Enabling Exit: Religious Association and Membership Contract

This paper investigates the right of exit from religious associations. The liberal state has a compelling interest in overseeing exit, even if it implies some loss in religious group autonomy. Members should not be bound by rules they find unconscionable. They should be free to leave and able to do so. To enable exit, the paper advocates the use of membership contracts. Religious associations should issue a contract for members working for, residing in, or donating money to the association under a regime of legal exemptions. The membership contract publicises the right of exit and offers a basis for negotiating and contesting its terms. It makes exit less “unthinkable” to members and helps tackle unreasonable economic costs to exit.

contradiction with the oft-repeated organisation's slogan: "The door to Opus Dei is only open a crack to come in but wide open to leave". 2 Exiting proved immensely costly for María. In her forties, she found herself back at her parents' house, without any form of compensation or social security. She struggled to account for her years of devout service to the US immigration authorities, since she was denied a work certificate or an official transcript from the religious institution.
María's experience is exceptional in its gravity, but not unique. Some individuals leaving religious organisations report having given years of work or donated assets to later find themselves deprived of substantive opportunities to start a new life elsewhere. Some give up on leaving altogether, and resign themselves to remaining in communities where they feel like outsiders and obey rules they find unconscionable. What is intriguing in María's story is the discrepancy between the formal proceedings surrounding membership to the organisation, and her own account of being held captive. Although Opus Dei devised highly formalised procedures to acquire and renounce membership, María's testimony reveals a wrenching process of exit.
This paper investigates the right of exit from religious associations and how to best regulate the process of dissociation. There is a wide agreement on the significance of the right of exit. 3 What is in dispute, however, is how to ensure substantive exit rights. 4 Drawing on the analogy with marriage and employment, I offer a response in terms of an adequate regulative framework. Religious associations should issue a membership contract for members working for, residing in, or donating money to the association under a regime of legal exemptions. The membership contract publicises the right of exit and offers a basis for negotiating and contesting its terms. It makes exit less "unthinkable" to members and helps tackle unreasonable economic costs to exit. 5 The paper is structured as follows. Section 1 discusses existing accounts of how to regulate exit from religious associations. Section 2 presents an analogy with marriage and employment and draws some useful lessons from divorce and labour law. Section 3 introduces and defends the idea of a membership contract. Section 4 argues that this proposal constitutes a justified interference in religious group autonomy.

Regulating Exit Adequately
In a liberal state, an adequate regulation of exit must avoid two pitfalls. The first pitfall consists in adopting a posture of laissez-faire, with the hope that individuals will find their way out of their communities with minimal state interference. The second consists in trading too much group autonomy to forcefully enhance individuals' ability to exit. Most proponents of the right of exit are nearer the first pitfall. They take it that the right of exit should be ensured by general 2 Ibid., 319. 3 See: Eisenberg and Spinner-Halev (2004), Spinner-Halev (2000), Spinner-Halev (2008), Kukathas (1992), Barry (2001), Galston (1995), Taylor (2017), Borchers and Vitikainen (2012), ibid. 4 Critics of the right of exit strategy deplore that it is not substantive enough. See: Green (1998), Okin (2002), Shachar (2001), Fagan (2018). 5 As the example of María illustrates, I do not draw a sharp distinction between chosen and unchosen exit. The frontier between exit and expulsion can be blurred. Members can leave reluctantly, because they make decisions at odds with the precepts of the group. Deliberate acts of dissent can lead to exclusion. And even when the contrast is clear, both dissenters and expelled members should have realistic exit opportunities. legislation preventing physical coercion, slavery or forced marriage, and ensuring basic education. Minimal interference should remain the default option. By contrast, proponents of interventionist measures often distance themselves from a right of exit strategy. For instance, Ayelet Shachar expresses scepticism about relying on the right of exit because it "forces an insider into a cruel choice of penalties: either accept all group practices-including those that violate your fundamental citizenship rights-or (somehow) leave." 6 She theorises instead the conditions of "fair remain". 7 In what follows I focus on proposals that are comparatively less interventionist and that take seriously the right of exit. I argue that these proposals are insufficient in one key respect: they neglect to incorporate a legal mechanism susceptible to regulate the terms of exit.
A prominent example of a laissez-faire proposal has been put forward by Chandran Kukathas. 8 A fervent defender of freedom of association, Kukathas presents the right of exit as a fundamental right. Religious and cultural groups are construed as voluntary associations. They are voluntary insofar as each member is free to associate and dissociate. Given that people are often born and raised in cultural or religious groups, the guarantee of the right of exit is essential. One does not chose to enter, but one is free to leave. The right of exit is meant to reconcile individual and group autonomy: it allows members to leave, especially when the rules of their group become oppressive. It is presumed that those who remain consent. 9 The laissez-faire strategy recommends minimal state interference. The option of exit justifies refraining from intervening in groups' internal affairs, even when they diverge from liberal tenets. For instance, the state does not intervene in education systems, however they are structured and whatever is being taught. Instances where state interference is required are very minimally circumscribed to cases of slavery and physical coercion. When other injustices are perpetrated within a group, the fact that members choose to remain should diminish our concern. Still, Kukathas insists that exit rights should be substantive, not merely formal. But what makes a right of exit substantive, in his view, is merely a free society susceptible to host dissenters. Leavers should find a haven to go to. A similar thought is expressed by Jacob Levy: "We outsiders are limited to providing a safe, free and just place to them to exit to" (Levy 2005). Otherwise, minimal interference should prevail.
I have two concerns with the laissez-faire approach. First, it is unclear whether a deregulated religious marketplace lowers individuals' exit costs from religious associations. On the one hand, the cost of joining other religious groups decreases with deregulation. Fewer restrictions on religious liberty, be they about property-rights, registration requirements, or even restrictions on building size, fuel religious vitality and diversity (Gill 2008;Finke and Stark 1998). New religious organisations can freely compete in the religious marketplace, 6 Ibid., 41. 7 Shachar theorises an institutional design to mitigate power hierarchies within groups. Her model of joint governance and "transformative accommodation" aims to alleviate the tension between the accommodation of differences and the situation of vulnerable minorities within minorities. Shachar, Multicultural Jurisdictions: Cultural Differences and Women's Rights. 8 Kukathas, 'Are There Any Cultural Rights? '. 9 Here the justificatory function of the right of exit relies on a presumption of consent. Members have the option to leave, yet they choose to stay, thus they are consenting to the norms of the group. The validity of this consent is highly contestable given the high costs of exit, be they economic, social, or psychological. In addition, the idea that individuals' consent justifies non-interference with illiberal norms is troublesome. Drawing on an analogy with the state, the fact that citizens have the option to emigrate does not justify illiberal laws. My own account of the right of exit does not endorse this justificatory strategy. I am generally favourable to some level of state interference. I hold that the protection of exit rights is necessary even in an interventionist model. giving thereby more opportunities for dissent. On the other hand, these "voluntary associations" demand much more commitment from their adherents: "Instead of passively accepting the offerings of the church, the individual is now called upon to make voluntary contributions of time and money" (Finke 1990). Individual believers invest more resources in religious associations when the religious marketplace is deregulated. As a result, they may find exit more costly.
Second, the appeal of a laissez-faire proposal very much depends on the actual opportunities the wider society provides. A society with an accessible educational system, a strong welfare state, or even a universal basic income, would plausibly lessen the need for state intervention. Paradoxically, a laissez-faire policy in cultural and religious matters must rely on strong state interventionism in socioeconomic matters to make the right effective. The right of exit from voluntary associations is indirectly guaranteed by a society ensuring a fair range of meaningful opportunities. In other words, the state, and society at large, become responsible to make exit feasible for dissenters. While I am sympathetic to progressive measures susceptible to widen the set of opportunities for everyone, I regret that a laissez-faire proposal loses sight of the fact that excessive exit costs may sometimes be unreasonable exit costs. We should remain open to the possibility that some individuals may have a legitimate claim against their former association because they were facing unreasonable exit terms, significantly obstructing their ability to leave-I come back to this idea in my proposal. Laissez-faire policies exempt religious associations from bearing any of the burdens of exit.
While a laissez-faire proposal insufficiently accounts for the responsibility of religious associations, some versions of targeted interventionism may run into the opposite problem, namely the presumption that religious groups should be the primary bearers of exit costs. Jeff Spinner-Halev makes an interesting proposal in this regard. 10 Spinner-Halev is generally wary of undue state interference in religious affairs. In his view, individuals should bear the burdens of the risky choices they make, not the state. And the state is not committed to alleviating the psychological burdens of dissenters. Yet Spinner-Halev is concerned about practical considerations obstructing the right of exit, namely economic obstacles. The case of the Hutterites from Canada illustrates the problem of having worked almost a lifetime for a religious group with communal ownership and having to leave completely destitute. 11 Drawing on this example, Spinner-Halev argues that the state should coerce some religious communities in developing an "exit fund", a small financial help for their former members, too small to constitute an incentive to exit, but enough to allow members to leave. The exit fund is not a compensation for past work, Spinner-Halev insists, but minimal means to exit and reintegrate.
The exit fund tackles an important problem. Spinner-Halev is right to give special attention to economic barriers to exit. They are critical in assessing whether exit is feasible. Recall the dramatic situation of María upon leaving Opus Dei. Economic barriers also very much bear on the psychological burden-leaving in a state of misery is likely to worsen the predicament of leavers. 12 Not all psychological costs can be remedied with material help, but some can and these should be addressed.
I have two worries with the exit fund proposal, however. The first is the following: the narrow scope of the policy makes it difficult to justify the seriousness of the interference. It trades a lot of (a few) religious group autonomy for addressing only a small subset of exit cases. Let me explain.
I am generally favourable to some infringement on religious group autonomy for the sake of lowering exit costs. There are inevitable trade-offs between the two. 13 Consider the example of homeschooling. Homeschooling is legal in some countries (such as the United Kingdom or Australia), illegal in some (Germany or Sweden), and closely regulated in others (France or Belgium). Motivations to provide home education vary, ranging from dissatisfaction with the school system, caring for children with special needs, or the desire to provide a specific religious or moral instruction. Some parents choose to homeschool because they find that the secular environment of the school system is not favourable to their children's spiritual growth. One way of ensuring substantive exit rights is to closely regulate homeschooling, to ensure that children receive a diverse basic education and are exposed to children from other backgrounds, which is favourable to a broadening of their future options. But a regulation or a prohibition of homeschooling does constitute an interference in parents' and religious group autonomy, if religious education is central to their practice. Tradeoffs are inevitable.
The exit fund is a bold interference-recall that it is made compulsory by the state and specifically targets a few religious communities-but its narrow scope makes the infringement on religious groups autonomy hard to justify. The policy only targets a few insular religious groups relying on collective ownership. This leaves out a lot of individual cases demanding similar attention: what about someone leaving a position of religious leadership without employment contract and therefore unable to access unemployment benefit schemes? What about long-term volunteering for a religious association that entailed relinquishing a paid job? What about people having made generous donations to their church, such as one's inheritance money, finding themselves deprived of the means to start a new life? In all these cases, individuals encounter high economic costs to exit that are not created by collective ownership. They too deserve attention. Burdening insular religious groups with compulsory exit fund seems disproportionate with regard to the few dissenters who would benefit from a minimal financial aid.
The second worry is that the proposal presumes that religious associations should be the primary bearers of the burdens of exit. This is not implausible in cases of residential living and collective ownership, but should it always be the case? What if, say, someone dissents after only a few months of joining an insular religious group? Should they receive a portion of the exit fund? In fact, the proposal eschews the question of fairness entirely by denying that exit fund should compensate for past work. This is surprising given that Hutterite case is precisely about dissenters claiming that their treatment is unfair-being left broke after long years of service. The terms of exit require a case-by-case assessment.
My proposal of a membership contract is also interventionist, but it is more wide-ranging and concerned with regulating the terms of exit. I take on board a solution that Spinner-Halev mentions, but quickly dismisses. Reflecting on the Hutterite case, he observes: The colony had a legal charter (as all Hutterite colonies do) stating that members who are expelled have no claim on the colony's property. A strictly legalistic approach, however, won't do. A legal document that says in exchange for work a person will be given food, along with clothes and shelter, but no money, makes the coercion of members quite possible. 14 Spinner-Halev rejects the idea of a legal document regulating the material aspects of the association on the grounds that it is perfectly compatible with an unreasonable clause-say, giving everything one's owns without getting anything in return-and therefore not protective enough. Yet, only a few lines later, one reads that "[n]ot all contracts are legal in the liberal state", giving the example of long-term unpaid hard work. Spinner-Halev provides an answer to his initial concerns: yes, some contract terms will be unreasonable, but if so, they can also be illegal.
My proposal, which I detail in Section 3, does not target only a handful of insular religious groups, but ranges across material relationships of various kinds, such as paid and unpaid work, residential living, and donations. I argue that the regulation of exit from religious associations should be brought closer to existing regulations of the exit of other voluntary associations, such as marriage and employment. Liberal states appear much more "hands-off" when it comes to "spiritual" relationships. Wrongly so, I argue.

The Analogy with Marriage and Employment
Nancy Rosenblum invites us to be inventive in the way we approach the associational landscape. She writes: "We can consider competing accounts of the salient features of an association, and draw alternative analogies between a given group and a business corporation, or church, or local government" (Rosenblum 1998). Analogies and disanalogies shed light on why we should treat different associations in the same way and on whether some special treatments are justified.
Any attempt to analogise "religious groups" with other associations is doomed to be challenged on empirical grounds. And rightly so: religious groups are incredibly diverse. They share features with both close associations, 15 such as families, and large corporations, such as multinational firms. 16 Some rely exclusively on personal relationships between individuals collectively engaged in a spiritual pursuit, while others function bureaucratically, manage collective assets, and hire employees. Religious groups include some insular communities and many loose networks. They can also combine different features. For instance, the pursuit of religious ends is sometimes combined with profit-seeking.
From a normative perspective, I am interested in religious groups as voluntary associations and from an empirical perspective, formally organised religious associations. In Cécile Laborde's disaggregative theory, how religious groups are construed depends on what is at stake. When a religious community is selling a property, it should be treated like a legal corporation. If a religious group is granted special exemptions from discrimination law, then it should to be treated as a voluntary association (Laborde 2017). This is the normative dimension I am focusing on. Laborde specifies that "only religious groups that are formally 14 Ibid., 73. 15 Lawrence Sager argues that organised religions are based on personal bonds, and it is in virtue of the right of close association that they can discriminate when employing ministers. Sager (2016). 16 Chiara Cordelli notes that a powerful case has been made for the democratisation of business firms, while the church remains "a democracy-free zone". A comparison between business firms and "large, hierarchical churches" runs throughout her argument. She concludes that the case for democracy in the church is even stronger than it is for the firm. Cordelli (2017). organized as associations (such as "organized religions" in UK law) are candidates for exemption from general laws". 17 My argument applies to these formal associations.
Bearing in mind the empirical limitations of the analogical exercise, I want to suggest that some lessons can be drawn from comparing the exit from religious associations with the exit from marriage and employment. In all three cases, the right of exit guarantees the consensual nature of membership and the possibility of opting out. Divorce protects marital freedom and helps prevent domestic abuse. The right to resign protects market freedom and helps prevent exploitation. When it comes to 'identificatory associations'-"groups that individuals join to pursue a conception of the good that is central to their identity and integrity", 18 the right of exit protects the freedom to freely pursue one's conception of the good, and prevents violations of integrity, being bound by rules that violate one's core commitments. The right of exit protects positive, as well as negative freedoms. Exit from different voluntary associations raise analogous concerns and it makes sense to think about them in relation with one another. 19 The analogy with marriage teaches us that a fair allocation of exit costs is crucial in a cooperative venture which is not profit-seeking. Spouses live and work together without formal monetary compensation. They are (presumably) motivated by the spirit of marital cooperation. Their engagement thus differs in a crucial respect from the economic exchange of an employer-employee relationship. Similarly, many religious members freely labour for and give to their community. It is (presumably) conviction and spiritual commitment, not economic gain, that move them.
Divorce law protects the economic interests of parties, regulates the division of shared assets, and when applicable the determination of maintenance (or alimony). Divorce law has many functions, but one of them is to ensure that the union remains voluntary and that the right of exit of the weaker party is substantive enough. For instance, maintenance aims at supporting a lesser-earning spouse for a specific duration after separation, allowing homemakers to reintegrate the labour market. 20 Maintenance lowers exit costs for the weaker party. The fact that spousal contribution was not motivated by economic gain does not exclude a duty of spousal support after separation, under certain circumstances specified by law. In the same vein, the spiritual motivation of religious labour should not eschew the discussion of the fair distribution of exit costs. The stronger party may have a duty to provide a temporary stream of income to the leaving party, given their contribution to the cooperative venture.
The state-sponsored marital institution organises a "contract-like moment" where parties publicly consent to a bundle of rights and duties set by the general social convention (to which they can add a premarital agreement). Parties make a commitment which entails legal obligations. If there is no "contract-like" moment in religious life, then it ought to be instituted, at least for certain types of relationships involving the exchange of goods. The state ought to 17 Ibid.,181. 18 Ibid.,174. 19 This is the approach endorsed by Robert Taylor who examines how exit can be facilitated by legal reforms and policy provisions. Taylor has in mind not only religious or cultural communities, but also intimate associations such as marriages, and corporations, such as business firms. Concrete examples include providing practical help to victims of abuse, such as restraining orders, shelters, and job opportunities. Taylor, Exit Left: Markets and Mobility in Republican Thought. 20 This is called "rehabilitative alimony". Rationales for alimony are contested and diverse, emphasising for instance the need of the lesser-earning spouse to maintain their prior lifestyle, a compensation for the unpaid work of the homemaker, or an insurance scheme against the costs of divorce. See: Scott and Scott (1998), Dagan and Frantz (2011). guarantee the voluntariness of religious associations formally recognised as voluntary associations. My contractual proposal fills this gap.
While a marriage can ambition to be governed by a principle of equality between parties, the relationship between a dissenter and a religious association remains intrinsically unequal. In that respect, a religious dissenter is more akin to an employee leaving their job. This is when the second analogy comes in. The state guarantees the freedom of employment contract, with an eye on the differential of bargaining power. Early dismissals must often be justified, and an employee can claim compensation in case of unfair dismissal. The work contract, signed by the two parties, detail the conditions of employment. They must align with employment rights. Labour regulation does not alleviate all exit costs-after resignation, an employee may still experience difficulties finding a new job-yet labour law attempts to mitigate them and enforce fair standards of termination.
If labour law offers adequate exit protection, why not simply challenge the politics of religious exemptions and extend labour law to a wider range of employment relationships within religious associations? I am sympathetic to this claim and generally favourable to more state interference in religious employment. I acknowledge, however, that religious associations differ from business firms in one crucial respect, namely their right to collective integrity. In a nutshell, religious associations are "special (but not uniquely special)" 21 insofar as they exhibit key associational interests in living by and interpreting their own conception of the good. 22 A policy that would make employment contracts compulsory for all labour within religious associations would clash with internal norms of volunteer contribution-unpaid (or minimally compensated) work may be a way of expressing one's deep religious commitment. Even if it desirable that the protections of labour law be extended to more employment relationships within the firm, some ought to remain voluntary.
Proponents of collective exemptions encounter what I call an "exit paradox". On the one hand, they insist that the voluntariness of the association is a necessary precondition for exemptions. In particular, Laborde writes that groups "must be voluntary, in the sense that members must be able to leave the group at no excessive cost". 23 She adds that this is necessary for presuming the consent of the members. Billingham also partly relies on the right of exit to justify granting corporate exemptions to religious organisations in spite of internal dissent. 24 In his view, the right of exit is a necessary feature of an argument for church autonomy-dissenters should not be coerced in practicing their religion in a way they find unconscionable. They should be able to leave.
On the other hand, proponents of collective exemptions must acknowledge that they have the unfortunate effects of undermining exit opportunities. The law has evolved to protect weaker parties in a context of power disparities-for instance, by protecting employees from unfair dismissal or tenants from ejection without prior notice. Members of religious associations benefiting from such exemptions are not adequately protected. The exit paradox is as follows: collective exemptions from some generally applicable rules are in part justified by 21 Laborde,Liberalism's Religion,177. 22 To be more precise, the two "integrity-related interests" outlined by Laborde are defined as follows: "Coherence interests refer to associations' ability to live by their own standards, purposes, and commitments. Competence interests refer to associations' ability to interpret their own standards purposes and commitments." Ibid., 175. 23 Ibid., 174. 24 I say partly because Billingham makes it clear that the right of exit is not sufficient to protect dissenters' interests, given the potentially high costs they incur. Billingham (2019a). relying on the right of exit; yet their very existence exacerbates the burdens of exit on a material level.
Given that many religious dissidents cannot benefit from some crucial legal protections, which alternative protection could be developed? I advocate a regulative framework that tackles the question of exit costs heads-on.

The Membership Contract
Consider the following proposal. Religious associations should issue a membership contract for members working for, residing in, or donating money to the association under a regime of legal exemptions. 25 The document should stipulate a policy of termination, e.g. whether the leaving member is eligible for a compensation from the institution and under which conditions. The state guarantees the validity of the contract and subsequent enforcement. Religious associations bear the responsibility of issuing it. To qualify, the exchange of good must be above a given threshold-to be specified in light of the legal and economic context. 26 Religious associations will be provided with clear guidelines about when a membership contract is required by law.
At the outset, it is worth highlighting that the proposal is limited in two ways. First, it does not tackle all the burdens associated with exit, such as social costs as well as some psychological costs. Some costs are simply beyond state intervention. Second, it targets specific relationships within religious associations that involve an exchange of goods. As such, it is not the "spiritual" relationship which is at stake, but relationships of employer-employee, landlordtenant, or recipient-donor.
Still the proposal enhances the protection of exit rights in three distinct ways. It improves publicity, facilitates the negotiation and the contestation of exit terms.
With respect to publicity, the membership contract brings the right of exit to the awareness of each member signing the contract. The association is under the obligation to acknowledge and communicate this right. Improved publicity contributes to making exit less "unthinkable" to members. 27 To draw another analogy with marriage, the legal fact of divorce undeniably makes very vivid in people's minds the possibility of separation from marriage. This is not to say that divorce as such causes separation, or that it substantially alleviates its costs. A potential by-product of publicity is normalisation. Exit is recognised by the association as belonging to the repertoire of available options. Members are informed that they can always opt out. Exit is not framed solely as a regrettable breach of loyalty, but as a protected right, publicly acknowledged by the association.
This proposal implies that religious associations must recognise exit as a legitimate option for all members. Is this a too burdensome requirement? It ought not to be. Religious associations may still condemn exit morally and depict it as both wrong and harmful. They must, however, acknowledge it legally when they issue membership contracts. The 25 Membership contracts are not mandatory for members who are merely attending religious events. They could, however, be recommended, as they improve publicity. Making them mandatory would unduly burden religious associations, by increasing the costs of entry, without significantly reducing the costs of exit. 26 This threshold allows to exclude insignificant exchanges of good. A one-off donation of a few pounds or a few hours of voluntary cleaning should not warrant the use of a membership contract. 27 Okin sees the unthinkable nature of departure as the biggest obstacle to exit. Okin, '"Mistresses of Their Own Destiny": Group Rights, Gender, and Realistic Rights of Exit'. requirement of publicity does not extend beyond contracts-religious leaders are by no means required to remind the audience of their right of exit before the worship. When they enter contractual relationships with some of their members, however, a formal acknowledgement of their right of exit is in order.
The idea of requiring associations to publicise the right of exit resonates with some views on freedom of association that advocate compulsory disclosure of relevant membership information. In particular, Jennifer Gerarda Brown argues that members should make an informed decision when joining an organisation that discriminates against homosexuals. 28 She advocates "Informed Association statutes" that compel organisations to reveal their discriminatory policies to individuals who wish to join or donate to the association. For instance, written consent forms can make explicit the basis on which discrimination is conducted. Individuals can thus associate-or choose not to associate-knowingly.
My proposal relies on the same intuition. Knowledge about the conditions of membership matter and should be explicit. It differs on a crucial point, however: organisations are not given any option. Denying the right of exit is a nonstarter, as the voluntariness of the association is nonnegotiable.
In addition to publicity, the contract enhances the possibility of negotiating and contesting the terms of exit. It does so mostly by facilitating information sharing. To be sure, it is plausible that only few new converts enter a religious organisation with the foresight of negotiating a favourable departure. However, it is also plausible that the normalisation of membership contracts could change that. A the very least, it opens a window of opportunity for doing so.
Contestation is a more promising path. To be adequately protective, the introduction of membership contracts must be accompanied by a legislative and judicial development on what constitutes unreasonable contract terms.
Similar regulations exist to protect vulnerable contracting parties, including the rights of consumers. 29 When it comes to consumers' rights, a typical definition of "unfair contract terms" reads as follows: "A term is unfair if, contrary to the requirement of good faith, it causes a significant imbalance in the parties' rights and obligations under the contract to the detriment of the consumer." 30 According to the Competition & Markets Authority, this phrasing includes a requirement of transparency-terms should be intelligible-, a respect for consumers' "legitimate interests", and the exclusion of terms that blatantly disadvantage the weaker party. 31 This leaves courts with a margin of appreciation to assess unfairness on a case-bycase basis. If deemed unfair, a term is not binding on the consumer. I advocate a similar approach with regard to membership contracts. Their introduction should be accompanied by a legislative act specifying in rather general terms what counts as unreasonable contract terms. The possibility of exit should be taken as a yardstick. Unreasonable contract terms could read as follows: "A term is unreasonable if it significantly burdens the possibility for a member to exit the association." In the same vein as the "unfair contract 28 Brown (2002). For a thorough discussion of "neo-libertarian views", see: Koppelman and Wolff (2009). 29 For a discussion of the Directive 93/13 on Unfair Terms in Consumer Contracts in European law, see Nebbia (2007). For a discussion of the Unfair Contract Terms Act 1977 in UK law, see Lawson (2017). 30 Consumer Rights Act 2015, section 62 (4): http://www.legislation.gov.uk/ukpga/2015/15/section/62/enacted [accessed 10 March 2020] 31 CMA, "Unfair contract terms explained", 31 July 2015: https://assets.publishing.service.gov. uk/government/uploads/system/uploads/attachment_data/file/450410/Unfair_Terms_Explained.pdf [accessed 11 March 2020] terms" in commercial law, this would include a requirement of transparency-the possibility of exit, as well as its terms should be clearly stated and intelligible. It would also require that the terms do not significantly disadvantage the weaker contracting party in a way that impedes the possibility of exit. Note that an asymmetry of contribution between parties is not a problem as such. It becomes a problem only when it obstructs exit.
This implies the following: (a) A clause requesting to give everything one owns without anything in return-not even a minimal exit fund-will most likely be found unreasonable under all circumstances, as it significantly burdens the possibility of exit for members. (b) A large donation-giving one's house or inheritance-will only be troublesome if it leaves the member in a state of deprivation which makes exit difficult. (c) The amount of volunteer work will begin to be problematic when it jeopardises the possibility of reconversion altogether. At the very extreme, a work which is full-time, long-term, unpaid, and without the possibility of certification or compensation after leaving should qualify as unreasonable. By contrast, temporary volunteer work around special occasions should not raise any concern. (d) Nonmaterial costs, such as excommunication or shunning remain beyond state intervention. Yet the fact that the right of exit is publicised by the membership contract may somehow mitigate the dissuasive power of informal sanctions by "normalising" exit.
Like in the case of consumer's rights, this leaves courts with a margin of appreciation to assess contract terms on an individual basis. A context-sensitive assessment is necessary to determine whether a specific arrangement generated unreasonable exit costs for a specific individual. The same arrangement will burden different individuals differently, depending on their respective financial circumstances. Plaintiffs will have to convince the court that their ability to exit was significantly burdened by the contract terms. They can request compensations or a temporary maintenance.
The fairness of the arrangement itself should not be the focus of assessment. Courts should not take as a premise that religious transactions function on a basis of reciprocity and aim at balancing gains and costs. Asymmetric contract terms are acceptable to the extent that they do not result in unreasonable exit costs, i.e. costs that significantly impede individuals' ability to leave the association. Court decisions will inevitably interfere in religious internal affairs. The following section argues that this interference is justified.

A Justified Interference
A strength of my proposal is its congruence with the existing legal framework. 32 Although there has been a historical reluctance to construe "spiritual bonds" in terms of contractual relationships, the law is now evolving (Sandberg 2014). This is especially visible in discussions about the "ministerial exception". In the case Hosanna-Tabor Evangelical Lutheran Church and School v. Equal Employment Opportunity Commission, 33 the United States Supreme Court denied ministers of religion the status of employees, crucially exempting the recruitment of office holders from nondiscrimination law. A similar legal pattern occurred in twentieth century UK law. From Re Employment of Church of England Curates in 1912 34 to Coker v. Diocese of Southwark in 1998, 35 curates of the Church of England have been denied the status of employees. Courts have issued similar rulings in cases involving religious groups other than the Church of England. It was held that there was a fundamental incompatibility between the spiritual nature of office holder positions, and the secular nature of a contract. In the House of Lord decision David v. Presbyterian Church of Wales, 36 Lord Templeman claimed that the plaintiff's "duties are defined and his activities are dictated not by contract but by conscience" and "by no stretch of imagination can such an agreement constitute a contract of service." 37 Yet we no longer need to stretch our imagination to construe "spiritual" relationships in contractual terms. In the United Kingdom, a significant legal change has occurred at the turn of the twenty-first century, following the case Percy v. Church of Scotland Board of Mission. 38 The alleged incompatibility between holding an office and being an employee was decisively challenged. Spiritual bond and contractual relationship need not be mutually exclusive. It is now established that spiritual leaders may be entitled to employment rights, although it remains the case that many ministers of religion in the United Kingdom are left unprotected. It is still unclear when an office should count as a work contract. 39 The historical resistance against employment contracts should not obscure the fact that religious associations are generally regulated by contract law. In UK law, non-established religious groups are treated like sport clubs. 40 They are unincorporated associations, which means that they do not have a separate legal personality from their members (unlike corporations), unless they acquire the status of corporation or trust by becoming limited companies if they want to own and sell property. There exists a "quasi contractual" bond between members, which are governed by the rules of the association. This is referred to as the doctrine of "consensual compact".
Not only does my proposal fit with the contractual framework regulating religious associations, but it can also be compatible with the legal principle of non-interference. The principle of non-interference generally prevents tribunals from interfering in religious affairs and enforcing contracts. 41 Religious associations fall under contract law, but the contracts are not legally enforceable by the state. 42 The ultimate sanction a religious association can impose on its members is expulsion. Yet, exceptions to the principle of non-interference occur in UK law when a property or a financial interest is at stake: this is known as the Forbes v. Eden 33 (2012) : Sandberg, Religion, Law and Society, 106-107. 38 [2005] UKHL The case concerned a sex discrimination claim against the Church of Scotland. 39 Russel Sandberg deplores that "there is little guidance as to what factors are likely to render the relationship contractual". Sandberg,Religion,Law and Society,115. 40 Sandberg writes: "Religion in the United Kingdom is organised on a private basis: in short, the legal status of religious groups is similar to that of a sports club." Sandberg (2011). 41 Sandberg,Law and Religion,74. 42 In addition, enforcing some religious contracts, such as religious vows, would clash with the protection of fundamental liberties. exception. 43 So religious associations are mostly self-regulated and their rules are recognised as binding on their members. Secular courts refrain from enforcing these rules in virtue of the principle of non-interference, but they do interfere on some practical matters. Because it focuses on the practical aspect of the dissolution of membership, the proposal need not violate the principle of non-interference. Instead, it leaves parties with a large degree of autonomy in stipulating the content of their mutual obligations. The set up of the contractual agreement gives them the opportunity to negotiate the terms of their relationship with one another, including the conditions of its cancellation. Parties benefit from the freedom to contract, limited by the exclusion of "unreasonable contract terms".
The devil is in this detail. Does judicial review of the reasonableness of contract terms potentially constitute an illegitimate interference in religious group autonomy? I argue that there is indeed a non-negligible interference, as unreasonable contract terms will be ruled out for obstructing exit, but this interference is justified by the compelling interest to protect individuals' right of exit.
Religious group autonomy (often called "church autonomy") refers to the right of religious organisations to manage their internal affairs free of government interference. Such a collective religious liberty is sustained by various justifications and challenged by different critiques. 44 Without giving justice to a large debate, I want to point out that my proposal is compatible with at least some accounts of religious group autonomy, based on the value of freedom of association and religious freedom.
I take it that the democratic state has final authority over questions of "jurisdictional boundary" (Kompetenz-Kompetenz as Laborde puts it), as it must arbitrate conflicts about justice while representing the interests of individual citizens. 45 I therefore reject the view that churches have a separate jurisdiction from the state, thereby limiting state sovereignty. But even leaving jurisdictional conceptions of church autonomy aside, concerns over potential infringement of church autonomy remain. As Paul Billingham rightly observes, even if we agree that the state is not "externally limited" by the jurisdiction of churches, it remains "internally limited by liberal democratic principles themselves" (Billingham 2019b). Given these internal limitations, some intrusions within groups' internal affairs may still be illegitimate. My proposal has the potential to rule out religious ways of life that involve irreversible commitments-for instance, giving everything that one owns to an association or some forms of long-term voluntary labour. This is a significant interference in collective autonomy. 46 This interference is ultimately justified. It should be clear by now that my proposal is compatible with an account of religious group autonomy based on freedom of association. Dissociation is an integral part of freedom of association, together with the right to exclude and the right to organisational autonomy (Brownlee and Jenkins 2019). Yet there is something special about the right of exit within the bundle of associational rights. The right to exclude and the right to organisational autonomy are collective rights that stem from freedom of association. The right to exit, as applied to individuals, has a different status. While the former are 43 Sandberg,Law and Religion,[76][77] For an illuminating discussion on collective religious liberty, see: Schwartzman et al. (2016), ibid. 45 Here I endorse Cécile Laborde's objection against religious institutionalism. Laborde responds to authors such as Richard Garnett, Steven D. Smith, and Victor Muniz-Fraticelli. See: Laborde, Liberalism's Religion. Chapter 5. I 46 I acknowledge that this is also an interference in individual autonomy. I would argue that the benefits of substantive exit rights in terms of protecting individual autonomy outweighs the costs of restricting irreversible commitments. collective associational rights, the latter is an individual right "vis-à-vis groups". 47 It is a right that members hold "against the community" 48 and that the state ought to protect, even if it implies some loss in organisational autonomy. The special status of the right of exit justifies some interference in the internal governance of religious associations. The yardstick of my account of unreasonable contract terms is precisely substantive exit rights.
In addition, my proposal is compatible with an account of religious group autonomy grounded in the special value of religious freedom. Douglas Laycock argues that the "right to church autonomy" is protected in US law under the free exercise clause (as opposed to the establishment clause, which targets government support for religion) (Laycock 1981). The free exercise clause protects not only the freedom to believe and to practice one's religion, but also the right of religious groups to conduct religious activities autonomously. Government interference in the internal affairs of a church undermines "the very process of forming the religion as it will exist in the future" 49 -and Laycock specifies that church labor relations fall within the realm of church autonomy. 50 In Laycock's view, volunteer work and donations should plausibly be protected against government oversight.
I argue, however, that Laycock's own account of legitimate government interference allows for my proposal. In Laycock's view, interference is justified when there is a sufficiently compelling state interest that outweighs church's interest in autonomy. "Voluntary affiliation" with the group is depicted as "the premise on which group autonomy depends". 51 The protection of individuals' right to exit must therefore be a compelling state interest. When it comes to obstacles to exit, however, Laycock, only considers the case of physical coercion and quickly dismisses "brainwashing", namely cases of psychological abuse. He completely overlooks the question of economic costs to exit. Interestingly enough, when Laycock provides an example of legitimate state interference under the establishment clause, he mentions state support to Catholic education for disadvantaged children. 52 In this particular case, socio-economic difficulties provide a compelling enough state interest to support religious education. 53 I do not see why the socio-economic conditions of leavers should not also be a matter of state concern, given that it bears on voluntariness, the very premise of group autonomy.
The normative significance of protecting exit rights outweighs the loss of collective religious autonomy. Yet the implementation of the proposal should aim to minimise unnecessary interference as much as possible. For example, it need not require large-scale collections of data about religious behaviours. 54 In particular, state oversight need not require the compulsory disclosure of the identity of group members. The possibility for members, or former members, to raise a complaint for not having signed a membership contract is a 47 Ibid. 48 Kukathas, 'Are There Any Cultural Rights? ', 116. 49 Ibid., 1391. 50 Ibid., 1398. 51 Ibid., 1405. 52 Ibid., 1387 Plausibly combined with the fact that they are children. I wonder if Laycock would make a similar case for say, state support to faith-based rehabilitation programmes. 54 This worry has been raised in relation to the application of procedural evenhandedness, a model of statereligion relation whereby government allocates resources to religious groups without being affiliated with one of them. See: Jobani and Perez (2017). I thank an anonymous reviewer for raising this point. possible tool of oversight. Arguably many will not complain. But if a few do complain, and an association is sanctioned for failing to provide membership contracts, this might create a real incentive for all associations to comply with their legal duty.

Conclusion
María's story with the Opus Dei happened some decades ago, in the 1950 and 1960s. The Opus Dei has been repeatedly criticised in the following decades. It has now made publicly available on its website a formal acknowledgment of members' right of exit: Obviously, if anyone is thinking of leaving Opus Dei, they are invited to think their decision over carefully in God's presence, since they joined the Work out of a conviction that God was calling them to this path. However, nobody is kept in Opus Dei against their will. When they formally request permission to leave, the request is always granted, including in cases of people who had been permanently incorporated into Opus Dei. 55 In light of the argument of the paper, this public acknowledgment inspires two thoughts. First, requesting that religious associations acknowledge the right of exit by issuing a membership contract might not be as controversial as it might seem at first glance. If even a controversial group such as Opus Dei publicises the right of exit on its website, it must be the case that many other less rigorous organisations would have no objection to officially acknowledge it. The formal acknowledgement of the right of exit may not be so unpopular after all.
And here is the second thought: if even the Opus Dei publicises some kind of formal commitment acknowledging the right of exit, isn't the proposal just adding another layer of formality to an already too formal right? What concrete difference do formal rules make in contexts of multifaceted pressures and diffuse coercion? Suppose that the "contract" endorsed by María would have been accompanied by a signature of the membership contract. María could have well signed an asymmetric deal where "she gives up everything without getting anything in return". But she should have been informed that this was conditional to her right of exit guaranteed by the state. Upon leaving, she could have sought redress in a secular court on the basis of this document and argue that (1) the coercive measures she underwent before leaving invalidate the contract (2) the very clause of "giving everything without anything in return" was unreasonable. There are good reasons to believe that María would have been in a better position to receive compensation.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.