The Arbiters of Faith: Legislative Assembly of BC Entanglement with Religious Dogma Resulting from Legislative Prayer

Daily sittings of the Legislative Assembly of British Columbia (BC) begin with a prayer or reflection delivered by a Member of the Legislative Assembly (MLA). MLAs have the option of using a sample prayer from a list of five provided by legislative staff, or of delivering a prayer of their own devising. Sittings began with ‘prayers’ until November 2019 when the Standing Orders were amended to ‘prayers and reflections’ (Legislative Assembly of BC 2019). This change is part of an ongoing process of reform, led by the Office of the Clerk. In addition to this nominal change, the Office of the Clerk is currently reviewing the list of sample prayers “to ensure that prepared prayers provide a breadth of non-religious reflections, as well as prayers from major religious groups” (K. Ryan-Lloyd, 2019, Acting Clerk of the House, correspondence with authors, August 21). Part of this review involves the Office of the Clerk reaching out to various faith groups in BC to solicit their feedback on the sample prayers (K. Ryan-Lloyd, 2020, Clerk of the House, correspondence with authors, March 10). Notably, the Office of the Clerk included nonreligious organizations in this consultation; such groups are very often excluded from these types of consultation (see for example Beaman, Steele & Pringnitz 2018: 52, referencing Fox v. Secretary of State for Education). This review is timely, as legislative prayer is fraught with problems: it trivializes a potentially sacred and private act;1 has a tendency to promote religions (and denominations within those religions) at the exclusion of others (Marshall 2002); is inherently exclusory to those from faith traditions which may not include practices that could be considered as ‘prayer’ or for whom the limitations presented by the legislature (location, time, duration, etc.) would make prayer in this context impossible; excludes non-believers; and seemingly represents a violation of the state’s duty of religious neutrality (Phelps Bondaroff et al. 2019: 17–38; Delahunty 2007; Mouvement laïque québécois v. Saguenay 2015). While the practice of starting sittings with prayers is exclusionary and potentially unconstitutional, the process involved in revising and amending the provided sample prayers is equally problematic. The review of the sample prayers is ongoing. To accomplish this task, bureaucrats in the Office of the Clerk face a number of challenges, both practical and constitutional. They will have to first identify religious and belief groups that are present in BC, then select a reasonable number of those groups to include on the list, and finally, identify prayers and reflections that are representative of those traditions. Each of these steps raises practical questions with respect to how bureaucrats can actually make these decisions, as well as to more fundamental questions regarding the appropriateness of the decisions being made. This article explores these difficulties. Given the exclusory nature and challenges surrounding legislative prayer, this practice should ultimately be Bushfield, I and Phelps Bondaroff, TN. 2020. The Arbiters of Faith: Legislative Assembly of BC Entanglement with Religious Dogma Resulting from Legislative Prayer. Secularism and Nonreligion, 9: 8, pp. 1–16. DOI: https://doi.org/10.5334/snr.140


Introduction
Daily sittings of the Legislative Assembly of British Columbia (BC) begin with a prayer or reflection delivered by a Member of the Legislative Assembly (MLA). MLAs have the option of using a sample prayer from a list of five provided by legislative staff, or of delivering a prayer of their own devising. Sittings began with 'prayers' until November 2019 when the Standing Orders were amended to 'prayers and reflections' (Legislative Assembly of BC 2019). This change is part of an ongoing process of reform, led by the Office of the Clerk. In addition to this nominal change, the Office of the Clerk is currently reviewing the list of sample prayers "to ensure that prepared prayers provide a breadth of non-religious reflections, as well as prayers from major religious groups" (K. Ryan-Lloyd, 2019, Acting Clerk of the House, correspondence with authors, August 21). Part of this review involves the Office of the Clerk reaching out to various faith groups in BC to solicit their feedback on the sample prayers (K. Ryan-Lloyd, 2020, Clerk of the House, correspondence with authors, March 10). Notably, the Office of the Clerk included nonreligious organizations in this consultation; such groups are very often excluded from these types of consultation (see for example Beaman, Steele & Pringnitz 2018: 52, referencing Fox v

. Secretary of State for Education).
This review is timely, as legislative prayer is fraught with problems: it trivializes a potentially sacred and private act; 1 has a tendency to promote religions (and denominations within those religions) at the exclusion of others (Marshall 2002); is inherently exclusory to those from faith traditions which may not include practices that could be considered as 'prayer' or for whom the limitations presented by the legislature (location, time, duration, etc.) would make prayer in this context impossible; excludes non-believers; and seemingly represents a violation of the state's duty of religious neutrality (Phelps Bondaroff et al. 2019: 17-38;Delahunty 2007; Mouvement laïque québécois v. Saguenay 2015). While the practice of starting sittings with prayers is exclusionary and potentially unconstitutional, the process involved in revising and amending the provided sample prayers is equally problematic.
The review of the sample prayers is ongoing. To accomplish this task, bureaucrats in the Office of the Clerk face a number of challenges, both practical and constitutional. They will have to first identify religious and belief groups that are present in BC, then select a reasonable number of those groups to include on the list, and finally, identify prayers and reflections that are representative of those traditions. Each of these steps raises practical questions with respect to how bureaucrats can actually make these decisions, as well as to more fundamental questions regarding the appropriateness of the decisions being made. This article explores these difficulties.
Given the exclusory nature and challenges surrounding legislative prayer, this practice should ultimately be Ian Bushfield and Teale N. Phelps Bondaroff Daily sittings of the Legislative Assembly of British Columbia (BC), Canada begin with a prayer or reflection delivered by a Member of the Legislative Assembly (MLA). MLAs have the option of using a sample prayer from a list of five provided by Legislative staff, or of delivering a prayer of their own devising. The Office of the Clerk is currently revising the list of sample prayers to include a greater diversity of beliefs. To accomplish this task, bureaucrats face a number of practical and constitutional challenges. This article explores these challenges. The Office of the Clerk will have to first identify religious and belief groups that are present in BC, then select a reasonable number of those groups to include on the list, and finally identify prayers and reflections that are representative of those traditions. Each of these steps raises practical questions with respect to how bureaucrats can actually make these decisions, and to more fundamental questions regarding the appropriateness of the decisions made. Given these challenges, this paper concludes that the practice of offering sample prayers should be avoided, and that MLAs should be called upon to deliver prayers and reflections of their own devising, or that legislative prayer be abolished.
abolished. However, presuming that 'prayers and reflections' will continue to open sittings of the BC Legislature for the foreseeable future, this study explores some of the challenges that the Office of the Clerk will likely face as it attempts to develop a list of 'sample prayers' to provide to BC legislators. In so doing, we offer an in-depth case study that will contribute to the broader discussion around separation of religion and government in BC and Canada.
This paper outlines the current practices surrounding legislative prayer in BC. It then examines existing jurisprudence relating to issues surrounding separation of religion and government to identify some of the possible constitutional and legal challenges that the Office of the Clerk may face as it considers expanding the list of sample prayers on offer. The various approaches that could be adopted by the Office of the Clerk to amend the list of sample prayers, and the myriad of practical challenges associated with these approaches, are then explored. The paper concludes that given these challenges, the practice of offering sample prayers should be avoided, and that MLAs should be called upon to deliver prayers and reflections of their own devising, or that legislative prayer be abolished.

Current legislative practice in BC
The daily routine business of the BC Legislature is established by Standing Order 25, which details that morning and afternoon sittings will begin with 'prayers and reflections' as the "the first item of business after the arrival of the Speaker's Procession in the Chamber" (Legislative Assembly of BC, Standing Order 25; Ryan-Lloyd et al. 2020: 97). Prayers and reflections are delivered by either an MLA, or "by the Speaker or invited faith leaders or Indigenous leaders or Elders," the latter of which typically occurs on days when a Speech from the Throne is read (Ryan-Lloyd et al. 2020: 97). Parliamentary practice in British Columbia (Ibid.) notes that: unlike other jurisdictions, Prayers and Reflections are held with both officers and strangers present and are the only proceedings not transcribed verbatim for publication in the Hansard transcript. Usually the Speaker will invite a Member to "lead the House in prayer or reflection," the Member having been previously designated by their Caucus Whip.
When delivering prayers, MLAs are given the option of delivering a prayer or reflection of their own devising, or of reading one of five sample prayers (see Appendix 1). These prayers "may be of any faith or denomination, may be reflective of different cultural traditions, may be a traditional land acknowledgement, and may also be a moment of reflection" (Ibid.).
The recent House of Prayers report, which analyzed the content and religiosity of these prayers between 2003 and 2019, found that MLAs chose to deliver a prayer from the sample prayer list precisely half of the time (50.0%) (Phelps Bondaroff et al. 2019: 73). This report also found that MLAs have tended to prefer some of the sample prayers over others, and that MLAs will occasionally make minor alterations to the prayers, or deliver sample prayers in a number of combinations (Table 1).
Apart from the five sample prayers and combinations thereof, the content of the prayers devised and delivered by MLAs varied considerably; they included religious invocations, acknowledgments of tragedies and natural disasters and those impacted by them (see prayer delivered by R. Visser, 2005, March 2), and well-wishes to members experiencing hardships. The House of Prayers report identified prayers where MLAs proselytized (see prayer delivered by V. Roddick, 2008, February 25), recited poetry (see prayer delivered by L. Popham, 2009, September 21), praised government policy (see prayer delivered by N. Letnick, 2011, October 19), extolled principles of democracy (see prayer delivered by D. Routley, 2014, February A cursory examination of the current list of sample prayers clearly underscores the need for amendment or abolition. All of the current sample prayers have a structure typical of Abrahamic faith traditions, and all end in ' amen.' While the House of Prayer report coded two of these prayers as 'secular' given their absence of overt religious language other than ' amen' (see prayers 3 and 4 in Appendix 1), the other three sample prayers (1, 2, and 5) are overtly religious and were coded as 'non-sectarian' (Phelps Bondaroff et al. 2019: 48). These three prayers contain overt references to a deity, requests for divine guidance, and are replete with religious language (Ibid.). While all five of the prayers can reasonably be seen as ecumenical, their structure and content essentially excludes the religious practices of non-Abrahamic religions and non-believers.
That this list could be considered 'secular' betrays what Beaman (2010: 280) refers to as the "tainted neutrality of the secular," whereby 'secular' is simply "another name for a vague Christianity" (Ibid. quoting Jakobsen & Pellegrini 2004: 114). In these prayers we can see "the ancestral ghost of majoritarian religion continue to haunt present-day iterations of the boundaries of religious freedom" (Beaman 2020). In an increasingly multicultural society, which includes a plethora of faith traditions and a growing number of non-believers, this limited list of five Abrahamic-style prayers is indeed a very narrow sample. This is ultimately an exclusionary practice, which entrenches a specific and narrow understanding of an acceptable "vision and version of a history that belongs only to a particular segment of society, which is in turn located at the apex of a hierarchy of citizenship and belonging" (Beaman 2020). As Beaman (2020) notes, this has an obvious impact: certain groups, particularly religious minorities and the nonreligious, are excluded from the narrative of 'we' in the public sphere, and as contributors to nations in the present tense, the past, and potentially the future. What is worth preserving and protecting does not belong to them and does not originate with them.
What we see in the current practice of beginning sittings of the BC Legislature with prayer is an unspoken understanding of what constitutes religion, and when this understanding is examined, it is revealed to be highly exclusionary. As US Supreme Court Justice Ginsberg noted in her dissent in American Legion v. American Humanist Association, those not included on the list are treated as "outsiders, not full members of the political community" (Ginsberg quoting County of Allegheny in American Legion v. American Humanist Association 2019). Here, in a similar vein to Beaman, the goal is not to "turn Christianity into the villain of the story, but rather to ask how existing power relations constitute an impediment to social inclusion, living well together, and equality" (2020).

Arbitrating what constitutes a religion
Before soliciting prayers and reflections for inclusion in a revised list, the Office of the Clerk must first decide what constitutes a religion. Simply approaching certain organizations or individuals at the exclusion of others exposes a judgement call on the part of the Office of the Clerk, regardless of the existence of explicit rules, procedures, or definitions relating to what constitutes a religion. By making such a choice, the Office of the Clerk is effectively validating or elevating some groups or individuals and invalidating others, or as Sullivan describes it, "creating a hierarchy of orthodoxies" (2005: 3; and see Beaman 2010: 267).
Naturally then, the first and most important question to consider is whether the government should have a role in determining what is recognized as a religion. If the state lacks the authority to adjudicate this question, then the question as to how it might go about accomplishing this becomes moot. In many countries, there is a "government office that is in charge of deciding what qualifies as religion, for legal purposes" and religion must be "organized and officially licensed in order to function" (Sullivan 2009(Sullivan : 1185. This is not the case in Canada.
Canadian jurisprudence has consistently found that it is inappropriate for the state to adjudicate what constitutes a religion (Ogilvie 2015). This conclusion is unsurprising given the fact that defining 'religion' continues to be a subject of intense debate in academia, where there are competing sociological and phenomenological approaches (see Siding Jensen 2019; Beaman 2010;Spiro 2004;Beaman 2003;Platvoet & Molendijk 1999;Sullivan 1998;Plavoet 1990;andFreeman 1982-1983). In addition, many of these approaches are derived from historically Judeo-Christian understandings, which have not always been able to accommodate spiritual beliefs of non-European cultures (Oman 2013;Southwold 1978). For example, in a legal dispute over whether a smudging ceremony in a Port Alberni, BC, school violated the school's requirement to operate on "strictly secular and non-sectarian principles", the Nuu-chah-nulth Tribal Council argued that "smudging is a cultural practice, not a religious one" (Servatius v. Alberni School District 2020: para9). Ultimately, the question remains alive as the Supreme Court of British Columbia decided this case on other grounds. Similar examples can be found in the United States of America (USA). In Lyng v. Northwest Indian Cemetery Protective Association (1988) the US Supreme Court "allowed ' development' to continue at the expense of the desecration of sacred spaces and the destruction of Native-American spirituality" (Beaman 2003: 319). If the road construction project in Lyng had been proposed through a church building or cathedral it is plausible that the verdict would have been different or that the case would not have been necessary at all. Another seminal example from the USA can be found in Employment Division v Smith (1990). In this case, the Supreme Court ruled that the First Amendment's clause concerning the free exercise of religion, "does not protect individuals from neutral laws that incidentally inhibit, or even preclude, the practice of their religion" (Marin 1990(Marin -1991(Marin : 1431. And as a result, the Court "refused to exempt the Native American respondents, dismissed from their jobs for participating in a religious ritual involving the ingestion of peyote, from Oregon's law prohibiting peyote use" (Ibid. 1432; see also Sullivan 2009Sullivan : 1188. Such cases touch upon attempts to differentiate between religious and spiritual practices, and incline one towards Sullivan's (2014Sullivan's ( : 1132Sullivan's ( -1133) conclusion that we do not know how to define religion, and the available evidence suggests that we should probably stop trying. Not because those practices that are gathered under the term religion are not important but because the religiousness or not of a particular activity cannot be determined with sufficient specificity and therefore should not matter in law. We will have to find our collective reassurance elsewhere.
The semantic debate over religion is far from settled. Time and again, bureaucrats may voluntarily, or be mandated to, weigh in on this debate in their official capacity. While there are a number of examples of this, here we will consider two: The determination of charitable status and recognition to solemnize marriages.
An organization may qualify for charitable status if it has as a purpose the ' advancement of religion.' The Canada Revenue Agency (CRA) therefore maintains a summary policy on religion to aid bureaucrats making a determination regarding the status of an applicant. The policy stipulates that: to advance religion in the charitable sense means to promote the spiritual teachings of a religious body and to maintain doctrines and spiritual observances on which those teachings are based. There must be an element of theistic worship, which means the worship of a deity or deities in the spiritual sense (CRA 2002).
While this definition attempts to be as inclusive as possible, it clearly excludes a number of religious traditions, most notably those that are non-theistic. Unlike theistic religions that "presuppose that ultimate reality is a personal God (as in monotheism)… other traditions presuppose that ultimate reality is impersonal" (Delahunty 2007: 541; and see GR. Stone cited in Ibid., p. 540-541). These non-theistic traditions tend to include concepts of "an allpervasive energy or force that may or may not include the individual" (Berry 2005: 636; and see Koenig et al. 2014: 530).
The contentious nature of this definition was reflected in a recent Canadian Federal Court of Appeal ruling where the court rejected the claim that the Church of Atheism of Central Canada constituted a religion. The ruling did note deficiencies in current efforts to define religion in the context of what constitutes a charity. In her decision, Justice Rivoalen (Church of Atheism of Central Canada v. Canada 2019: para.21), stated that: I agree with the appellant that the requirement that the belief system have faith in a higher Supreme Being or entity and reverence of said Supreme Being is not always required when considering the meaning of 'religion'. The appellant rightfully pointed to Buddhism as being a recognized religion that does not believe in a Supreme Being or any entity at all (citing South Place Ethical Society, Barralet and Others v. AG., 1980, 1 W.L.R. 1565, at p.1573).
In other words, the Court disagreed with the CRA's policy. Justice Rivoalen (Church of Atheism of Central Canada v. Canada 2019: para.10), also noted that: For something to be a 'religion' in the charitable sense under the Act, either the Courts must have recognized it as such in the past, or it must have the same fundamental characteristics as those recognized religions. These fundamental characteristics are not set out in a clear 'test.' Justice Rivoalen then proceeded to outline some of these characteristics, which included that "the followers have a faith in a higher power such as God, entity, or Supreme Being; that followers worship this higher power; and that the religion consists of a particular and comprehensive system of faith and worship" (Ibid. citing Syndicat Northcrest v. Amselem, 2004 SCC 47, 2 S.C.R. 551, at para.39).
We can see the "ancestral ghost of majoritarian religion" lurking in Justice Rivoalen's characteristics of a religion (Beaman 2020). Here, "Protestantism, and to some extent Catholicism, are constructed as the normal against which the ' other' is established" and from which the characteristics of a religion are determined (Beaman 2003: 313). Such is the hegemony of these religions that they are used to establish what constitutes a 'normal' religion and to form the basis of a definition of religion itself (Beaman 2003: 214).
In other contexts, the laws governing marriages in BC, Ontario, and Quebec give bureaucrats the power to grant authority to religious representatives to solemnize marriages (BCHA 2017). Notably though, each of these acts does not define 'religion' explicitly. For example, the Marriage Act of BC defines a 'religious body' as "any church, or any religious denomination, sect, congregation or society" (Marriage Act 1996). Such a definition is circular and not useful when attempting to effectively define what constitutes a religion.
It is again telling the definition would begin with the word ' church' and then append other examples. The use of this term also potentially muddies the definition, as it can have different meanings to different Christian sects. As Sullivan (2014Sullivan ( : 1130 notes, "there is arguably no analogy to the church outside Christianity," and within Christianity there is considerable debate as to whether this refers to a group of two or three believers gathered in Jesus Christ's name, or a hierarchical structure, or even a building (Ibid. 1127; and see inter alia Hebart 2009;Dulles 2002). While this is not the place for a detailed exploration of ecclesiology, suffice it to say the use of this term without additional elaboration makes it ill-suited to serve as a benchmark.
As the branch responsible for overseeing the registration of religious representatives in BC, the Vital Statistics Agency maintains a policy that any organization seeking recognition "must demonstrate that it is sufficiently well established as to continuity of existence and as to recognized rites and usages respecting the solemnization of marriage to warrant the registration of its religious representatives" (Vital Statistic Agency 2016). Organizations demonstrate this by submitting "documentary proof of existence in British Columbia or existence and recognition in another province, for a minimum period of five years demonstrating such factors as stability, growth, and continuing viability" (Ibid.). This five-year requirement exists in spite of the Supreme Court of Canada ruling that "[t]he Charter protects all sincere religious beliefs and practices, old or new" (Ktunaxa Nation v. British Columbia 2017: para.69; and see also Sullivan 2009Sullivan : 1185. Again, we can see a strong bias in favour of a specific kind of religious structure. Given the paucity of legislative definitions of religion, we must instead look to existing jurisprudence to offer some insight. One of the first cases that explored the meaning of 'freedom of religion' as enshrined in the Charter of Rights and Freedoms was a 1985 Supreme Court of Canada case that considered whether the Lord's Day Act, which prohibited the sale of goods on a Sunday, infringed on the religious freedoms of those wishing to do business on this day, in this case a drug store in Calgary. Finding the purpose of the Act to amount to religious compulsion and therefore be unconstitutional, Justice Dickson (R. v. Big M Drug Mart Ltd.1985: para.135) wrote for the majority: In an earlier time, when people believed in the collective responsibility of the community toward some deity, the enforcement of religious conformity may have been a legitimate object of government, but since the Charter, it is no longer legitimate. With the Charter, it has become the right of every Canadian to work out for himself or herself what his or her religious obligations, if any, should be and it is not for the state to dictate otherwise.
In other words, the court specifically eschewed any role for government in deciding the validity of claimed religious beliefs. The Court found that the Act had as its purpose "the compulsion of sabbatical observance" and, as a result, it violated the Charter's protection of freedom of religion (Ibid., para.85). The language of the ruling is adamant, noting that "the diversity of belief and nonbelief, the diverse sociocultural backgrounds of Canadians make it constitutionally incompetent for the federal Parliament to provide legislative preference for any one religion at the expense of those of another religious persuasion" (Ibid., para.134).
This view was expanded upon by the Court in Syndicat Northcrest v. Amselem in 2004. In this case, several Orthodox Jews in Montreal sought to construct temporary huts on the balconies of their units, in violation of their building's bylaws, to fulfill what they claimed was a religious requirement. Writing for the majority, Justice Iacobucci (Syndicat Northcrest v. Amselem 2004: para.50) noted that: the State is in no position to be, nor should it become, the arbiter of religious dogma. Accordingly, courts should avoid judicially interpreting and thus determining, either explicitly or implicitly, the content of a subjective understanding of religious requirement, ' obligation', precept, ' commandment', custom or ritual. Secular judicial determinations of theological or religious disputes, or of contentious matters of religious doctrine, unjustifiably entangle the court in the affairs of religion.
Again, religion is viewed by the Court as an inherently subjective and therefore personal thing. Developing the definition, Justice Iacobucci (Ibid., para.39) elaborated, noting that: In order to define religious freedom, we must first ask ourselves what we mean by 'religion'. While it is perhaps not possible to define religion precisely, some outer definition is useful since only beliefs, convictions and practices rooted in religion, as opposed to those that are secular, socially based or conscientiously held, are protected by the guarantee of freedom of religion. Defined broadly, religion typically involves a particular and comprehensive system of faith and worship. Religion also tends to involve the belief in a divine, superhuman or controlling power. In essence, religion is about freely and deeply held personal convictions or beliefs connected to an individual's spiritual faith and integrally linked to one's self-definition and spiritual fulfilment, the practices of which allow individuals to foster a connection with the divine or with the subject or object of that spiritual faith.
While this statement defines religion in narrower parameters than leaving it to up to "every Canadian to work out," (R. v. Big M Drug Mart Ltd. 1985: para.135) it still connects religion with the personal experience. It also leaves a considerable degree of flexibility, noting that religion 'typically' and 'tends to' display certain characteristics without establishing anything definitive. Even with those characteristics, words like 'faith', 'worship' and 'spiritual' create further room for semantic disputes.
The challenge for bureaucrats tasked with vetting applicants for a religious benefit (whether these include receiving tax benefits or inclusion in a list of approved sample prayers) is that Justice Iacobucci was clear that Courts, and by extension agents of the state, are not to be 'the arbiter of religious dogma'; however, any vetting process inherently demands some criteria upon which to judge an application. These bureaucrats are often therefore placed in the unenviable position of determining if something is religious without making any judgement about the religion itself. But how can one determine whether something is or is not religious without making a judgement concerning its content? For example, the Church of Atheism was found not to be a religious organization based on the Minister's review of the content of the materials the Church submitted describing its belief system (Church of Atheism of Central Canada v. Canada 2019: para.23).
Justice Iacobucci attempted to address this apparent paradox with respect to freedom of religion claims. He noted that "while a court is not qualified to rule on the validity or veracity of any given religious practice or belief, or to choose among various interpretations of belief, it is qualified to inquire into the sincerity of a claimant's belief, where sincerity is in fact at issue" (Syndicat Northcrest v. Amselem 2004: para.51). He then provided the following test: at the first stage of a religious freedom analysis, an individual advancing an issue premised upon a freedom of religion claim must show the court that (1) he or she has a practice or belief, having a nexus with religion, which calls for a particular line of conduct, either by being objectively or subjectively obligatory or customary, or by, in general, subjectively engendering a personal connection with the divine or with the subject or object of an individual's spiritual faith, irrespective of whether a particular practice or belief is required by official religious dogma or is in conformity with the position of religious officials; and (2) he or she is sincere in his or her belief. Only then will freedom of religion be triggered (Ibid., para.56).
This has become known as the 'Amselem Test' and was expanded upon in S.L. v. Commission scolaire des Chênes. In this case, Catholic parents argued that a mandatory ethics and religion class in Quebec schools violated their freedom to raise their children according to their own religious beliefs. Writing for the majority, Justice Deshamps (S.L. v. Commission scolaire des Chênes 2012: para.24) elaborated on the aspects of the test for sincerity of belief, explaining that: It follows that when considering an infringement of freedom of religion, the question is not whether the person sincerely believes that a religious practice or belief has been infringed, but whether a religious practice or belief exists that has been infringed. The subjective part of the analysis is limited to establishing that there is a sincere belief that has a nexus with religion, including the belief in an obligation to conform to a religious practice. As with any other right or freedom protected by the Canadian Charter and the Quebec Charter, proving the infringement requires an objective analysis of the rules, events or acts that interfere with the exercise of the freedom. To decide otherwise would allow persons to conclude themselves that their rights had been infringed and thus to supplant the courts in this role.
Applying this to the facts of the case, Justice Deschamps concluded that while the parents "sincerely believe that they have an obligation to pass on the precepts of the Catholic religion to their children", their ability to observe that practice was not being interfered with (Ibid., para.26). Rather than effectively define what constitutes a religion, therefore, the Courts have sought to establish the sincerity of religious beliefs.
Of course, the job of interpreting what constitutes 'sincerity' and evaluating this is no small task. Evaluating the sincerity of someone's beliefs is incredibly challenging, particularly given the limitations imposed by the Amselem Test. As Charney (2010: 48) elaborates, because "the Supreme Court of Canada held that religious beliefs are personal and so cannot be subject to an objective evaluation," one cannot rely on factors such as consistent practice and expert opinion to evaluate the sincerity of an individual's beliefs (Ibid., 55). For example, because beliefs change over time (Syndicat Northcrest v. Amselem 2004: para.53 and 54), one cannot rely on an individual's actions in the past to serve as an indication as to the sincerity of their beliefs at the moment (Charney 2020: 59). As Beaman (2010: 277) elaborates, "courts have struggled to reconcile the rather static notion of religion with the dynamic ways in which people live out or practice religion on a day to day basis. Like law, religion on paper and religion as lived are often two very different phenomena." Or to put it another way, the fact that someone enjoyed a ham sandwich last year has no impact of the sincerity of their belief in the need to keep Kosher today. Ultimately, as Charney notes, "the state does not have a 'window into men's souls'" (2010: 56, citing Queen Elizabeth I), and must instead wait until that belief manifests itself in some external way -when it becomes embodied, practiced, and public (Beaman 2010: 278).
In exploring the extent to which prison regulations substantially burdened a prisoner's exercise of religion, Sullivan (1998) explored two different models of what constitutes religion and how one would test if the exercise thereof were substantially burdened. The first view considers religion as "doctrinally definite and authoritatively determined by an institutional church," and as such, "being religious -exercising religion -is being obedient to the legal prescriptions of that religion" (p.446). Under this formulation, a court would use expert testimony to identify "the mandates of a particular religion… to establish the requirements of that religion with respect to a particular practice," and a religious practitioner would only need to "testify convincingly to membership in a religious community" (Ibid.). The second view sees religion "as personally determined -a matter of individual choice. Being religious -exercising religion -is about being faithful to one's own religious understanding" (Ibid.). Under this formulation, "the appropriate test should be a subjective one -whether the practices are central to, and sincerely and religiously motivated in terms of, the individual's religious exercise" (Ibid.).
The aforementioned rulings provide some guidance for those bureaucrats tasked with evaluating the claims of individuals and groups seeking recognition for a benefit awarded on the basis of religion, and seem to suggest the second of Sullivan's two tests. 3 Bureaucrats should query the sincerity of belief by those who seek recognition on an individual case-by-case basis. As religion has been considered by the courts as a deeply personal issue, there is no need to consider religious texts or other authorities.
As we shall see, however, this approach contrasts with what is being undertaken in the Office of the Clerk's present review of the sample prayers. Here, the Office of the Clerk aims to identify a manageable number of religious traditions from which to provide prayers. Rather than treating religion as subjective and therefore evaluating the sincerity of various beliefs, bureaucrats are instead treating it as "doctrinally definite and authoritatively determined by an institutional church" (Ibid.), and making determinations based on religious authorities and legal prescriptions of those religions. The very first step necessary to select sample prayers for legislators -simply defining what constitutes a religion -finds itself on shaky constitutional ground.

Which religions make the cut?
Once the difficult constitutional hurdle of how to determine what constitutes a religion has been cleared, there remains the challenge of selecting which recognized religious traditions will be included on the list of sample prayers, and by extension, which will be excluded.
Making this determination, however, risks violating the state's duty of religious neutrality, as outlined in the 2015 Supreme Court of Canada ruling Mouvement laïque québécois v Saguenay. This case examined the constitutionality of starting municipal council meetings with a prayer (Phelps Bondaroff et al. 2019: 16). The ruling found that prayers at municipal council meetings violated the state's duty of religious neutrality, with Justice Gascon (Mouvement laïque québécois v. Saguenay 2015: para.72) noting that: the evolution of Canadian society has given rise to a concept of neutrality according to which the state must not interfere in religion and beliefs. The state must instead remain neutral in this regard. This neutrality requires that the state neither favour nor hinder any particular belief, and the same holds true for non-belief. It requires that the state abstain from taking any position and thus avoid adhering to a particular belief.
Justice Gascon underscored that the state's duty of religious neutrality was a ' democratic imperative', stipulating that [t]he state may not act in such a way as to create a preferential public space that favours certain religious groups and is hostile to others. It follows that the state may not, by expressing its own religious preference, promote the participation of believers to the exclusion of non-believers or vice versa (Ibid., para.75).
Anytime the Office of the Clerk decides to include a specific religious tradition on its list of 'sample prayers' at the exclusion of another, it violates its duty of religious neutrality. Any time it includes religious traditions at the exclusion of non-religious traditions, the Office of the Clerk likewise violates this duty.
Beyond these constitutional quagmires, the staff in the Office of the Clerk are faced with the very pragmatic question about how to go about selecting which religious traditions to include. Here we will examine several possible approaches, although, as we shall see, each is fraught with uncertainty.

Relying on Bureaucratic Familiarity
The (then acting) Clerk informed the authors by email that her staff's review of the sample prayers intended "to ensure that prepared prayers provide a breadth of nonreligious reflections, as well as prayers from major religious groups" (K. Ryan-Lloyd, 2019, Acting Clerk of the House, correspondence with authors, August 21). In further communications, the Clerk pointed to the inclusion of "representatives of faith communities" in the Government's Table of Precedence, though did not specify that this list was being used to inform the process (K. Ryan-Lloyd, 2020, Clerk of the House, correspondence with authors, March 10; and see Government of BC n.d.; Government of Canada 2015). Taken together, this suggests that the approach the Office of the Clerk is taking is effectively to select the traditions most familiar to them and to other branches of government. The problem with such an approach should be immediately apparent, as it relies almost entirely on the biases and anecdotal familiarity of those within the bureaucracy.
A public version of the province's Table of Precedence is not currently available; however, the official federal Table  of Precedence lists "Representatives of faith communities" after "Premiers of the Territories of Canada" and above judges from a wide range of courts, senators, members of parliament, and other high ranking officials (Government of Canada 2015). This list, an amended version of which applies at provincial events, includes the following footnote with respect to representatives of faith communities: The religious dignitaries will be senior Canadian representatives of faith communities having a significant presence in a relevant jurisdiction. The relative precedence of the representatives of faith communities is to be governed by the date of their assumption in their present office, their representatives being given the same relative precedence (Ibid.).
In a conversation with a spokesperson from the Department of Canadian Heritage, we were told there is no formal list of religious officials associated with the Table (M. Belanger, 2020, Senior Ceremonial and Protocol Officer, Canada, conversation with authors, March 13). Rather the Table instead serves as a guide for those hosting official government events to identify the symbolic hierarchy of those present. As these events are typically invite-only, organizers are aware of who among the attendees are religious representatives. However, the spokesperson did not describe a robust procedure for ensuring that a diversity of religious (and non-religious) traditions would be represented. Organizers might well rely, for example, on a contact list held by the Chaplain General of the Canadian Armed Forces. Event organizers would also research and vet individual representatives to exclude those not deemed worthy of attending; the Church of the Flying Spaghetti Monster was given as an example.
Of course, none of this addresses the central issues raised above. By relying on other parts of the bureaucracy, the Office of the Clerk is merely abdicating responsibility for making the judgement concerning what is and is not a religion and allowing others to make this contentious call. It is clearly evident that with the approach taken by organizers of federal government events, these decisions are often still ad hoc and subject to the same biases and preconceptions. Even the list compiled by the Chaplain General will invariably be biased toward larger, hierarchal religious traditions that have pastoral programs.
Alternatively, the Office of the Clerk could connect with those traditions that have been actively involved -or permitted to be involved -with larger multi-faith organizations, such as the Canadian Multifaith Federation (formerly the Ontario Multifaith Council) (n.d.), the Faith Alliance (Faith in Canada 150 n.d.), or the Multifaith Action Society (n.d.). Among the deficiencies with this approach are that each of these organizations relies on groups that represent individual faith traditions to apply for membership and they often do not include non-religious traditions. This risks privileging proselytizing traditions and ones that take a more proactive role in wider society. Some religious traditions, such as Jehovah's Witnesses, actively limit their interaction with those of other faiths and of no faith, and abstain from engaging in 'politics' (Jehovah's Witnesses n.d.).
Ultimately, there does not appear to be any official government list of religious traditions that can be used to decide which should be included on the list of sample prayers upon which the Office of the Clerk can draw. To withstand judicial scrutiny this process cannot be arbitrary, so another method must be adopted.

Population Demographics
Arguably, the next most intuitive way to identify the traditions to be included would be to use census data in order to identify the most common religious and belief traditions with which people in BC identify. Under this approach, the Office of the Clerk would simply seek to include prayers and reflections from the religious and belief traditions with the most adherents in the province.
There are a number of challenges with this method however. First, there is the issue of how Statistics Canada constructs the question regarding religion in the census. Specifically, the census directs respondents to record "the specific denomination or religion with which they identify." The problem of this approach is that this provides a measure of religious affiliation rather than of active practice (see Beaman 2010: 270). Statistics Canada further instructs respondents to identify their religion "even if they are not currently a practicing member of that group" (Statistics Canada 2015: Question 22, Religion). As such, the precise rates of belief may not be accurately captured by the census, as those who may have left their religious tradition and no longer ascribe to its tenets are told to report their (former) religious identity (Jukier & Woehrling 2015: 156-157). This is not just a theoretical concern as 75% of Quebecers identified as Catholic in the most recent Statistics Canada survey (Statistic Canada 2013), but under 28% say they strongly believe in God, according to a 2019 poll by Léger Marketing (Riga 2019). This clearly suggests that a large number of Quebecers are ' cultural Catholics' and simply identify as Catholic in the census, while not ascribing to the most basic tenet of the religion, namely belief in God (The Vatican n.d.).
Next, by relying on the categories used by Statistics Canada, the Office of the Clerk is continuing to use decisions made by other government bureaucrats. This presents essentially the same problems as reliance upon the Table of Precedence, as it presumes that Statistics Canada knows what a religion is, and is able to successfully navigate all the previously raised issues. This is not merely a theoretical concern. In fact, several respondents to the agency's consultation for Census 2021 cited data gaps in the list of religious identities provided for respondents in their feedback (Statistics Canada 2019). In response to these comments, Statistics Canada has modified its list of examples of religions for Census 2021 (Ibid.). This raises a further challenge for bureaucrats attempting to determine which religions to represent within a list of sample prayers, in so far as the categories may change from census to census.
Assuming the demographic data are accurate, bureaucrats will need to set a threshold that must be attained in order for a religious or belief tradition to be allocated a sample prayer, otherwise the list would quickly become unwieldy, as there were 99 religious identities listed in the most recent census (Statistics Canada 2013). Ultimately, any threshold that the bureaucracy picks will be arbitrary.
A major problem with this approach is that the manner in which religious traditions and identities are sub-divided in the census can make determinations as to the relative size of those groups difficult. In his analysis of the 1973-1996 General Social Survey, for example, Sherkat found that "almost half the ' other' category, which were previously thought to be non-Judeo Christian faiths, [were] actually misclassified Christians" (Sherkat 1999: 551 cited by Beaman 2003. The census includes roughly 70 sects of 'Protestant Christianity,' but the distinction between these is often relatively haphazard, and even this number does not seem to accurately capture the diversity of Protestant sects. By one estimate, there are roughly 33,000 Protestant denominations (Barrett et al. 2001). 4 On the other hand, the census provides no distinction between interpretations of Buddhism, Islam or Judaism, despite the deep divisions present in each of these traditions.
Judaism, for example, is often categorized into Reform, Conservative, and Orthodox branches and there exist sub-categories within each of these. For example, Orthodox Judaism can be refined into Modern Orthodox, Open Orthodox, Haredi, and Ultra Orthodox (which includes Hasidic and Yeshivish) (My Jewish Learning n.d.). The differences between a Hasidic Jew and a Reform Jew are vast, likely more extensive than the differences between an Anglican and a Catholic. It seems arbitrary therefore, that Christianity is subdivided into a wide range of denominations, whereas divisions within other religious traditions are ignored.
What level of granularity is acceptable for the purposes of including a religious tradition on the list of sample prayers? A risk here is that religions with more sects and denominations may be over-represented by 'sample prayers' or individual sects risk being under-represented when subsumed into a larger category.
Let us consider one possible threshold. Given Judaism's historical significance and the fact that it was the only religion, other than Christianity, that was found to be over-represented in recent legislature prayers (Phelps Bondaroff et al. 2019: 70), establishing a threshold that includes Judaism would be consistent with current practice. In this instance, religions that can claim followings with at least as large a portion of the province could present a strong a case to have one of their prayers included on the sample list.
In the 2011 National Household Survey (Table 2), 23,130 people reported their religious identity as Jewish, representing 0.5% of the province (Statistics Canada 2011). Of the 99 religious identities listed on the survey, 17 constitute at least 0.5% of the population of BC. Including 17 separate prayers represents a more than tripling of the size of the current list; assuming only one prayer is included for each tradition and that larger traditions are not apportioned a number of sample prayers commensurate with their demographic representation.
A possible solution to this unwieldy list would be for bureaucrats to amalgamate the prayers of the 10 sects of Protestant Christianity present on that list. While this may seem like a reasonable course of action at first blush, it raises further questions regarding how such a decision might be made, given that many of these sects formed because of sometimes nuanced and subtle differences in interpretations of the same texts. Would members of an Anglican church be comfortable being included in the same category as members of a Pentecostal church? Would members of both of these groups feel adequately represented by the same prayer? And how are bureaucrats to make these determinations, given that they are likely untrained in the subtle theological differences between these traditions?

Demographics of the Legislature
Rather than relying on the demographics of the province, the Office of the Clerk could choose to draft sample prayers reflecting the diversity of the Legislative Assembly. This might prove more useful for MLAs, as they would most likely choose from those prayers that reflect their belief system. However, this would mean that the Office of the Clerk would have to query individual MLAs about their religious affiliation, identity, and belief, and the practices stemming from them. This raises questions about member's privacy rights, as well as challenges with operationalizing these beliefs into practice.
The Office of the Clerk would also need to continually update its religious data on members following every election and by-election, in the event of someone with a new religious identity being elected. Similarly, some sort of reporting mechanism would need to be in place to allow MLAs to update the Office of the Clerk on any changes in their beliefs, because "[o]ver the course of a lifetime, individuals change and so can their beliefs. Religious beliefs, by their very natures, are fluid and rarely static" (Syndicat Northcrest v. Amselem 2004: para.53;and see Charney 2010: 55).
Knowing the religious affiliation of each MLA would also not tell the Office of the Clerk how each individual MLA practiced that religion. As Sullivan notes, "most people have a complex and changing relationship to one or more religious traditions… They simply do not fit neatly into a model which could be labeled 'Presbyterian' or 'Buddhist' or 'Black Muslim'" (Sullivan 1998: 454). Likewise, simply knowing an MLA's religious affiliation tells the Office of the Clerk nothing concerning that MLA's beliefs or how these beliefs manifest themselves in practice. Knowing an MLA's religious affiliation does not, for example, inform the Office of the Clerk where that MLA stands on theological differences that might exist within that religion (see  Sullivan 1998). This raises all manner of thorny questions concerning belief, practice, and even sincerity of belief, questions that the Office of the Clerk is not capable of resolving. Ultimately, this approach seems far more cumbersome than simply leaving the entire endeavour to the MLAs themselves, as they are already empowered to draft their own prayers, and certainly more capable of accurately representing and expressing their beliefs than another actor. Bureaucrats are on shaky constitutional grounds when it comes to making decisions regarding what is and what is not a religion and further lack a practical rational method for deciding which traditions to include.

Which prayers to use?
Presuming the Office of the Clerk is able to identify those traditions to represent on its list of sample prayers and reflections, bureaucrats would then be faced with the even more onerous task of identifying specific prayers (and other content) to reflect each of those traditions.

The bureaucracy is ill-suited to select specific prayers
For a particular tradition, there may be a considerable number of commonly uttered prayers, each with a specific intent or purpose. The Anglican Church of Canada lists a number of resources for prayers, including 'The Apostles' Creed,' 'The Athanasian Creed,' 'The Nicene Creed,' as well as daily Bible readings and a year-long calendar of prayers (The Anglican Church of Canada n.d.). Each of these prayers has a specific context and, while some may not be suitable to the average Anglican for use in starting the Legislature's proceedings, there remains the likelihood that more than one would be appropriate.
The directive that the state not become the arbiter of religious dogma bears repeating (Syndicat Northcrest v. Amselem 2004: para.50). Clearly bureaucrats should not be put, nor put themselves, in the position to decide what constitutes an acceptable prayer for a specific religious tradition. Instead, they should be looking for ways to extricate themselves from, or avoid, any situation where such determinations might arise. After all, what grounds could the state have for selecting between two competing options without attempting to arbitrate religious dogma?
Aside from simply selecting prayers from well-established and hierarchal traditions, how can bureaucrats select sample prayers from non-hierarchal religions or from ones without established dogmas?

Practical limitations on soliciting religious prayers from religious representatives
A seemingly reasonable alternative approach might be to simply canvas the largest organizations representing those selected belief systems. While finding representative prayers for highly structured religious groups such as the Roman Catholic Church should be straightforward -bureaucrats could simply call up the local Archbishop and ask him 5 to submit a prayer -this is not as straightforward for most traditions. Or as Sullivan (1998: 452) explains, one advantage of a religious establishment is that you can ask the authorities. Without a religious establishment there is no way of making these determinations. All we have is a set of divergent opinions about religion and religious practice and no way of adjudicating among them.
Deciding who to contact within a religious tradition, such as the highly decentralized charismatic Christian tradition or the various courts and dynasties of Hasidism is an impossible task. Any effort to solicit prayers from various faith traditions will tend to favour religious traditions with established hierarchies and structures that are approachable. Less hierarchal religious traditions may find themselves being overlooked for lack of an official spokesperson.
Most religions are at least somewhat decentralized and often have competing interpretations of the same texts. This is, as previously noted, the reason why many sects and denominations emerge in the first place. Minor differences in beliefs, dogma, language, and practices have fuelled religious conflict for millennia. Ultimately, there are numerous reasons why sects differentiate themselves from coreligionists. As explored above, combining two smaller but distinct sects introduces the possibility for disagreements over the content of a representative prayer. While Anglicans and Lutherans might find the 'Lord's Prayer' acceptable, it is unlikely that Hasidic and Reform Jews would agree on a single acceptable and representative Jewish prayer.
Even within a particular church, temple, or synagogue, there is likely to be a diversity of opinion as to which prayers individuals might consider representative. Take for example the recent splits that have emerged within the Methodist Church (Burns 2020) and within the Canadian Anglican Church over same sex marriage (Valpy 2007), or the emerging schism within Orthodox Christianity around the emergence of an independent church in the Ukraine (The Economist 2020). As the Supreme Court of Canada has said, religion is "about freely and deeply held personal convictions or beliefs" (Syndicat Northcrest v. Amselem 2004: para.39), and while two members of the same faith tradition may, for example, share the same book, it is unlikely that they share identical interpretations of that book. As such, we might reasonably expect there are as many religious viewpoints as there are British Columbians. It is seemingly counterproductive then to attempt to select a representative prayer without the input of the congregants themselves. As such, any approach that solicits feedback from the largest religious organizations will privilege the viewpoints of their spokespeople over the actual congregants and worshippers whose beliefs the bureaucrat is trying to reflect.
There are further complications relating to the nature of the spokesperson. First, there is the matter of whether a bureaucrat can verify that the person or people from whom they are soliciting prayers is a legitimate representative of that tradition. And in so doing, do they risk weighing in on matters of dogma and the internal affairs of that group?
Another issue relating to the identity of spokespeople and representatives of faith traditions is that they tend to be disproportionately men. Evidence of this is reflected in the practice adopted by the Scottish Parliament. On Tuesday afternoons, the first item of business is 'Time for Reflection.' Guest speakers are invited to deliver reflections to the legislature for a period of four minutes and the "pattern of speakers reflects the balance of beliefs in Scotland (based on the census)" (Scottish Parliament 2020). While this practise does result in a wide range of guests representing the diversity of beliefs in Scotland (religious and non-religious), an examination of the guests reveals a serious shortcoming: Between 2016 and 2020, over 70% of the guests delivering addresses were men (Ibid).
This raises a question of gender equity: Will men, who have traditionally held positions of power in most religions, identify the same prayers and rituals as women or gender non-binary people of the same religious tradition? Similar questions could be asked concerning other identities such as race, class and ability. The question is then whether or not the Office of the Clerk should make an effort to seek diverse viewpoints within any given faith tradition. At the same time, we should also ask whether this is an appropriate role for bureaucrats.

The challenges of reflecting non-religious diversity
Selecting a prayer that is representative of a specific religious tradition is an incredibly challenging and constitutionally dubious task. Choosing a reflection to deliver on behalf of the non-religious, on the other hand, is distinctly more challenging.
Consider the category of people who identify themselves as having 'no religion.' The diversity of beliefs encompassed in this category is vast. A 2016 Insights West survey commissioned by the BC Humanist Association (BCHA) found that 69% of British Columbians claimed to not "practice or participate in a particular religion or faith" (BCHA 2016). Of those, 44% believed in a higher power. Some people within this category may believe in a god, gods, higher power, or force, but simply not fit within the confines of a specific religion; others may have esoteric spiritual beliefs; while yet others may be staunch atheists. Clearly selecting one or more reflections to represent this diverse group is a Herculean task.
Even just among those who identify as 'non-believers' or ' atheists' there exists a considerable and diverse group of people. Non-believers and atheists are united in one thing: the lack of a belief in a god or gods. Knowing that someone is an atheist tells you absolutely nothing about that person's position on any particular moral question. How does one deliver a reflection that captures a lack of belief, or the diversity of beliefs of the non-religious? Any individual poem, reading, or reflection will, by its very nature, speak more to some people than to others. And often the more broadly unobjectionable the piece, the less meaningful it becomes.
Furthermore, it is unclear which organization or organizations can claim to speak for the entire body of nonreligious British Columbians. The issue of identifying a legitimate spokesperson is magnified, given a general lack of proliferation of organizations and groups representing this demographic. Coalescing around a lack of belief in something is not a common activity. Most people go about their lives not believing the things they do not believe. Most people do not believe in unicorns and do not see the need to join an organization dedicated to asserting this lack of belief. As such, the vast majority of non-believers are not, and likely would not want to be, represented by some group or organization. Who then should the Office of the Clerk approach to solicit reflections from the mass of non-believing British Columbians?
The BC Humanist Association (BCHA), for example, represents a subset of atheists and agnostics who hold Humanist values but not all atheists or non-religious people share these values. Some atheistic philosophies, like nihilism, are incompatible with Humanism, and some atheists simply do not ascribe to Humanist ethics.
One thing that comes close to uniting most non-religious individuals is a rejection of formalized ritual practice. As such, even asking the BCHA for sample reflections to include on a list of sample prayer is requesting an organization participate in an activity in which they do not believe nor support. As Phelps Bondaroff et al. (2019: 86)

argued in House of Prayers:
Humanism 'is undogmatic, imposing no creed upon its adherents,' and as such, there is no one specific 'prayer' upon which humanists will universally agree. Rather, when asked, individuals often craft a humanist ' declaration' based on their personal values and connections to the philosophy.
This becomes a fundamental structural challenge with seeking to engage individuals in a practice in which they do not believe.

Soliciting contributions directly from British Columbians
Rather than attempting to draw prayers and reflections from religious and belief groups, the Office of the Clerk could decide to solicit feedback directly from British Columbians; asking individuals to submit sample prayers for inclusion.
This would help circumvent some of the issues regarding legitimizing organizations, beliefs and spokespeople as explored above. However, this approach presents a number of drawbacks. Relying on submissions comes with an inherent bias problem. Highly motivated individuals and organized groups are likely to drown out voices from traditionally underrepresented groups. For example, a consultation by the Ontario Legislative Assembly on the practice of beginning its sessions with the 'Lord's Prayer' generated 11,000 responses and "[e]ffectively, the conversation was taken over by those who wanted to keep the Lord's Prayer and would see to it that it remained an important part of the daily opening of the Legislature" (Fizet 2010). Furthermore, as detailed by Phelps Bondaroff et al. (2019: 25-26), not all religious traditions include prayer or support the practice of offering prayers or reflections in a public setting. As a result, people from these faith traditions will necessarily be excluded from this process.
There would likely be a disproportionate number of submissions from religious British Columbians in so far as the idea of a prayer (or even reflection) is not generally present in secular worldviews. Again, there is a structural flaw inherent in asking an extended group of people containing both believers and non-believers to submit prayers: those who come from a religious tradition that utilizes prayers will be more willing, comfortable, and able to submit a prayer. Those whose traditions do not include prayer or who are non-religious are more likely to decline the invitation rather than trying to contort their beliefs to fit this structure.
There are further issues that may be irresolvable, expressly the issue of prayers and reflections that may be foundational to one group, but offensive to another. Many prayers assert the existence of one god at the exclusion of others; believers in those other gods may find such prayers exclusory. The reflections of many non-believers may come in direct conflict with the beliefs of people from faith traditions. Take for example the sample 'Humanist Declarations' that were included in Phelps Bondaroff et al. (2019: 106), like the following: "There are almost certainly no gods; therefore, let us commit ourselves to tackling the challenges that face our province with reason, wisdom, and empathy" (see Appendix 2). This kind of declaration is foundational to Humanist thought, but might be considered ' offensive' to religious people. In fact, a 2009 ad campaign by the British Humanist Association put ads with the following wording on the side of London buses: "There's probably no god, now stop worrying and enjoy your life" (Sweney 2009a). This sparked a controversy as a number of religious groups filed complaints with the Advertising Standards Authority who eventually dismissed them (Sweney 2009b;Sweney 2009c;Beckford 2009).
Is the Office of the Clerk equipped and capable of performing the delicate and potentially impossible balancing act of accommodating the vast and sometimes directly conflicting viewpoints which might be submitted as potential prayers or reflections? Controversies that are likely to arise will distract from the general operations of the Office of the Clerk and likely tend to decrease the dignity of the Legislature.
The Office of the Clerk in all likelihood is not equipped to review the submissions and adjudicate between them, both from a practical and legal perspective. Evaluating hundreds or thousands of submissions would be a monumental undertaking, one for which the Office of the Clerk is likely ill-equipped. Assuming it does have the capacity to evaluate submissions, doing so would surely impede other projects planned by the Office of the Clerk. Apart from criteria such as formatting, spelling, and length, any effort on the part of bureaucrats to decide between two prayers would further entangle the state in arbitrating religious dogma, again violating the clear directive from the Supreme Court of Canada.

Conclusion
Bureaucrats are not in the position, constitutionally or practically, to evaluate what constitutes a valid religious belief, who can speak for those beliefs, nor what prayers and reflections represent those belief systems. The provision of 'sample prayers,' which are used in half of all prayers delivered by MLAs in the BC Legislature, fails to accurately represent the diversity of BC. Efforts to update the list unjustifiably entangle the bureaucracy into arbitrating religious dogma and inevitably embroil the Office of the Clerk, and by extension the Legislature, in sectarian disputes. In this instance, efforts at incremental reform are more problematic than wholesale change. Rather than struggling with the constitutionally dubious task of drafting a list of sample prayers, the Office of the Clerk is better served by leaving the crafting of prayers and reflections to MLAs themselves. Alternatively, the practice of beginning sessions of the BC Legislature with prayers and reflections could be ended.
The challenges faced by the Office of the Clerk in BC are not unique to this province. Prayer is common in legislatures across the country: New Brunswick, Prince Edward Island and Ontario open their daily sittings with the 'Lord's Prayer' (Lanouette 2009). Since 2008, Ontario has followed the Lord's Prayer with a prayer from a rotating schedule of prayers "reflecting Indigenous, Buddhist, Muslim, Jewish, Baha'i and Sikh faiths" (Bueckert et al. 2017: 25; and see Lanouette 2009). Sittings of the Nova Scotia House of Assembly open with a shortened version of the 'Lord's Prayer' (Fizet 2010: 2). In Alberta, the Speaker devises and delivers the daily prayer, while in Saskatchewan and Manitoba the Speaker delivers a standard 'non-denominational' prayer, and in the Yukon, the Speaker reads one of four standard prayers (Bueckert et al. 2017: 25; N. Clarke, 2019, Speaker Yukon Legislative Assembly, correspondence with authors, September 3; Yukon Legislative Assembly 2018). The National Assembly of Quebec begins with silent 'time for reflection,' a practice that was adopted following the abolition of the practice of opening sittings with prayer in 1976 (Bueckert et al. 2017: 25;Lanouette 2009: 6). Newfoundland and Labrador has never opened sittings of its House of Assembly with prayer (Ibid.). And finally, Nunavut and the Northwest Territories join BC in giving MLAs the opportunity to deliver a prayer of their own devising (Fizet 2010: 2; Legislative Assembly of the Northwest Territories 2015: 12).
Both the Canadian House of Commons and Senate begin with the Speaker reading a standard 'non-denominational' prayer, followed by time for silent reflection (Fizet 2010: 2; Parliament of Canada n.d.; Canadian Senate 2013: Canadian Senate 2015: Chapter 4). The practice of opening sittings with prayer is a common feature in legislatures around the world. Any legislature that seeks to amend or reform its practices relating to prayer in order to make them more inclusive will encounter many of the problems outlined in this paper, and as such, may wish to discontinue the practice rather than put the state in the inappropriate position of having adjudicate on matters relating to dogma, thereby violating its duty of religious neutrality. 4. Rather than bowing our heads and closing our eyes in deference, we should open our eyes to face the challenges that confront us. Let us commit ourselves to improving the lives of all British Columbians with reason, wisdom, and empathy. 5. We have within us all a shared humanity. Let us therefore treat one another with respect and dignity. Let us focus on what we have in common, and not what divides us. And let us commit ourselves to applying reason and science, strengthened by empathy and compassion in order to improve the lives of all British Columbians. 6. Let us celebrate our shared humanity, our shared capacity for reason and compassion, our shared love for the people of our Province. Let us commit ourselves to fulfilling the great responsibility we have been given by the people of British Columbia, with reason informed by science, compassion, and empathy. ). 2 Here the term 'non-sectarian' refers to a prayer that "invokes the divine or transcendent, a deity, power, or supernatural entity, or relies on religious language," but which could not be identified with a specific religion (Phelps Bondaroff et al. 2019: 54). Secular "includes any invocation, or call of thanks not specifically invoking, or directed towards a deity or the transcendent," and while these types of prayers may still end in ' amen' they don't otherwise rely on overtly religious language (Ibid. 53). 3 Whether the state's duty of religious neutrality should arguably preclude the government from providing benefits on the basis of religion is something that is explored at length in Phelps Bondaroff et al. 2019