Anti-Discrimination Law, Religious Organizations, and Justice

Abstract

In many jurisdictions the list of factors for which anti-discrimination law applies has been expanded to include sexual orientation. As a result, moral and legal difficulties have arisen for religious organizations whose basic beliefs include the belief that sexual acts between persons of the same sex are immoral. In light of these difficulties, is anti-discrimination law of this sort unjust? Recently John Finnis has argued that, as commonly applied, such anti-discrimination law is disproportionate and therefore unjust. In this essay, I critically examine Finnis's argument, and argue that, on account of the conception of disproportionateness that is employed, it does not succeed. So as to enable the argument from disproportionateness to succeed, I develop and defend a modified conception of disproportionateness.

In many jurisdictions, the list of factors for which anti-discrimination law applies – ethnicity, nationality, race, sex, and so forth – has, in recent years, been expanded to include sexual orientation. As a result, moral and legal difficulties have arisen for religious organizations whose basic beliefs include the belief that sexual acts between persons of the same sex are immoral. For instance, in 2006, as a result of this expansion in anti-discrimination law (among other factors), Catholic Charities of Boston decided to no longer offer adoption services. In the view if its leadership, the organization could not both be faithful to its religious and moral beliefs, and comply with the legal requirement not to discriminate against same-sex partners in its child placement efforts.¹

In this essay, I address the question of whether anti-discrimination law of the sort confronted by Catholic Charities of Boston is unjust. In Section I, I critically examine John Finnis's brief analysis of the justice of such anti-discrimination law. In Section II, I follow Finnis's lead, arguing that such anti-discrimination law is unjust because it is disproportionate, but to do so I first develop an alternative conception of disproportionateness. In Section III, I address a number of potential objections to the argument of Section II. Section IV concludes.

I

Recently, John Finnis has addressed the question of whether anti-discrimination law of the sort confronted by Catholic Charities of Boston, what I will refer to as exemption-less anti-discrimination law – anti-discrimination law that (i) has been expanded to include sexual orientation as one of the factors for which discrimination is forbidden, and (ii) in which exemptions are not granted with respect to this factor, even when, on grounds of religion or conscience, it is believed that sexual acts between persons of the same sex are immoral – is unjust. His position is that such anti-discrimination law is unjust, and is so in virtue of being “disproportionate.”²

Finnis is not opposed to efforts by governments to do away with wrongful discrimination. For instance, he claims that “the eliminating or minimizing of the use of irrelevant considerations in decision-making in public life broadly conceived is a legitimate aim (where the considerations are truly irrelevant).”³ Further, he claims that “the opening up of public spaces and services, broadly conceived, to everyone properly within the realm, without discrimination on grounds not sufficiently related to the public welfare, is a legitimate aim” of government.⁴

However, in Finnis's view, that a government's end – such as to “honor . . . the radical basic equality of all human beings”⁵ – is legitimate, and that the means it employs – such as (what I have for convenience labeled) exemption-less anti-discrimination law – is an effective means of achieving the legitimate end is not sufficient for the employment of such means to be just. Why so? According to Finnis, justice requires that the “side effects . . . for other elements of the common good” such as any “negative impact on established constitutional rights such as freedom of association, freedom of religion and conscience, and freedom of parents to educate their children towards good forms of life”⁶ must also be taken into account.⁷

Given this view of what justice involves, Finnis is critical of recent applications of exemption-less anti-discrimination law. For instance, referring to a number of English cases, he notes:

In all these cases, the courts have proceeded straight from affirming the legitimacy of an anti-discriminatory aim, and the efficacy of the anti-discrimination policy's means, to concluding that the policy was justified and the conduct it prohibited was discrimination in the genuine sense: unjustified differentiation. These courts neglected their duty to consider whether the means (the policy) was not only effective but proportionate, i.e., did not affect people's legitimate interest in other recognized rights more than was needed by the legitimate aim.⁸

But what if a government's initiative (the means it employs) has negative side effects with respect to “other elements of the common good”? What if, for instance, exemption-less anti-discrimination law has negative side effects with respect to, say, freedom of religion, or freedom of conscience, or both?

Finnis does not claim that such negative side effects necessarily make an initiative unjust. Instead, relying on the (European) legal concept of “proportionate means,”⁹ he invokes what can be seen as a standard for determining whether such an initiative – one that employs effective means to a legitimate end but also has negative side effects with respect to some other component(s) of the common good – is just. According to this standard, “means of pursuing a legitimate aim are proportionate . . . if and only if they are rationally connected with a legitimate aim (an objective sufficiently important to warrant the negative side-effects of pursuing it), and are ‘no more than is necessary to accomplish the objective.’”¹⁰ (I will refer to this two-part standard as the proportionate means test.) What does it mean for the means to be “no more than is necessary to accomplish the objective”? Finnis accepts what he sees as being a common interpretation: “having no more adverse impact on the enjoyment of other rights or protected interests than any available alternative effective means of pursuing the legitimate aim.”¹¹

The utility of this test of the proportionality of the means employed for addressing questions of the justice of the means employed resides in the relationship that Finnis assumes to exist, if I understand him correctly, between a given means being disproportionate and being unjust: if the means are disproportionate, they are thereby unjust (that is, they are unjust in virtue of being disproportionate).¹²

With (what I refer to as) the proportionate means test in hand, Finnis addresses a number of controversial government initiatives.¹³ With respect to applying exemption-less anti-discrimination law (in England) in the case of religious organizations that provide adoption services but, on account of their beliefs, are adverse to placing children with same-sex partners, he claims that:

disproportionateness by needlessness was . . . vividly manifested by the law prohibiting adoption agencies from continuing to give effect to their judgment . . . that both the unchastity and the lack of complementarity involved in adoption by same-sex sex-partners should count at least as a negative factor, if not a disqualification, in decisions about adoption. This coercion (resulting in some cases in the agency's withdrawal from providing adoption services at all) was imposed by the English enactments even though would-be same-sex adopters had available other suitable and vastly more numerous adoption agencies willing to cater for them.¹⁴

It can be recalled that, as noted above, Finnis's position is that “means of pursuing a legitimate aim are proportionate” if two conditions are met: (1) “they are rationally connected with a legitimate aim (an objective sufficiently important to warrant the negative side-effects of pursuing it)” and (2) the means are “no more than is necessary to accomplish the objective.”¹⁵ With respect to the application of exemption-less anti-discrimination law to adoption agencies, Finnis's complaint is with the latter condition. His complaint, I believe, can be put as follows: even if it is granted that exemption-less anti-discrimination law would, as intended, have good effects with respect to the legitimate aim of “honoring the radical basic equality of all human beings,” the means to this end – exemption-less anti-discrimination law – are more than necessary, and hence disproportionate and therefore unjust.

For this complaint to be warranted (given Finnis's interpretation of what it means for the means to be no more than necessary – “having no more adverse impact on the enjoyment of other rights or protected interests than any available alternative effective means of pursuing the legitimate aim”¹⁶), exemption-less anti-discrimination law needs to be shown to “hav[e] . . . more adverse impact on the enjoyment of other rights or protected interests than [an] available alternative effective means of pursuing the legitimate aim.”¹⁷ Based on Finnis's analysis with respect to adoption agencies, it appears that he believes that exemption-less anti-discrimination law is disproportionate because its negative side effects with respect to freedom of religion or freedom of conscience (or both) are more than necessary. This is so, in his view, because there are plenty of other adoption agencies that are “willing to cater for” same-sex partners seeking to adopt.¹⁸

However, even if plenty of other adoption agencies are willing to serve same-sex partners, and thus there is no “need” for adoption agencies for which serving such persons would be contrary to their moral or religious beliefs to be required to serve them, this is not sufficient to show that exemption-less anti-discrimination law is disproportionate (and thereby unjust), at least by the lights of the proportionate means test. To see this, suppose that there is an available alternative to exemption-less anti-discrimination law wherein adoption agencies whose basic beliefs include the belief that sexual acts between persons of the same sex are immoral are granted an exemption with respect to the factor of sexual orientation. Let us refer to this alternative as accommodating anti-discrimination law. It is plausible to hold that, compared to exemption-less anti-discrimination law, accommodating anti-discrimination law (a) has less “adverse impact on the enjoyment of other rights or protected interests,” and (b) is an effective means of providing adoption services to all (including same-sex partners).

Given the proportionate means test, however, the presence of an alternative with these attributes is not what is needed to show that exemption-less anti-discrimination law is disproportionate (and thereby unjust). For, recall that on the interpretation of what it means for the means to be no more than necessary to accomplish the objective that Finnis accepts, for exemption-less anti-discrimination law to be disproportionate, an available alternative (such as accommodating anti-discrimination law) must have a less “adverse impact on the enjoyment of other rights or protected interests” and be an “effective means of pursuing the legitimate aim.”¹⁹ Hence, by the lights of the proportionate means test, to show that exemption-less anti-discrimination law is disproportionate (and thereby unjust), what is needed is an available alternative that, compared to it (a) has less “adverse impact on the enjoyment of other rights or protected interests,” and (c) is also an effective means of pursuing the legitimate aim – that of “honoring the radical basic equality of all human beings.”

The problem then with Finnis's analysis (as I understand it) is that it is not plausible to hold that (b) is the same as (c), nor is it plausible to hold that if an initiative is (b) it is necessarily also (c). If this is so, then proponents of exemption-less anti-discrimination law can grant that while accommodating anti-discrimination law would have a less “adverse impact on the enjoyment of other rights or protected interests” and thus do better than exemption-less anti-discrimination law on account of (a), it does not measure up as well with respect to (c) since it is not an effective means of pursuing the legitimate aim of honoring the radical basic equality of all human beings. Such proponents can ask, for instance: “How can letting some adoption agencies discriminate against same-sex partners be consistent with honoring the radical basic equality of all human beings?” Further, they can argue that accommodating anti-discrimination law fails to honor this basic equality in virtue of the accommodation it makes. Therefore, such proponents can plausibly argue that the availability of accommodating anti-discrimination law (and plenty of other agencies willing to serve same-sex couples) is not sufficient to show that exemption-less anti-discrimination law is disproportionate (and thereby unjust).

II

One possible response to the argument of Section I is to offer an alternative conception of disproportionateness that circumvents the problems encountered when the proportionate means test is employed. I suggest that such an alternative conception can be discovered by examining what is commonly accepted as a morally acceptable way of handling a commonly arising situation of incompatibility between fulfilling a legitimate governmental aim (such as honoring the radical basic equality of all human beings) and respecting fundamental rights.

Suppose that a situation of incompatibility arises between fulfilling a legitimate governmental aim (honoring the radical basic equality of all human beings) and respecting the fundamental right of freedom of speech. What ought to be done? It is commonly accepted that appropriately responding to the moral importance of the right of freedom of speech typically requires that costs (in terms of the fulfillment of legitimate governmental aims) be incurred. In such cases of conflict, even speech that is widely viewed as being not only immoral but also such that it hinders the fulfillment of legitimate governmental aims is commonly held to be rightly tolerated so as to appropriately respond to the moral importance of respecting freedom of speech.

For instance, suppose that some persons claim that those of a certain race or sex are unfit to serve in positions of responsibility in government. While it could be argued that, so as to fulfill the legitimate aim of honoring the radical basic equality of all human beings, governments ought to forbid speech of this sort, the commonly accepted (and plausible) response is to hold that doing so would fail to appropriately respond to the moral importance of respecting freedom of speech. Underlying this response is a simple rationale: tolerating such speech (and the foreseeable hindrance it causes to the objective of honoring the radical basic equality of all human beings) is simply a cost that ought to be borne (at least at times) so as to appropriately respond to the moral importance of freedom of speech. In other words, the moral importance of respecting fundamental rights (such as freedom of speech) entails that some costs (in terms of the fulfillment of legitimate aims of governments) ought to be borne in cases wherein fulfilling a legitimate governmental aim and respecting fundamental rights are incompatible.

Were governments to refuse to bear any such costs, it is plausible to hold that their actions would be “disproportionate,” but in a sense different from that captured by the proportionate means test. The sense of disproportionateness that is germane here is that of not properly recognizing the moral importance of each of the various values at stake. This alternative sense of disproportionateness can be put as follows: when the fulfillment of one morally important value is incompatible with the fulfillment of another morally important value (and each value is a value that is relevant to justice, not merely to morality per se), the means employed so as to fulfill one value are disproportionate if they fail to appropriately respond to the moral importance of the other value.²⁰

If we return to the question of the justice of exemption-less anti-discrimination law, and if we assume that Finnis was correct in holding that there is a direct relationship between means being disproportionate and being unjust, given this alternative conception of disproportionateness, it is plausible to hold that exemption-less anti-discrimination law (as the means to the presumably legitimate end of honoring the radical basic equality of all human beings) is disproportionate, and thereby unjust. In this case, the disproportionateness is readily identifiable, given that employing exemption-less anti-discrimination law entails a refusal to appropriately respond to the moral importance of freedom of religion and freedom of conscience – there is a refusal to bear any costs so as to (appropriately) respond to the moral importance of respecting the fundamental rights of freedom of religion and freedom of conscience. Assuming that it is in fact of significant moral importance for freedom of religion and freedom of conscience to be respected, such a response cannot be appropriate, and hence is disproportionate.While it may be difficult to determine just how much costs should be borne so as to appropriately respond to the moral importance of respecting these fundamental rights, given that no costs at all are borne, and that respecting these fundamental rights is morally important, exemption-less anti-discrimination law is readily identifiable as disproportionate, akin to an approach to the freedom of speech that refused to bear any costs (an “absolutist” approach to freedom of speech) – refused to tolerate any speech that has negative side effects with respect to a given legitimate governmental aim. In cases in which there is incompatibility between fulfilling a legitimate governmental aim and respecting the fundamental rights of freedom of religion and conscience, an absolutist and exception-less means such as exemption-less anti-discrimination law cannot plausibly be held to be an adequate response to the moral importance of respecting these fundamental rights, and hence is disproportionate.²¹

This alternative conception of disproportionateness is attractive because it allows the difficulties encountered when the proportionate means test is employed to be circumvented. In particular, the difficulty of showing that an alternative effective means to the legitimate end is available is circumvented because a verdict on the justice of the means employed can be made on analysis of the means in question itself; there is no need for a comparative judgment among available alternatives, as was the case with the proportionate means test.

III

In this section, I address three possible objections to the argument of Section II.

A.

One possible response to the argument of Section II is that whatever persuasiveness it has is due to the noted incompatibility being framed in terms of the mere fulfillment of a legitimate governmental aim on the one hand, and respecting fundamental rights on the other. The worry here is that underlying my argument is an implicit assumption that the moral importance of fundamental rights is such that it is always more important for governments to respect them than to fulfill whatever legitimate aims they may have.

One form that this response could take would be to argue that at least in some cases, the fulfillment of an aim may itself be necessary so as to respect fundamental rights. For instance, it could be argued that implementing exemption-less anti-discrimination law is not simply a laudable and legitimate thing for governments to do but rather is necessary for them to respect fundamental rights – necessary to respect a fundamental right to equal concern and respect, for instance.

However, even if it is granted that persons have a fundamental right to equal concern and respect,²² and that implementing exemption-less anti-discrimination law is necessary for this right to be respected, it can still be held that exemption-less anti-discrimination law is disproportionate, as claimed in Section II. In the absence of a compelling rationale for why the moral importance of respecting one fundamental right is always greater than respecting another fundamental right, in cases of incompatibility between such rights, it would be disproportionate for respecting one fundamental right to always take precedence over respecting other fundamental rights, never bearing any costs when such cases of incompatibility arise. (Disproportionate because such an asymmetrical treatment of fundamental rights would fail to appropriately respond to the moral importance of the always-trumped fundamental rights.)

To see this, recall that, as I argued above, the moral importance of respecting fundamental rights entails that some costs (in terms of the fulfillment of legitimate aims of governments) ought to be borne in cases wherein fulfilling a legitimate governmental aim and respecting fundamental rights are incompatible. Pursuing this line of reasoning a little further, and assuming that the fundamental rights are equally fundamental, we can set forth the following principle: in the absence of a rationale for why the moral importance of respecting one fundamental right is greater than respecting another fundamental right, when properly respecting one fundamental right is incompatible with properly respecting another fundamental right, properly respecting the moral importance of fundamental rights entails that some costs may rightly be borne with respect to each and any fundamental right. If this is correct, we can still hold that exemption-less anti-discrimination law is disproportionate and therefore unjust, even if it is the case that persons have a fundamental right to equal concern and respect, and even if it is granted that actions such as refusing to place children with same-sex partners fails to appropriately respect this right. The reason for this is that even if the values that are in conflict are re-described as being, say, a fundamental right to equal concern and respect on one hand, and a fundamental right of freedom of religion on the other, the application of exemption-less anti-discrimination law entails that the former always trumps the latter (or, in other words, that the costs are always borne by the latter) and such systematic asymmetrical treatment of fundamental rights cannot plausibly be held to be proportionate in the sense of appropriately responding to the moral importance of each of the (justice relevant) values at stake.

B.

Another possible response to the argument of Section II is to take issue with the free speech example I appeal to. It could be said that this example is merely an “intuition pump,” and held that an alternative intuition pump can be develop that yields “intuitions” that conflict with those I invoke. For instance, one could appeal to the historical example of the actions taken by the federal government in the United States to put an end to (race-based) segregation. A plausible evaluation of these actions by the federal government is that they were, in general, morally acceptable. Thus, one could argue that just as it was morally acceptable for the Unites States to require integration, even by coercion, it is morally acceptable for governments to require adoption agencies to place children with same-sex partners.

The problem with this counterargument is that it fails to take into account a morally relevant difference between the two cases. In the case of the practice of segregation, there simply was not a plausible (let alone compelling) moral (as opposed to prudential) argument to be made for the practice of segregation itself. Of course it is morally important to respect the fundamental rights that were at issue in this case (such as freedom of association), but when we look to the practice of segregation itself (apart from the mere assertion of the moral importance of the freedom(s) at stake), no plausible moral justification was available. The situation is different in the case of the practice of adoption agencies refraining from placing children with same-sex partners. In this case, it is at least possible to make a plausible moral argument – not merely assert the moral importance of the freedoms at stake – but defend the practice itself, grounded in ideas that all reasonable persons (religious and non-religious alike) can recognize and grant as being reasonable, such as: that marriage is a basic good and therefore a constitutive element of human well-being, that marriage is only possible between a man and a woman, that all non-marital sexual relations, being contrary to the basic good of marriage, are morally wrong, and that governments have an interest in protecting traditional marriage.²³

C.

Another possible response to the argument of Section II is that when dealing with adoption agencies whose basic beliefs include the belief that sexual acts between persons of the same sex are immoral and that are therefore disinclined to place children with same-sex partners is that any form of anti-discrimination law that is less stringent than exemption-less anti-discrimination law, such as accommodating anti-discrimination law, would be morally unacceptable because tolerating such discrimination would entail government officially recognizing and hence, symbolically, at least, stamping such discrimination with a seal of approval or legitimacy, which itself is morally unacceptable. However, this response relies on an implicit and unwarranted conception of tolerance – tolerance as recognition. However, tolerance does not entail recognition. Indeed, the traditional conception of tolerance is in terms of non-interference rather than recognition. Tolerance need not entail even symbolic recognition (and legitimacy). Hence, tolerance can be exercised at the same time it is made clear that what is being tolerated is morally unacceptable.²⁴

IV

In this essay, I have addressed the question of whether a certain prevalent sort of anti-discrimination law, what I have termed exemption-less anti-discrimination law, is unjust. I critically examined John Finnis's argument for the injustice of such anti-discrimination law and claimed that, on account of the conception of disproportionateness it employs, it does not succeed. So as to enable the argument from disproportionateness to succeed, I developed and defended a modified conception of disproportionateness. With this alternative conception of disproportionateness in hand, it is plausible to argue that exemption-less anti-discrimination law is unjust.