able to advance their publishing missions after the tax exemption at issue than they were before the sales tax itself was enacted, yet the court declined to characterize the exemption as a mere accommodation of religious freedom, instead finding the exemption impermissibly favored religious publications. (Texas Monthly, supra, 489 U.S. at p. 15 [109 S.Ct. at pp. 899-900]; id. at p. 28 [109 S.Ct. at pp. 906-907] (cone. opn. of Blackmun, J.).)

Because, as Justice O’Connor points out, most government assistance to religion could be characterized as a mere accommodation, the test of whether an exemption goes too far cannot be a simply formal one: it cannot be enough to observe that religious organizations granted exemption from an inconvenient or costly general law are no better off than they would have been had the general law not been enacted. Rather than rely on formalism, we must examine a special exemption for religious entities in context to determine whether it is designed and will serve to accommodate free exercise or, instead, to differentially promote and endorse religion or any particular religion. “To ascertain whether the statute conveys a message of endorsement, the relevant issue is how it would be perceived by an objective observer, acquainted with the text, legislative history, and implementation of the statute.” (Corporation of Presiding Bishop, supra, 483 U.S. at p. 348 [ 107 S.Ct. at p. 2875 ] (cone. opn. of O’Connor, J.).) Crucial to the analysis is whether or not the law was designed to lift a discemable significant burden on free exercise, for, where there is no such burden, the law “ ‘provide [s] unjustifiable awards of assistance to religious organizations’ and cannot but ‘conve[y] a message-of endorsement’ to slighted members of the community.” (Te xas Monthly, supra, 489 U.S. at p. 15 [ 109 S.Ct. at p. 899 ], quoting Corporation of Presiding Bishop, supra, 483 U.S. at p. 348 [107 S.Ct. at pp. 2874-2875] (conc. opn. of O’Connor, J.).) I believe an objective observer would see the statutes at issue here as going well beyond accommodation into the prohibited realm of preferential support and endorsement.

*735 B. Landmark Designation and Religious Freedom

Sections 37361 and 25373 authorize cities and counties, respectively, to create and apply reasonable land use controls to protect the “special character or special historical or aesthetic interest or value” of places and buildings. More will be said later in this opinion about the value of preserving landmarks; suffice it here to note that landmark regulation has in common with much zoning and other land use regulation the “entirely permissible governmental goal” of “enhanc[ing] the quality of life by preserving the character and desirable aesthetic features of a city.” (Penn Central Transp. Co. v. New York City (1978) 438 U.S. 104, 129 [ 98 S.Ct. 2646, 2661 , 57 L.Ed.2d 631 ].)

Localities’ efforts to shape and preserve the physical community through land use regulation do impose costs on property owners—in time, money and lost economic opportunities. As applied to religious individuals and institutions owning property, however, the incidental costs of such broadly applicable regulations have not generally been deemed constitutionally significant burdens on free exercise. Whether assessed under the lenient federal free exercise regime established in Employment Div., Ore. Dept, of Human Res. v. Smith (1990) 494 U.S. 872 [ 110 S.Ct. 1595 , 108 L.Ed.2d 876 ] (Smith), 2 or under more stringent state or pre-Smith federal law, neutral land use regulations that merely add inconvenience or cost to a person’s or group’s religious practices do not unconstitutionally restrict freedom of religion.

As the United States Supreme Court recently explained in City of Boerne v. Flores (1997) 521 U.S. 507 [ 117 S.Ct. 2157 , 138 L.Ed.2d 624 ], rejecting a free exercise challenge to the impact of landmark restrictions on a church, “It is a reality of the modem regulatory state that numerous state laws, such as the zoning regulations at issue here, impose a substantial burden on a large class of individuals. When the exercise of religion has been burdened in an incidental way by a law of general application, it does not follow that the persons affected have been burdened any more than other citizens, let alone burdened because of their religious beliefs.” (Id. at p. 535 [ 117 S.Ct. at *736 p. 2171 ].) 3 City of Boerne , and the other cases cited in footnote 3, fit the general pattern, apparent even under pr e-Smith law, of high court decisions holding that religion-neutral government regulations that incidentally burden religious individuals or groups by making their practices more costly or inconvenient do not thereby infringe adherents’ freedom of religion. 4

I recognize that “ ‘[t]he limits of permissible state accommodation to religion are by no means co-extensive with the noninterference mandated by the Free Exercise Clause.’” (Corporation of Presiding Bishop, supra, 483 U.S. at p. 334 [ 107 S.Ct. at p. 2867 ], quoting Walz v. Tax Commission, supra, *737 397 U.S. at p. 673 [90 S.Ct. at pp. 1413-1414].) That religious individuals and groups may generally be required to bear the economic cost and inconvenience of compliance with neutral laws without violation of the free exercise clause does not necessarily imply that no relief from such costs is permissible under the establishment clause. Nonetheless, the two are related; the absence of even an arguably substantial burden on religious practices eliminates the chief justification for targeted religious exemptions. As one noted scholar of the religion clauses, George Washington University’s Ira Lupu, has written, “One of the strongest arguments for permissive accommodations is that they might be mandatory after all. In a world with no mandatory accommodations [i.e., after Smith], this argument disappears. Government may burden religion incidentally, and may, by way of exemptions, seek to lift such a burden; as long as such burdens do not violate the Constitution, however, the case for lifting them from religious entities but not from those engaged in counterpart secular activity remains hard to fathom.” (Lupu, The Trouble with Accommodation (1992) 60 Geo. Wash. L.Rev. 743, 771, fn. omitted.) Where a neutral law, such as a landmark regulation, imposes costs on a variety of individuals and organizations, but the costs to religious adherents do not rise to a level even arguably cognizable under free exercise principles, a legislative decision to target relief solely to religious adherents becomes difficult to justify as an accommodation of religious liberty. 5

If the incidental costs of compliance with land use laws, including landmark restrictions, do not generally constitute substantial interference with free exercise justifying specially targeted exemptions, are there nonetheless some costs or impacts from landmarking that do threaten free exercise and call for accommodation? Courts and commentators agree that such conflicts between landmark laws and free exercise can occur, but the occasions are limited to a few types of circumstances and, in practice, have proven rare.

First, application of a landmark regulation to control the arrangement or appearance of the interior of a house of worship, or exterior features with *738 religious significance, poses a significant threat of government interference in religious practices. (See Society of Jesus v. Boston Landmarks (1990) 409 Mass. 38 [ 564 N.E.2d 571, 573 ] [designation of church interior held impermissible restraint on freedom of worship under state Constitution].) Second, a significant burden would exist “[w]here a church can prove that unyielding enforcement of a landmark ordinance will result in a forced cessation of religious worship and practice in the landmarked building . . . .” (Weinstein, The Myth of Ministry vs. Mortar: A Legal and Policy Analysis of Landmark Designation of Religious Institutions (1992) 65 Temp. L.Rev. 91, 151 (hereafter Weinstein); see St. Bartholomew’s Church v. City of New York, supra, 914 F.2d at pp. 355-356 [suggesting landmark restriction that prevented church from “continuing] its religious practice in its existing facilities” would be impermissible burden].)

Beyond these areas, however, the incidental costs and physical restrictions on repair and maintenance of property exteriors, the administrative cost and inconvenience of complying with landmark regulation, and the opportunity cost imposed by restraints on demolition of landmarked properties for commercial redevelopment are not constitutionally significant burdens on religious freedom. (Weinstein, supra, 65 Temp. L.Rev. at pp. 148-152; Nunez & Sidman, California’s Statutory Exemption for Religious Properties from Landmark Ordinances: A Constitutional and Policy Analysis (1995-1996) 12 J.L. & Religion 271 , 289-290 (hereafter Nunez & Sidman); see St. Bartholomew’s Church v. City of New York, supra, 914 F.2d at p. 355 [opportunity cost not constitutionally significant]; Open Door Baptist Church v. Clark County, supra, 995 P.2d at pp. 42-43 [same as to application fee for conditional use permit].) In Weinstein’s summary, “[i]t seems clear that interior designation, and exterior designation that would constrain the ‘theological aspects of building design’ or have the effect of forcing a church to cease religious worship at a given site because of physical or financial exigency, would constitute a burden. Conversely, the denial of permission for commercial development would not appear to constitute a burden.” (Weinstein, supra, 65 Temp. L.Rev. at p. 148.)

Actual conflicts between religious practice and landmark restrictions arise infrequently. One survey of the reported cases of asserted conflict found that “most controversies involving historic preservation and religious organizations have little to do with the free exercise of religious beliefs. Instead, the arguments raised by religious institutions as property owners are more aptly viewed as secular in nature. For example, these owners often argue that existing facilities need expansion, maintenance and repair costs are too high, local land use regulation interferes with the opportunity to attain the ‘highest *739 and best’ (i.e., most profitable) use of the property, government red-tape is too burdensome, or the need for additional parking requires the demolition of adjacent buildings. All of these arguments are directly analogous to arguments raised by secular property owners in land use disputes, and are generally rejected by the courts.” (Nelson, Remove Not the Ancient Landmark: Legal Protection for Historic Religious Properties in an Age of Religious Freedom Legislation (1999) 21 Cardozo L.Rev. 721, 729-730, fns. omitted.)

Not surprisingly, in light of both the rarity of situations in which land-marking actually restricts religious freedom and the flexibility built into most landmark laws, even cities with large numbers of landmarked religious buildings have seen few conflicts that could not readily be resolved administratively. In Philadelphia, with 139 landmarked religious properties, a 1989 study found religious institutions had applied for 127 building permits under the landmark ordinance’s standards. Only one was denied, and that application was subsequently withdrawn. Most of the 126 approvals came within one week of application. (Weinstein, supra, 65 Temp. L.Rev. at p. 111.) In New York City, with 211 religious institutions designated as landmarks or in historic districts, a 1991 study found all but nine of 423 applications for permits were approved. Two were partially approved, and two more were resolved by approval of subsequent proposals. (Id. at p. 112.)

With this background understanding of the limited constitutional conflict between landmark laws and religious freedom, we can examine the text and history of the California exemption to see if it is designed as a reasonable accommodation of religion or as an unjustified and preferential benefit to religious organizations.

C. Text and History of California’s Religious Entities Exemption

Section 25373(d) provides that counties’ authority to control the use and appearance of historic properties, granted in subdivision (b) of the same section, “shall not apply to noncommercial property owned by any association or corporation that is religiously affiliated and not organized for private profit, whether the corporation is organized as a religious corporation, or as a public benefit corporation, provided that both of the following occur: [^Q (1) The association or corporation objects to the application of the subdivision to its property. [<[] (2) The association or corporation determines in a public forum that it will suffer substantial hardship, which is likely to deprive the association or corporation of economic return on its property, the reasonable use of its property, or the appropriate use of its property in the *740 furtherance of its religious mission, if the application is approved.” Section 37361(c) places the same restriction on cities’ historic preservation authority.

Although the statutes require a self-determination of hardship to be made in a “public forum,” the nature of that forum is not further specified. (§§ 25373(d)(2), 37361(c)(2).) No application for exemption need be made to a governmental body. Nor does either statute provide for any governmental or judicial review of the religious entity’s hardship declaration; indeed, both statutes expressly disavow any authorization for local legislative bodies to “override the determination made pursuant to paragraph (2).” (Gov. Code, §§ 25373, subd. (e), 37361, subd. (d).) The laws, therefore, must realistically be seen as blanket exemptions for noncommercial properties owned by religious entities or, more precisely, as blanket grants to such organizations of the power to exempt their own noncommercial properties from local historic preservation regulations. 6

Neither statute defines “noncommercial,” but the opposite term, “commercial,” is a familiar one in land use law. As commonly employed in zoning regulations, “commercial” denotes a category of real property use, standing in distinction to such other common use zones as residential, industrial, and agricultural. (See, e.g., Williams, Cal. Zoning Practice (Cont.Ed.Bar 2000 supp.) §§ 6.2-6.3, pp.. 287-295 [zone definition charts for the City of Los Angeles].) “Noncommercial property” would thus appear to include property used for residential (including multiple-unit rentals), industrial, and agricultural purposes. Moreover, while the property must be owned by a religiously affiliated entity, nothing in the statutes requires that it be in use by the organization; nor do the statutes exclude property leased to others. The exemption coiild therefore be applied to property owned by a church but leased to a nonreligious individual or corporation for industrial or residential use.

Whether or not the property is being used for worship or other religious practices at the time of the self-declared exemption, nothing in the statutes requires that it be so used after exemption, or even that it remain in the religious entity’s ownership. Under sections 25373(d) and 37361(c), then, a religiously affiliated organization could, for example, declare exempt a historic building still suitable for and in use as a house of worship, tear it down, and build a modem office building for sale on the commercial market.

*741 Finally, even when the building is in use for worship or other religious practice, and the organization intends to continue that use, the statutes call for only an unreviewable declaration of economic impact. Thus, the exemption could be invoked where application of the landmark regulation would impose only incidental costs and inconvenience on the religious entity and would not prevent the property’s continued religious use. Under sections 25373(d) and 37361(c), a religiously affiliated organization that simply did not want to go through the permit process to make minor changes to a building, or wished to save some money by using inexpensive but anachronistic materials to rebuild an exterior feature, could exempt itself from the landmark regulation process despite the lack of any demonstrable threat to freedom of religion.

The permanent provisions in sections 25373 and 37361 allowing religious entities self-exemption were added in 1994, by Assembly Bill No. 133 (1993-1994 Reg. Sess.). (Stats. 1994, ch. 1199, §§ 1, 2 (Assembly Bill No. 133).) The history of these amendments bears out what the text indicates, that they were designed to favor religious organizations with regulatory relief much broader than necessary for protection of religious freedom.

“The Archdiocese of San Francisco, as part of a cost-cutting move, planned to close nine parish churches.” (Nunez & Sidman, supra, 12 J.L. & Religion at p. 275.) The archdiocese, facing millions of dollars in seismic retrofitting costs as well as declining attendance in some parish churches, had systematically evaluated the market value and “highest and best use” of its properties, and decided on a consolidation plan. (Fernandez et al., Church for Sale?, S.F. Examiner (Aug. 3, 1995) p. Al.) However, the archbishop’s “controversial plan to consolidate the city’s 54 parishes . . . sparked a firestorm of protest in churches across the city” (Lattin, Altars in Escrow, S.F. Chronicle (Mar. 2, 1997) p. 1) and “ ‘arousfed] preservationists’ fears that [the churches might] be sold, demolished, or remodeled’ ” (Comment, Holy War: In the Name of Religious Freedom, California Exempts Churches from Historic Preservation (1996) 37 Santa Clara L.Rev. 213, 229, fn. omitted). “In response to the San Francisco Catholic Archbishop’s desire to close and demolish churches, [Assembly] Speaker [Willie] Brown amended AB 133 . . .so that cities and counties could not apply special conditions or regulations to protect historic properties owned by religious organizations.” (Assem. Com. on Local Gov., Concurrence in Sen. Amends., Assem. Bill No. 133 (1993-1994 Reg. Sess.) as amended Aug. 26, 1994, p. 2.) The bill thus had its genesis in a specific conflict between the San Francisco Catholic Archdiocese, and parishioners and preservationists in San Francisco, although ultimately wider support came from a number of San Francisco *742 churches and church groups from around the state, and wider opposition from an array of historic preservation organizations and local governments. (Sen. Rules Com., Off. of Sen. Floor Analyses, 3d reading analysis of Assem. Bill No. 133 (1993-1994 Reg. Sess.) as amended Aug. 26, 1994, pp. 2-3.) 7

As this history and the law’s text show, the amendments to sections 25373 and 37361 were designed to do much more than protect religious adherents’ right to worship and practice their religions as they choose in historic buildings, dr to remodel the interior of such structures for a different use, retrofit historic buildings for earthquake safety, or abandon historic houses of worship that, for physical or financial reasons, can no longer be used for that purpose. (Indeed, San Francisco’s landmark ordinance does not restrict alterations to the interiors of privately owned structures or prohibit the demolition of landmarked properties; nor is any special approval required for ordinary maintenance and repairs, or for work needed to comply with seismic retrofit regulations, or for work needed to correct an unsafe condition. (S.F. Planning Code, §§ 1005, subd. (e)(2)-(4), 1006.6, subd. (b), 1007.) 8 Rather, the amendments’ design and effect were to award religious organizations the unique ability to manipulate their property holdings without regard for landmark regulations, by altering, demolishing, and selling *743 properties, if they choose, for maximum economic benefit and without the incidental cost and delay attendant on obtaining special permits.

D. Unrestricted Self-exemption Is Not a Reasonable Measure for Accommodation of Religious Freedom

The Legislature could have responded to the San Francisco controversy over church closings, and any similar problems in other cities, by requiring that all local landmark laws have hardship provisions permitting approval for alteration of a historic building when necessary to continued use or rehabilitation. (See, e.g., San Jose Mun. Code, § 13.48.260 [permit for noncomplying work may be issued if denial “would cause immediate and substantial hardship on the applicant because rehabilitation in accordance with the chapter is infeasible from a technical, mechanical, or structural standpoint, or if the economics of rehabilitation in accordance with this chapter would require an unreasonable expenditure in light of the feasible uses of such property”]; Berkeley Mun. Code, § 3.24.270 [same, “if the applicant presents clear and convincing evidence to the commission that such disapproval will work immediate and substantial hardship because of conditions peculiar to the particular structure or feature involved, and that failure to disapprove the application will be consistent with the purposes of this chapter”].) Alternatively, the Legislature might have mandated a hardship exemption particularly tailored to religious and charitable institutions. Columbus, Ohio, for example, provides relief to any nonprofit organization that can show “it *744 is infeasible to financially or physically achieve its charitable purposes while conforming to the pertinent architectural standards and guidelines.” (Columbus City Code, tit. 31, ch. 3116.16, subd. (3) (enacted 1989), quoted in Weinstein, supra, 65 Temp. L.Rev. at p. 105 [also discussing targeted hardship provisions in both New York City and Buffalo, New York, landmark ordinances].)

Instead, the Legislature created an exemption for religious organizations only, and made the exemption available through the organization’s own, unreviewable, declaration of hardship. Self-exemption was not necessary either to protect religious freedom, as we have seen, or to guard against government entanglement in religious affairs in the permit process itself. A true hardship exemption, available on application and based on reviewable agency findings, would not involve the administering agency in any intrusive inquiry into religious affairs or otherwise entangle the local government in religion; the question before the planning commission or other board deciding such an application is not whether the organization’s use of the property is truly religious—that should be left to the sincerely held tenets of the organization itself—but only whether the physical and economic circumstances justify a variance from the ordinary requirements of historic preservation. 9

Not only is self-exemption unjustified by any need to prevent government interference with religious affairs, but by effectively abdicating exercise of a government function and delegating it to a private religious group, the self-exemption feature of sections 25373(d) and 37361(c) actually tends to create a “forbidden ‘fusion of governmental and religious functions.’ ” (Board of Ed. of Kiry as Joel Village School Dist. v. Grumet, supra, 512 U.S. at p. 702 [ 114 S.Ct. at p. 2490 ].) As with New York’s delegation of school district powers to a religious group, held unconstitutional in Grumet , here the *745 state has effectively placed the government power to determine applicability of land use laws in the hands of groups “chosen according to a religious criterion.” (Id. at p. 698 [ 114 S.Ct. at p. 2488 ].) Like Massachusetts’s delegation to churches of veto power over nearby liquor licenses, held unconstitutional in Larkin v. Grendel’s Den, Inc. (1982) 459 U.S. 116 [ 103 S.Ct. 505 , 74 L.Ed.2d 297 ], our self-exemption provision “substitutes the unilateral and absolute power of a church for the reasoned decisionmaking of a public legislative body acting on evidence and guided by standards, on issues with significant economic and political implications.” (Id. at p. 127 [ 103 S.Ct. at p. 512 ].)

An unnecessary blanket exemption for religious property owners from the constraints of landmark regulation, moreover, actually promotes the missions of religious organizations relative to the sometimes competing missions of secular educational and persuasive organizations. “Exempting religious institutions from landmark designation creates the potential for significantly advancing religious ideas over competing secular ideas. If St. Bart’s [the plaintiff church in St. Bartholomew’s Church v. City of New York, supra, 914 F.2d 348 ] is free to reap millions of dollars from the commercial development of its property and then apply those funds to support its religious and charitable programs, but secular charitable institutions must comply with the landmark ordinance and so are denied access to funds derived from property development, then religious institutions and their ideas are given a significant advantage by government action. Denying government the right to prefer religion over secularism lies at the core of the Lemon test . . . .” (Weinstein, supra, 65 Temp. L.Rev. at p. 157, fn. 405.)

An objective observer familiar with the text, history, and social context of the self-exemption power granted in sections 25373(d) and 37361(c) would thus perceive it not as a reasonable attempt to prevent local government interference in religious affairs or to lift a significant burden on religious liberty, but as the unjustifiably broad award, to religiously affiliated property owners, of a unique and valuable privilege—the privilege of determining their own properties’ subjection to generally applicable land use laws. Such favoritism in the political distribution of valuable privileges “cannot but ‘conve[y] a message of endorsement’ to slighted members of the community.” (Texas Monthly, supra, 489 U.S. at p. 15 [ 109 S.Ct. at p. 899 ].) The state’s relative endorsement of religious organizations’ missions “is impermissible because it sends the ancillary message to . . . nonadherents ‘that they are outsiders, not full members of the political community, and an accompanying message to adherents that they are insiders, favored members *746 of the political community.’ ” (Santa Fe Independent School Dist. v. Doe (2000) 530 U.S. 290, 309-310 [ 120 S.Ct. 2266, 2279 , 147 L.Ed.2d 295 ].) 10

II. Generality of Benefits

A. The Role of Evenhandedness

Even when the government relief accorded religious institutions cannot be characterized as an accommodation of free exercise, the proper governmental stance of neutrality toward religion may manifest itself in an evenhanded provision of benefits to religious and secular institutions alike. This is true not only when the scope of benefits approaches universality (see Everson v. Board of Education, supra, 330 U.S. at pp. 17-18 [67 S.Ct. at pp. 512-513] [citing such general government services as police and fire protection and sewage disposal]), but also, as seen below, when religious institutions are included in a limited set of recipients rationally defined by a secular governmental purpose.

In Walz v. Tax Commission, supra, 397 U.S. 664 (Walz), the high court ■ considered a New York law exempting from property taxation corporations and associations “ ‘organized exclusively for the moral or mental improvement of men and women, or for religious, bible, tract, charitable, benevolent, missionary, hospital, infirmary, educational, public playground, scientific, literary, bar association, medical society, library, patriotic, historical or cemetery purposes.’ ” (Id. at p. 667, fn. 1 [ 90 S.Ct. at p. 1410 ].) In rejecting an establishment clause challenge to the exemption, the court relied prominently on the generality with which it had been offered: “The legislative purpose of the property tax exemption is neither the advancement nor the inhibition of religion; it is neither sponsorship nor hostility. New York, in common with the other States, has determined that certain entities that exist in a harmonious relationship to the community at large, and that foster its ‘moral or mental improvement,’ should not be inhibited in their activities by property taxation or the hazard of loss of those properties for nonpayment of taxes. It has not singled out one particular church or religious group or even *747 churches as such; rather, it has granted exemption to all houses of religious worship within a broad class of property owned by nonprofit, quasi-public corporations which include hospitals, libraries, playgrounds, scientific, professional, historical, and patriotic groups. The State has an affirmative policy that considers these groups as beneficial and stabilizing influences in community life and finds this classification useful, desirable, and in the public interest.” (Id. at pp. 672-673 [ 90 S.Ct. at p. 1413 ].)

In Texas Monthly, supra, 489 U.S. 1 , again facing the question whether a tax exemption benefiting religious organizations violated the establishment clause, the court distinguished the evenhanded exemption of Walz from the narrow religious benefit provided by the Texas statute: “In [Walz and other cases], ... we emphasized that the benefits derived by religious organizations flowed to a large number of nonreligious groups as well. Indeed, were those benefits confined to religious organizations, they could not have appeared other than as state sponsorship of religion . . . .” (Texas Monthly, supra, at p. 11 [ 109 S.Ct. at p. 897 ] (plur. opn. of Brennan, J.).) In contrast, “Texas’ sales tax exemption for periodicals published or distributed by a religious faith and consisting wholly of writings promulgating the teaching of the faith lacks sufficient breadth to pass scrutiny under the Establishment Clause. Every tax exemption constitutes a subsidy that affects nonqualifying taxpayers, forcing them to become ‘indirect and vicarious “donors.” ’ [Citations.] Insofar as that subsidy is conferred upon a wide array of nonsectarian groups as well as religious organizations in pursuit of some legitimate secular end, the fact that religious groups benefit incidentally does not deprive the subsidy of the secular purpose and primary effect mandated by the Establishment Clause. However, when government directs a subsidy exclusively to religious organizations that is not required by the Free Exercise Clause and that either burdens nonbeneficiaries markedly or cannot reasonably be seen as removing a significant state-imposed' deterrent to the free exercise of religion, as Texas has done . . . , it ‘provide[s] unjustifiable awards of assistance to religious organizations’ and cannot but ‘conve[y] a message of endorsement’ to slighted members of the community.” (Id. at pp. 14-15 [ 109 S.Ct. at p. 899 ], fn. omitted.) The main concurring opinion agreed that “by confining the tax exemption exclusively to the sale of religious publications, Texas engaged in preferential support for the communication of religious messages.” (Id. at p. 28 [ 109 S.Ct. at p. 907 ] (conc. opn. of Blackmun, J.).)

In a variety of other factual contexts, as well, the high court has considered the evenhandedness, or generality, of benefits an important factor in its establishment clause analysis. Besides Walz and Texas Monthly, one may *748 look to such cases as Rosenberger v. Rector and Visitors of Univ. of Va. (1995) 515 U.S. 819, 839-840 [ 115 S.Ct. 2510, 2522 , 132 L.Ed.2d 700 ] (holding that use of a student activity fund to pay costs of a Christian student publication would not violate the establishment clause, because the fund was used to create an open forum for speech and publication, evenhandedly supporting a wide variety of student publications); Estate of Thornton v. Caldor, Inc. (1985) 472 U.S. 703, 710 , footnote 9 [ 105 S.Ct. 2914, 2918 , 86 L.Ed.2d 557 ] (law requiring employers to give religious employees their Sabbath days off was held unconstitutional, in part because it gives Sabbath observers a “valuable right” not provided to employees with “legitimate, but nonreligious, reasons” for wanting a particular day off); Mueller v. Allen (1983) 463 U.S. 388, 397 [ 103 S.Ct. 3062, 3068 , 77 L.Ed.2d 721 ] (state tax deduction for educational expenses was upheld, in part because its benefits extend to families with children in public schools and nonsectarian private schools); and Committee for Public Education v. Nyquist (1973) 413 U.S. 756, 794 [ 93 S.Ct. 2955, 2976 , 37 L.Ed.2d 948 ] (tax benefits were held unconstitutional, in part because they would go primarily to parents of children in sectarian private schools).

Just last term, in Mitchell v. Helms, supra, 530 U.S. 793 [ 120 S.Ct. 2530 ], the plurality, concurring, and dissenting justices all recognized the important role of evenhandedness in assessing an establishment clause challenge. Though the justices disagreed sharply as to whether evenhandedness in the distribution of benefits is, by itself, a sufficient condition of constitutionality under the establishment clause, all agreed it is an important, in some cases necessary, condition of constitutionality. 11

More specifically, the evenhandedness aspect of the neutrality principle bars government from engaging in “religious gerrymanders.” (Walz, supra, 397 U.S. at p. 696 [ 90 S.Ct. at p. 1425 ] (conc. opn. of Harlan, J.).) With the *749 exception of relief reasonably needed to alleviate substantial government burdens on free exercise, religious adherents and institutions must neither be denied nor granted government benefits on the basis of religion itself. In the absence of a demonstrable need for special accommodation of religion, religious institutions must not be singled out for special treatment, though they may, of course, be included in rationally drawn regulatory groups. “In any particular case the critical question is whether the circumference of legislation encircles a class so broad that it can be fairly concluded that religious institutions could be thought to fall within the natural perimeter.” (Ibid.)

Two recent cases from the same federal district court, Renzi v. Connelly School of Holy Child, supra, 61 F.Supp.2d 440 (Renzi), and Concerned Citizens of Carderock v. Hubbard (D.Md. 2000) 84 F.Supp.2d 668 (Concerned Citizens), illustrate the role of generality in establishment clause challenges to land use law exemptions.

In Renzi , a county zoning ordinance that required property owners to obtain a special exception for nonconforming uses in residential zones exempted any private school located on the grounds of a “church or religious organization.” Renzi, supra, 61 F.Supp.2d at p. 442 .) In finding an establishment clause violation, the court relied both on the lack of a substantial burden needing accommodation (id. at pp. 446-447) and on the ordinance’s unnecessary limitation to schools on religious property, a lack of evenhandedness the court concluded violated the principle of government neutrality. “In the final analysis, it is on this principle that [the exemption] founders because it fails to be neutral in a context where neutrality is possible. By favoring sectarian schools over most other non-profit private educational institutions, it belies any secular purpose and has a principal effect of advancing religion.” (Id. at p. 444, fn. omitted.)

Concerned Citizens involved a different part of the same county zoning ordinance, setting out a lengthy list of nonresidential uses permitted in the county’s RE-2 zones. These included “churches . . . and other places of worship,” but also included such diverse institutions as embassies, small bed and breakfasts, group homes, child and adult daycare facilities, adult foster care homes, libraries and museums. Nonpermitted uses could be located in the zone only if they obtained a special exception. (Concerned Citizens, supra, 84 F.Supp.2d at p. 670 & fn. 2.) In finding the ordinance passed establishment clause muster, the court again relied on generality as a measure of neutrality, concluding in this case that the ordinance “grants a benefit ... to a wide array of ‘nonsectarian [uses] as well as religious [uses] *750 . . . [Texas Monthly, supra,] 489 U.S. at 14 , 109 S.Ct. 890 , consistent with its purpose of preserving single family residential neighborhoods. The Ordinance envelops a broad circle of beneficiaries designating some 41 categories of permitted uses in the RE-2 zone, into which ‘churches . . . and other places of religious worship’ naturally fit.” (Id. at p. 674, first ellipsis added.) Renzi was distinguished as involving a law “the operative characteristic [of which] was religion.” (Concerned Citizens, supra, at p. 675, italics omitted.)

Both cases seem to me correctly decided, and, as discussed below, I find the present case more like Renzi than Concerned Citizens. 12 Here, too, the law’s operative characteristic is religion, for sections 25373(d) and 37361(c) provide the valuable privilege of self-exemption from landmark laws solely to “religiously affiliated” corporations and associations. Here, too, the blanket grant of this self-exemption power was unnecessary to prevent government interference with, or to alleviate a significant burden on, the free exercise of religion. Like the ordinance in Renzi, supra, 61 F.Supp.2d at page 444 , our law “fails to be neutral in a context where neutrality is possible.”

B. Sections 25373(d) and 37361(c) Create a Religious Gerrymander

To state the obvious, sections 25373(d) and 37361(c) do not distribute government benefits evenhandedly to the religious and nonreligious alike; the self-exemption power they provide is given only to “religiously affiliated” associations and corporations. Since, as we have seen in part I, ante, the unrestricted grant of self-exemption cannot be justified as needed to alleviate a significant burden on, or prevent government interference in, the free exercise of religion, no reason appears for restricting relief to religious property owners. 13 Without such reason or justification, such a law granting valuable privileges only to religious organizations advances those organizations’ missions relative to nonreligious counterparts and thus conveys a *751 message of comparative government endorsement of religious viewpoints. While evenhandedness may not itself guarantee constitutional neutrality, a lack of evenhandedness, without a need for the discrimination, is a departure from the neutral stance required of government.

The majority appears to argue that the restriction of benefits to religious organizations under sections 25373(d) and 37361(c) is constitutionally insignificant. It quotes a single sentence from Corporation of Presiding Bishop : “Where, as here, government acts with the proper purpose of lifting a regulation that burdens the exercise of religion, we see no reason to require that the exemption come packaged with benefits to secular entities.” (Corporation of Presiding Bishop, supra, 483 U.S. at p. 338 [ 107 S.Ct. at p. 2869 ].)

In context, however, that remark says .little about the role of a lack of evenhandedness in a case like this one. One must note, first, that the high court did not say that an exemption’s singling out of religious organizations was always permissible, only that it was not “per se invalid.” (Corporation of Presiding Bishop, supra, 483 U.S. at p. 338 [ 107 S.Ct. at p. 2869 ], italics omitted.) The court, indeed, observed that on at least two prior occasions it had given weight to the breadth or narrowness of an exemption. (Ibid., citing id. at p. 333, fn. 11 [ 107 S.Ct. at p. 2867 ], citing Mueller v. Allen, supra, 463 U.S. at p. 397 [ 103 S.Ct. at p. 3068 ], and Committee for Public Education v. Nyquist, supra, 413 U.S. at p. 794 [ 93 S.Ct. at p. 2976 ].) Second, as the quoted sentence makes explicit, Corporation of Presiding Bishop was a case in which the general law at issue, title VII’s bar on religious discrimination in the workplace, in fact placed a significant burden on religious organizations’ religious freedom. Indeed, the prohibition against religious discrimination in employment, by its nature, uniquely burdened the exercise of religion, because only religious employers were likely to suffer significant impairment of their legitimate activities if bound by this prohibition of title VII. Congress’s decision to exempt only religious employers from the prohibition can, therefore, readily be characterized as a neutral measure to prevent interference with religion. Where, as here, however, the costs of a regulation fall more generally on secular and religious institutions, the lack of evenhandedness in granting relief by exemption becomes much more significant.

Second, the majority attempts to distinguish Texas Monthly , a sales tax exemption case, as involving a forced subsidy from nonexempt taxpayers to *752 exempt religious publishers, relying on the Texas Monthly plurality’s observation that “[e]very tax exemption constitutes a subsidy that affects non-qualifying taxpayers, forcing them to become ‘indirect and vicarious “donors.” ’ ” (Texas Monthly, supra, 489 U.S. at p. 14 [ 109 S.Ct. at p. 899 ].) In contrast, the majority asserts, “an exemption from landmark status does not create a subsidy for religious activity by forcing other property owners to be vicarious donors.” (Maj. opn., ante, at p. 714.)

In my view, no salient distinction appears between the manner in which sections 25373(d) and 37361(c) favor religious property owners over their nonreligious counterparts and the manner in which the Texas sales tax exemption favored religious publishers over their nonreligious counterparts. None of the Texas Monthly opinions suggest that, under Texas law, a specified dollar amount had to be raised each year from taxes on sales of books and magazines, so that the cost of the religious exemption would have to be borne exclusively by other publishers. Rather, the involuntary donation effect was described as indirect', to the extent the exemption for religious publications deprived Texas of tax revenue, the state would have to raise it from other sources or reduce the cost or amount of public services funded by the tax. The subsidy to religious taxpayers would be borne by other, nonreligious taxpayers, by the general population of state service recipients, or by both.

Similarly, California has required that localities subsidize religious property owners with a valuable privilege—the power to exempt themselves from costly land use regulations. If a city, prevented by the exemption from preserving the historical and cultural value of a structure, wishes to make up as well as possible the historical and cultural deficit thereby created, it will have to designate additional landmarks and/or apply its restrictions that much more strictly to already designated buildings. Alternatively, the city may choose to forgo that amount of the public good created by historical preservation. Either way, nonadherents of the religion—either other historic property owners, the citizenry as a whole, or both-—become involuntary donors to the religious organization’s mission.

To the extent, then, that land use regulation is understood to create a public good, or prevent a public harm, it stands for present purposes in a comparable position to the collection of tax revenue. “Honoring religion’s negative claims . . . also may produce forced subsidies. Tax exemption shifts burdens from some to others, and exemptions from regulatory schemes result in uncorrected harms of the sort with which the scheme is concerned. Exemption for religious entities from land-use regulation, for example, *753 inescapably imposes social and economic costs upon others in the community.” (Lupu, The Trouble with Accommodation, supra, 60 Geo. Wash. L.Rev. at pp. 750-751, fn. omitted.)

Perhaps, then, the majority is relying on an unarticulated belief that historic preservation is not a sufficiently important public good, and the loss of landmarks not a significant civic threat, for Texas Monthly's subsidy rationale to be applicable. If so, I disagree. Particularly in California, with its relative paucity of historic buildings and its population perpetually rich in newcomers, preserving what landmarks we have is all the more vital to creating and continuing a sense of community. As Winston Churchill observed, “We shape our buildings; thereafter they shape us.” (Quoted in Simpson’s Contemporary Quotations (Houghton Mifflin 1988).)

Preservation of historic buildings and sites benefits the citizens in obvious ways by conserving aesthetically pleasing features of the urban landscape and by bolstering aspects of the local economy, such as tourism, that are dependent on the city’s distinctive character. Beyond that, landmarks• are crucial to the coherence of the community itself. “In essence, ‘we strive to save our historic and architectural heritage simply because ... it has become part of us.’ It has become so on two levels, the individual and the communal. ... [U] ... At the most basic level, a landmark can help physically orient a person in his environment; we oftentimes use landmarks to find our way about a city. A landmark, therefore, may be an important part of the ‘legible’ cityscape. In addition, a landmark can serve to orient a person in her historical or cultural environment. . . . ffl] More importantly, a landmark can impact an individual psychologically. Buildings function as symbols, imparting meaning beyond their historical context or architectural characteristics. [^] Preservation saves these meanings and associations. . . . [^[] On a greater scale, the cumulative effect of the icon-individual subjective experience becomes the basis for a shared communal identity with roots in the past, a present manifestation, and a foundation for the future. . . . [H] Aside from transmitting knowledge of the past, landmark preservation functions to provide a basis for a separate, presently shared community experience. For example, an individual who has recently moved to Boston need not be specifically aware of Fenway Park’s storied past, of the great baseball players who have competed there over the years, to appreciate its value as an icon in the community. By doing so, she has an immediate tie to individuals who are knowledgeable about the stadium’s past, or who have resided in the community for many years. This is one, admittedly trivial, example, but taken as a whole, landmarks can serve to presently facilitate communal bonds, and thereby stability, among a spectrum of residents who might *754 otherwise have little in common.” (Nunez & Sidman, supra, 12 J.L. & Religion at pp. 311-313, fns. omitted.)

Texas Monthly's rationale, therefore, appears fully applicable here. The Legislature’s unjustifiably unequal treatment of religious and nonreligious property owners forces the latter, along with the citizenry as a whole, to subsidize the former’s missions. The California law “fails to be neutral in a context where neutrality is possible.” (Renzi, supra, 61 F.Supp.2d at p. 444 , fn. omitted.)

Again, this is not to suggest that every exemption for a religious landmark would properly be regarded as an unconstitutional forced subsidy of religion. Where unyielding application of landmark restrictions would prevent a religious group from continuing to worship as it chooses in' the landmarked building, for example, an exemption, even if targeted to a particular religious group, would be seen not as favoritism but as a reasonable accommodation of religious freedom. Nor would any establishment clause problem be presented by a hardship exemption designed and applied more generally to a rationally defined set of nonprofit religious, philosophical, charitable, educational, and civic organizations. (See Texas Monthly, supra, 489 U.S. at pp. 15-16 [ 109 S.Ct. at p. 900 ].) Unfortunately, the Legislature took neither of these approaches in sections 25373(d) and 37361(c), instead granting solely to religious organizations a unique and unjustified power of self-exemption.

Conclusion

Part I of this opinion’s discussion measured the self-exemption provisions of sections 25373(d) and 37361(c) along the axis of advancement versus accommodation of religion, concluding that the grant of an unrestricted, unreviewable privilege of exempting one’s own property from historic preservation regulations could not reasonably be characterized as mere accommodation of free exercise. Rather, this drastically overbroad measure— which would, for example, allow a religious organization with a historic and still usable house of worship to demolish and sell it for commercial development, without even the waiting period for preservation efforts most landmark laws provide, simply in order to maximize the economic value of its real estate portfolio—could be seen by an objective observer only as comparatively advancing and promoting the missions of religious organizations by abdicating to them the government function of administering generally applicable land use regulations.

In part II, I measured the neutrality of the same provisions along a different axis, that of evenhandedness or generality in the granting of

*755 government benefits. Because they cannot be reasonably characterized as measures necessary for the accommodation of religious exercise, sections 25373(d) and 37361(c) could survive scrutiny in this dimension only if the class of recipients of the government benefit bestowed was rationally defined by a secular legislative purpose. A more general hardship exemption, such as those contained in many local landmark laws, would meet this criterion even if some of its benefits were to go to religious organizations, since relief of property owners from unusual regulatory hardships would be a proper secular goal. But because religious organizations are not the only property owners that occasionally need relief from strict enforcement of landmark regulations in order to continue using their structures, the grant of a self-exemption privilege solely to religiously affiliated corporations and associations cannot be deemed neutral along this axis either.

However one looks at them, in short, the self-exemption provisions of sections 25373(d) and 37361(c) fail to display the neutrality toward religion required of the government under the establishment clause. For this reason, I dissent.

George, C. J., concurred.

Because I conclude the law fails scrutiny under the United States Constitution, I need not discuss the religion clauses of the California Constitution. If required to address the question, however, I would construe our own establishment and no-preference clauses (Cal. Const., art. I, § 4) as barring preferential government advancement and promotion of religious missions at least to the same extent as I believe the federal establishment clause bars such preference.

In Smith , the high court disavowed the balancing test it had sometimes applied to adjudicate religious freedom challenges to generally applicable laws, holding instead that, in general, the free exercise clause does not require government to grant exemptions for religiously motivated conduct from neutral, generally applicable laws. ( Smith, supra, 494 U.S. at pp. 883-888 [110 S.Ct. at pp. 1602-1605].)

(See also St. Bartholomew’s Church v. City of New York (2d Cir. 1990) 914 F.2d 348 , 353-356 [where landmarking prevented church from replacing its midtown Manhattan community house with commercial office tower, but did not prevent continuation of religious programs in the current community house, loss of economic opportunity, though large, was not an unconstitutional burden]; Christian Gospel Church v. San Francisco (9th Cir. 1990) 896 F.2d 1221, 1224 [preceding Smith : burdens caused by denial of conditional use permit for church wishing to locate in residential zone “minimal” ones of “convenience and expense,” where church made no showing that worshiping in the particular home at issue was important]; Lucas Valley Homeowners Assn. v. County of Marin (1991) 233 Cal.App.3d 130, 143-147 [ 284 Cal.Rptr. 427 ] [applying pre-Smith law: Orthodox Jewish congregation not constitutionally entitled to exemption from zoning ordinance requiring special use permit for a house of worship in a residential area, though application of ordinance would make practice of its religion more onerous].)

Even the Supreme Court of Washington, which has construed both the First Amendment and the free exercise clause of its own state Constitution so as to give maximum protection to religious exercise, has acknowledged that not all financial burdens created by neutral land use laws pose a significant threat to free exercise. (Compare Open Door Baptist Church v. Clark County (2000) 140 Wash.2d 143 [ 995 P.2d 33, 42-43 ] [application fee for conditional use permit, estimated at more than $5,000, is not a cognizable burden without a showing of inability to pay] with First Covenant Church v. Seattle (1992) 120 Wash.2d 203 [ 840 P.2d 174, 183 ] [financial cost an unconstitutional burden only if “too gross”; landmark restrictions reducing church building’s value by half is impermissible burden].)

(See, e.g., Swaggart Ministries v. Cal. Bd. of Equalization (1990) 493 U.S. 378, 391 [ 110 S.Ct. 688, 696-697 , 107 L.Ed.2d 796 ] [application of sales tax to religious publications did not burden evangelical organization in a constitutionally significant manner, though it imposed an economic cost; “such a tax is no different from other generally applicable laws and regulations-—-such as health and safety regulations—to which appellant must adhere”]; Hernandez v. Commissioner (1989) 490 U.S. 680, 699 [ 109 S.Ct. 2136, 2148-2149 , 104 L.Ed.2d 766 ] [expressing doubt whether disallowance of tax deduction for purchase of religious training and counseling from Church of Scientology imposed substantial burden: at most, disallowance meant “adherents have less money available to gain access to such sessions”]; Bob Jones University v. United States (1983) 461 U.S. 574, 603-604 [ 103 S.Ct. 2017, 2035 , 76 L.Ed.2d 157 ] [“Denial of tax benefits will inevitably have a substantial impact on the operation of private religious schools, but will not prevent those schools from observing their religious tenets”]; see also Smith v. Fair Employment & Housing Com. (1996) 12 Cal.4th 1143, 1172-1173 [ 51 Cal.Rptr.2d 700 , 913 P.2d 909 ] [that landlord with religious objection against renting to unmarried couples might incur some cost in shifting from residential real estate to other investments does not, even under pre-Smith law, support her claim of constitutional entitlement to exemption from marital status discrimination bar].)

A dictum in Smith, upon which the majority relies, is not authority for the constitutionality of all legislative measures labelled as accommodations of religion. Though Smith alluded with apparent equanimity to the prospect of “leaving accommodation to the political process” (Smith, supra, 494 U.S. at p. 890 [ 110 S.Ct. at p. 1606 ]), the court was considering only the free exercise question of whether exemptions from generally applicable law are constitutionally required; no establishment clause question was decided or discussed in Smith. Moreover, even if the Smith court’s dictum were taken as a suggestion that legislative accommodation would be permissible under the establishment clause, it would say nothing about cases where the general law imposes no significant burden on religious practice, since the law at issue in Smith—a criminal prohibition on the use of certain drugs—clearly placed a substantial burden on the claimants’ sacramental use of peyote.

The majority opinion (ante, at p. 712) asserts that I fault the optional (self-exemption) character of the statutes “even though the Legislature could have simply exempted the property.” (Italics added.) Needless to say, I disagree that “the Legislature could have simply exempted the property.” A blanket exclusion would suffer from the same types of non-neutrality as the present self-exemption provisions.

The majority opinion, citing a letter to a legislative committee from then San Francisco Mayor Frank Jordan, as well as a letter to the Governor from the city’s future mayor, Willie L. Brown, Jr., leaves the' impression the City and County of San Francisco officially supported Assembly Bill No. 133. (Maj. opn. ante, p. 710, fn. 5.) In fact, the San Francisco Board of Supervisors adopted, and Mayor Jordan signed, a resolution opposing the legislation on the ground that it would permanently preempt the city’s ability to protect its historic resources, protection that “contributes to the economic vitality of the City by maintaining a City that is unusual in its beauty and attractive to tourists.” (S.F. Res. No. 887-93 (Nov. 12, 1993).) On June 23, 1994, the city’s Director of Intergovernmental Affairs sent the resolution to Marian Bergeson, Chair of the Senate Committee on Local Government, with a letter urging her opposition to Assembly Bill No. 133.

Relying on anecdotal evidence in Speaker of the Assembly Brown’s letter to the Governor, the majority suggests that religious organizations in San Francisco “were faced with very high seismic retrofit costs for older buildings that had been or were subject to landmark designation but could not qualify for assistance from the Federal Emergency Management Agency (FEMA).” (Maj. opn., ante, at pp. 710-711, fn. 5.) Without disputing that religious groups may sometimes face special barriers to renovating or retrofitting their facilities, perhaps justifying individual hardship exemptions from landmark laws, I note that FEMA regulations appear to contain no general prohibition on religiously affiliated organizations receiving disaster assistance; although houses of worship do not seem to fall within the definition of an eligible “private nonprofit facility” ( 44 C.F.R. § 206.221 (e) (2000)), hospitals, daycare and senior citizen centers, homeless shelters, schools and similar, institutions operated by religiously affiliated organizations appear eligible so long as their facilities are open to the general public (id., § 206.221(e)(4)-(6)) and, in the case of schools, are not “used primarily for religious purposes or instruction” (id., § 206.221(e)(1)). I further observe that San Francisco’s Landmarks Preservation Advisory Board, writing legislators to oppose Assembly Bill No. 133, pledged “to continue to work with religious leaders to find solutions to address *743 State imposed mandates to seismically retrofit Unreinforced Masonry Buildings (UMB’s) including finding potential sources of funding such as bonds, loans and grants to repair and maintain religious properties.” (S.F. Landmarks Preservation Advisory Bd., letter to Sen. Bergeson, June 27, 1994.) The Advisory Board President also wrote separately to Senator Bergeson, pointing out that of the nine churches proposed by the Catholic Archdiocese for closure, only two were UMB’s, and these two were already landmarked and hence unaffected by the proposed legislation. (Patrick McGrew, Pres., S.F. Landmarks Preservation Advisory Bd., letter to Sen. Bergeson, June 28, 1994.) If seismic retrofit of San Francisco’s older religious buildings was the problem before the Legislature, Assembly Bill No. 133 was a grossly overbroad approach to a solution.

(See Lucas Valley Homeowners Assn. v. County of Marin, supra, 233 Cal.App.3d at pp. 149-151 [placement of conditions on congregation’s use permit did not entangle county with religion, as the conditions were “concerned with mundane matters such as numbers, hours, location and noise restrictions,” and did not call for the county to “divin[e] religious content or otherwise pass[] on the religious affairs of Chabad”]; Renzi v. Connelly School of Holy Child (D.Md. 1999) 61 F.Supp.2d 440, 446 , revd. sub nom. Ehlers-Renzi v. Connelly School of the Holy Child (4th Cir. 2000) 224 F.3d 283 [no substantial government interference with religion would be threatened by requiring religious school to obtain permit for operation in residential area, as permit process “would merely subject its use of the property to such basic zoning considerations as constructing facilities generally harmonious with the neighborhood and preventing excess traffic congestion”]; cf. Hernandez v. Commissioner, supra, 490 U.S. at pp. 697-698 [109 S.Ct. at p. 2148] [Internal Revenue Service’s administration of charitable contribution deduction does not create impermissible entanglement because it does not require government to determine value of religious services, though it does require religious institutions to disclose cost information to the IRS].)

That plaintiffs have challenged sections 25373(d) and 37361(c) on their face rather than in any particular application does not reduce the force of this overbreadth analysis. That a law designed to favor religious adherents with unjustified assistance may in some cases be applied in a neutral manner does not alleviate the law’s destructive effect, prohibited under the establishment clause, of conveying endorsement by differentially advancing religious causes over nonreligious ones. (See Santa Fe Independent School Dist. v. Doe, supra, 530 U.S. at p. _[ 120 S.Ct. at p. 2281 ] [upholding facial challenge to school district’s policy sponsoring invocations at football games, despite possibility some invocations would be nonreligious, finding constitutional violation in “the mere passage by the District of a policy that has the purpose and perception of government establishment of religion”].)

(See Mitchell v. Helms, supra, 530 U.S. at pp. 809-810 [120 S.Ct. at p. 2541] (plur. opn. of Thomas, J.) [“[I]f the government, seeking to further some legitimate secular purpose, offers aid on the same terms, without regard to religion, to all who adequately further that purpose, then it is fair to say that any aid going to a religious recipient only has the effect of furthering that secular purpose”]; id. at p. 838 [120 S.Ct. at p. 2557] (cone. opn. of O’Connor, J.) [“I do not quarrel with the plurality’s recognition that neutrality is an important reason for upholding government-aid programs against Establishment Clause challenges. Our cases have described neutrality in precisely this manner, and we have emphasized a program’s neutrality repeatedly in our decisions approving various forms of school aid”]; id. at p. 883 [120 S.Ct. at pp. 2580-2581] (dis. opn. of Souter, J.) [“There is, of course, good reason for considering the generality of aid and the evenhandedness of its distribution in making close calls between benefits that in purpose or effect support a school’s religious mission and those that do not. . . . [T]he breadth of evenhanded distribution is one pointer toward the law’s purpose, since on the face of it aid distributed generally and without a religious criterion is less likely to be meant to aid religion than a benefit going only to religious institutions or people”].)

A split panel of the Fourth Circuit Court of Appeals recently reversed Renzi . (Ehlers- Renzi v. Connelly School of the Holy Child, supra, 224 F.3d 283 .) I agree with the dissenting judge in that case that the majority, in upholding the preferential treatment as an accommodation of religious freedom despite the lack of a showing of significant burden, was “inappropriately expanding the [Corporation of Presiding Bishop] principle and, as a result, traveling down a path that will ultimately render the Establishment Clause meaningless.” (Id. at p. 293 (dis. opn. of Murnaghan, J.).)

Nothing in the record or briefing indicates that landmark regulation imposes costs only or primarily on religious property owners. Religious properties comprise only 1.2 percent of landmarks in New York City, 1.1 percent in Philadelphia (Weinstein, supra, 65 Temp. L.Rev. at pp. 111-112), and approximately 12.5 percent in San Francisco. Other charitable, educational, cultural and civic organizations must bear the incidental costs of landmark designation, such as application costs and delays, the increased expense of repairing, rehabilitating and expanding their properties in a historically appropriate manner, and the opportunity cost of forgoing commercial development of landmarked noncommercial buildings occupying prime *751 commercial real estate. In some instances, to be sure, religious property owners may face unique problems—such as the desire to change liturgically significant features, or disqualification from participating in government-funded rehabilitation programs. As discussed in the previous part, however, sections 25373(d) and 37361(c) were not designed to alleviate any such unique burdens, but to grant a blanket power of self-exemption from landmark laws.