IMPERFECT EVIDENCE AND UNCERTAIN JUSTICE: AN EXPLORATORY STUDY OF ACCESS TO JUSTICE ISSUES IN CANADA'S ASYLUM SYSTEM.

INTRODUCTION

Canada's refugee determination system is multi-faceted and complex. So too are the individual characteristics and collective circumstances of the asylum claimants who seek the protection of this system. While not all of these individuals are able to successfully demonstrate that they are legal refugees entitled to protection, the system itself must aim to guarantee access to justice for each of them, regardless of the underlying and compounding complexities of their cases. In this article, we identify and cluster a series of issues that arise for asylum seekers in relation to the acquisition, presentation, and assessment of the evidence that is needed to support their claim for refugee status in Canada. Ultimately, we conclude that the ability of claimants to access justice is rendered uncertain by shortcomings in the way that the refugee determination system engages with imperfect evidence.

Our observations are based on close examination of 40 refugee claims submitted within the Canadian asylum system. A consent-based information sharing agreement with Canada's Immigration and Refugee Board (IRB) provided researchers with complete case files for each of these claims. A series of custom-designed tools were then created to assess the files from an access to justice perspective, with a specific focus on the acquisition and presentation of evidence. This paper offers an overview of the cases and key findings, and notes two overall observations: first, that the system appears to respond unevenly to issues relating to both the sufficiency and consistency of key evidence, particularly in relation to how it acknowledges and accommodates the impact of the social context of claimants; and second, that certain features of Canada's refugee claims system have the potential to create or exacerbate evidence-related issues. In combination, these circumstances raise concerns about whether the system is providing adequate access to justice in all cases. At a time when Canada's asylum system is under both increased pressure (1) and increased scrutiny, (2) both of these exploratory findings warrant more comprehensive study.

The research presented in this article emanates from the University of Ottawa Refugee Assistance Project (UORAP), an initiative founded in 2010 in response to pending changes to Canada's refugee claims system. Draft bills and statements by the Government of Canada about the purpose of the new legislation indicated to many experts that the modified system would very likely cause serious access to justice issues for vulnerable claimants. (3) The UORAP was created both to study the modified system from an access to justice perspective, and to mitigate the anticipated deficits via direct programming. (4)

Canada's reformed refugee system came into effect in 2012. (5) As anticipated, a number of the new features raised access to justice concerns--a phenomenon we explore in detail in "Troubling Signs", an article devoted to comparing the predicted outcomes with what actually materialized when the modified claims system was introduced. (6) The research that underlies the current article has a different objective: rather than compare anticipated and actual access to justice deficits, researchers undertook a detailed study of 40 cases from within the modified system, with the goal of identifying areas of concern and considering their systemic dimensions.

This paper reports on that research and proceeds over four parts. Section one outlines the key features of Canada's refugee claims process and explains the conceptualization of access to justice that frames our analysis of the system. Section two summarizes our methodology and profiles the 40 cases that are at the heart of this work. Section three builds on the first two sections by clustering and exploring--from an access to justice perspective--three specific evidence issues that appeared across the UORAP cases. Finally, the last section describes four specific features of Canada's asylum system that have the potential to create or exacerbate evidence-related access to justice deficits. We conclude by noting the need for further study in this area.

I. REFUGEE ACCESS TO JUSTICE: CLAIMS SYSTEM AND SOCIAL CONTEXT

Both Canada's asylum system and the personal characteristics and circumstances of refugee claimants are multi-faceted and complex. In this section, we provide a brief introduction to each of these considerations, with the goal of informing the analysis of evidence issues that follows.

A. KEY FEATURES OF CANADA'S REFUGEE CLAIMS SYSTEM

A refugee claimant seeking protection in Canada must prove her identity and establish that she meets the legal definition of a refugee. (7) Claims are submitted to the IRB's Refugee Protection Division (RPD), (8) a body that is comprised of independent board members who are empowered to grant or deny refugee status. In order to determine whether protection is legally warranted, board members make a series of decisions about both the process that will be followed and whether all elements of the claim have been successfully demonstrated. (9) For example, members routinely consider whether to allow late submission of evidence, whether to postpone hearings, and whether to grant the special allowances associated with a successful vulnerable person application. They must also assess the substantive merits of the claim itself, a determination which very frequently hinges on key findings regarding the claimant's credibility. (10)

Central to the Canadian refugee system is the guarantee that every asylum seeker has the right to present her claim for protection at an oral hearing. (11) While the RPD is bound by certain legislative requirements, hearings are not conducted in accordance with the usual legal or technical rules of evidence, and the process is thus imbued with a significant degree of flexibility. Further, the guidance and principles provided to decision makers regarding both procedural and substantive decision making are generally non-binding and not public (except where they also exist in some independent legal authority, such as relevant legislation or jurisprudence), (12) placing even greater significance on the broad power and discretion given to individual board members. IRB-produced National Documentation Packages (NDP) provide board members with information about conditions in specific countries of origin and are publicly available. (13)

Most asylum seekers choose to prepare for, and appear at, their oral hearing with the support of a lawyer or immigration consultant (14) who assists with documenting and presenting the claim for protection. (15) Access to counsel has been established as a major factor affecting outcomes in refugee determinations, and those who are not represented are much more likely to have their cases withdrawn or declared abandoned. (16) The primary evidence at most refugee hearings is the claimant's own oral testimony, and language barriers frequently necessitate the use of a simultaneous interpreter. The IRB provides these services via accredited interpreters who are either present in person or available via phone. (17) Interpreters are under a legal obligation to interpret exactly what is being said by the claimant, (18) and IRB training materials instruct them to adjust their tone and level of language to match the claimant's style. (19) The claimant must also submit a written version of her story in advance of the hearing on a special "Basis of Claim Form" (BOC), and may choose to submit additional corroborating evidence, including witness testimony or supporting documents. (20) Claimants are responsible for absorbing the cost of translating all written materials into either French or English. (21)

While Canada's system is generally designed to be non-adversarial, the ministers of immigration and public safety have the right to intervene in cases where there are concerns about fraud, credibility, system integrity, or certain forms of criminality which may affect the claim. (22) These interventions can occur either at the request of the board member (23) or at the discretion of government officials who may determine that a particular aspect of the application triggers the need for further review. Procedurally, the minister must issue notice to the claimant of an intention to intervene and provide the RPD with both a copy of this notice and a written statement indicating how and when the claimant was advised. (24) The notice itself must specify the reason for the intervention, and indicate whether it will occur in person via ministerial submissions at the hearing, in writing, or both. (25)

While protection must ultimately be provided to any asylum seeker who meets the legal requirements, Canada's status determination system differentiates between certain groups of claimants with regards to the claims process itself, including in relation to applicable timelines. For example, the system stipulates via legislation that claimants from certain countries (those termed "designated countries of origin" (DCOs)) must have their hearings heard within 30 days of filing a claim from within Canada or 45 days from filing a claim at a port of entry. (26) Further, if the claim originates at a port of entry, a completed BOC must be submitted within 15 days of receiving the form. (27) Claimants from all other countries must also file their BOC within 15 days of making a claim at a port of entry but their hearings can occur within 60 days, rather than 45. (28) All claimants must submit any supporting documentary evidence no more than 10 days before the hearing date. (29)

Asylum seekers who are successful in their claim for protection before the RPD receive permanent resident status in Canada. Those who are unsuccessful generally have the option of appealing to the Refugee Appeal Division (RAD) for a full fact-based appeal,although access to the RAD is restricted for certain categories of claimants. (30) Individuals who are unsuccessful before the RAD--or who are statutorily barred from appealing to it in the first place--may apply to the Federal Court within 15 days of the negative decision for leave for judicial review. In 2017, 1926 proceedings were commenced at the Federal Court and 384 applications for leave were granted. (31)

Like any complex process, Canada's refugee determination system is comprised of hundreds of procedural and substantive features which collectively create the system described here. For individual claimants, however, each specific feature has the potential to play a significant role in their ability to access justice. While our study was not designed to comprehensively assess the impact of every system component, our work did reveal four particular features whose ability to create or exacerbate evidentiary concerns was apparent across multiple files. These four features are ministerial interventions, legislated timelines, interpretation, and board member discretion. (32) Each of these features is discussed in detail in Part IV of this paper.

B. ACCESS TO JUSTICE AND SOCIAL CONTEXT IN REFUGEE DETERMINATION

The UORAP's research is grounded in what we refer to as a social context conception of access to justice. As is explained in our earlier articles, (33) this framing adopts Roderick Macdonald's three dimensions of access to justice--procedural, substantive, and symbolic--and reformulates his analysis into a focus on the ways in which individual and collective circumstances impact justice seekers. (34) Within this frame, we use the term social context to refer to the multiple individual attributes and circumstances that have significance because they have been socially constructed to produce positive, negative, or neutral experiences within (or in relation to) institutions and processes, including the justice system. (35) To the extent that a collective of individual justice seekers engaging with a particular institution or process has common social context factors, this commonality creates a general social context within which the particular institution or process operates. Further, the extent to which institutions and processes enable access to justice is, in our view, conditioned by the degree to which they adequately acknowledge, engage, and accommodate the social context of their stakeholders. (36) Enabling access to justice thus becomes an issue of understanding and engaging both the general social context of the population accessing the system and relevant variations arising from the specific social context factors that attach to each individual justice-seeker.

The range of social context factors that might be relevant for a particular asylum claimant includes various well-known identity markers that have socially-constructed significance in many forums. These include race, (37) disability, family status, sexual orientation, and gender identity. In addition, there are various circumstances that may be of heightened significance for those engaging with the refugee system in particular, including factors such as a lack of familiarity with the Canadian legal system and Canadian society more generally; limited or no knowledge of English or French; poverty; cultural and social differences; poor physical or mental health or both (often as a result of trauma); and a country of origin with precarious institutions and infrastructure. (38) Alone and in combination, these factors have the potential to create or exacerbate evidentiary challenges for refugee claimants and to impact their overall ability to access justice within the system. According to our social context conception of access to justice, the refugee claims process needs to be procedurally, substantively, and symbolically aware of, and responsive to, the impact of the social context of refugee claimants on their ability to gather, generate, present, and explain their evidence.

We have a dual basis for elaborating and applying this particular lens to evidentiary issues in the refugee claims process. First, a social context understanding of access to justice is consistent with other work examining Canada's refugee claims process and drawing attention to problems that can be regarded as impacting one or more of the procedural, substantive, and symbolic dimensions of access to justice. (39) For example, and perhaps most notably, a seminal study that assessed the decision-making process of the IRB over a decade ago clearly recognized the need for the individual circumstances of claimants to be considered by the system. (40) That study also highlighted explicitly the particular difficulties faced by board members, lawyers, and other participants in the claims process in treating evidence related to the psychological and cultural context of refugee claims, including challenges with vicarious traumatization and cultural insensitivity. (41) This study was followed by more specific works that analyzed and critiqued the claims process on various grounds. For example, there is a body of literature that addresses gender-based violence and persecution on the grounds of sexual identity. This literature emphasizes that refugee decision makers should appreciate the social context of dominant attitudes towards domestic violence, same-sex relationships, bisexuality, and other gender identities in countries of origin. These studies warn against the application of a Canada-centric lens in assessing the behaviour and credibility of claimants. (42) Related work has underscored the reality that the stark difference between a refugee claimant's own racial and cultural identity and that of the decision maker (the binary of the minority Other versus the dominant Canadian) can itself contribute to a risk of inequality and unfairness in determining credibility. For example, Sherene Razack describes modern or cultural racism as "the more covert practice of domination encoded in the assumption of cultural or acquired inferiority" (43) and argues that this "culturalized racism" is embedded in and affects decision-making in all contexts, but can be most prevalent in systems that raise (and debate) issues of race, minority cultures, and gender, including refugee law. (44) Other work has likewise highlighted challenges associated with the intersection between different aspects of refugee social context and specific features of the claims process, including how a lack of knowledge of Canada's legal system interacts with the availability and quality of legal representation; (45) how a lack of capacity in English or French interacts with the adequacy of interpretation services; (46) and how natural or trauma-induced memory loss interacts with expectations around detailed accounting of historical events. (47) Our social context conception of access to justice is informed by, and consistent with, these and other works that have assessed issues relevant to access to justice in Canada's refugee claims system, both before and since the 2012 reforms. (48)

Our second basis for elaborating and applying a social context conception of access to justice is that it is consistent with the approach that is evident in the general objectives and internal rules that structure the Canadian refugee system. (49) In particular, it is our view that the procedural framework governing Canada's asylum system contains a number of mechanisms aimed at enabling both flexibility and rigour. The flexibility of the system--evident in, for example, the general relaxation of the rules of evidence and broad reliance on the discretionary judgment of individual board members--appears to reflect Canada's commitment to offering protection to legitimate refugees, even in circumstances where it may be difficult to prove various elements of the claim. On the other hand, the same system seeks in various ways to maintain its own integrity by rigorously scrutinizing and testing the genuineness of the claimant's narrative and all supporting evidence. The presence of these dual aims--flexibility and rigour--can, in our view, be read as reflecting an understanding that the social context of refugee claimants poses inevitable challenges for the system tasked with determining their status and, correspondingly, that the legitimacy of the system requires an appropriate degree of recognition and accommodation of that social context. This reading of the system's internal features reinforces the suitability of the access to justice framework used in this study.

Applying a social context conception of access to justice to Canada's refugee determination process draws attention to the challenging circumstances in which refugees make their claims for protection, and in which decision makers must assess those claims. One key area in which challenges arise is gathering, presenting, and assessing evidence, as the social context of refugees frequently creates impediments to the provision of robust and reliable supporting material. However, while the evidence supporting a claim to protection may be uncertain, a claimant's entitlement to access to justice is not. Our review of 40 cases provides a glimpse of the degree to which the social context of claimants is considered and accommodated by Canada's refugee system in the context of evidentiary imperfections. Our methodology and details of the underlying cases are elaborated in the following section.

II. CASE DETAILS

Potential files for the UORAP's exploratory study were identified by a national network of refugee lawyers who were asked to recommend cases in which one or more evidentiary issues had arisen during the hearing process. The research team confirmed these recommendations and accepted cases exhibiting a range of issues. Once a case was selected for study, relevant refugee lawyers approached their clients to obtain written consent for the IRB to release its entire file to the research team (which generally includes documentary evidence filed by both parties, administrative notes by IRB case officers and Members, correspondence between parties and the IRB, and records of interim and final decisions), in accordance with a unique information-sharing agreement between the UORAP and the IRB. (50)

Materials collected through this process included the IRB's full written file, the audio recording of the hearing, and a copy of the decision. The UORAP team also acquired any additional documents in the lawyer's possession. Using this process, comprehensive materials were gathered for a total of 40 claims. Claimant names and other identifying details have been altered both here and in all other outputs from our work. (51)

The UORAP research team analyzed case files using a custom-designed analytical framework that aimed to capture key information about claimants, cases, evidence issues, and potential access to justice considerations in a consistent way. To develop the framework, the UORAP team read and analyzed a sample of cases and identified potential data collection fields. A draft tool was then developed and reviewed by expert advisers before being applied to 20 cases. The results from this pilot analysis were again reviewed, and the framework was further modified to ensure maximum effectiveness and consistency. The UORAP research team then applied the framework to all 40 cases, with all work being completed in duplicate, such that any discrepancies between independent researchers could be vetted and reconciled. The completed frameworks were then used to create a database of key features across all cases, allowing for information to be easily clustered and assessed. This database was used to generate the summary case information presented in this section, as well as to identify and categorize the various types of evidence issues that occurred across our 40 files.

It is important to explicitly underscore that the UORAP's files were not randomly selected and contain an inherent selection bias in favour of cases exhibiting evidence issues. It is also worth noting that we only collected files from trusted counsel, and our cohort thus also contains a bias in favour of cases that involved experienced refugee lawyers. (52) These factors, plus our very small sample size, render it impossible to use the UORAP cases to draw any statistical conclusions about the prevalence of evidence issues in Canada's asylum system. Our collection of materials does, however, provide an uniquely comprehensive view of how individual cases are being treated, making it possible to illuminate the kinds of issues that are arising within a system that is traditionally closed to scrutiny. Given the rarity of this access, we begin by providing a brief overview of our cohort.

Eleven of the claimants included in this study identified as female, and 29 identified as male. They ranged in age from 16 to 61 years old, with the majority being between the ages of 26 and 45 years old. (53) The top countries of origin for these claimants were: Afghanistan and Djibouti (four cases each); the Democratic Republic of the Congo and Iran (three cases each); and Burundi, Cameroon, Cote d'Ivoire, and Nigeria (two cases each). The remaining claimants originated from Colombia, Cuba, El Salvador, Eritrea, Ghana, Jamaica, Jordan, Kenya, Lebanon, Pakistan, Palestine/Jordan, (54) Peru, Rwanda, Senegal, St. Lucia, St. Vincent and the Grenadines, Syria, and Yemen. (55) Twenty-one of the 40 claims for protection were made inland and 19 were made at a Canadian Port of Entry. There were no DCO or Designated Foreign National (DFN) cases amongst our sample, although there was one co-claimant from a DCO.

The majority of the claimants included in our study pursued their claims for protection on multiple grounds. In 34 cases, the claim included a general fear of persecution and in 32 cases the claimant cited risk to life. (56) In 29 cases, non-state actors were identified as the primary agents of persecution.

Fifteen of the 40 claims in our sample were initially approved by the RPD and 3 rejected cases were reversed on appeal, for a total of 18 successful claims. Of the remaining 22 claims rejected by the RPD, 9 were unsuccessfully appealed to the RAD or the Federal Court, and 13 were either not appealed or the appeal result was unknown at the time the study concluded. (57)

A number of the files used in this study reflected similar social context factors. Perhaps most obviously, all of the claimants had experienced displacement and many fled countries that were in a prolonged state of instability. Twenty-six files revealed additional hardships associated with a lack of family or social support upon arrival to Canada. Poverty was also explicitly evident in 13 files, while the claimant's affluence was mentioned in four files. The majority of cases did not contain a direct or indirect reference to the claimant's socio-economic status.

A lack of facility with English or French was evident in a majority of cases, and 27 claimants relied on an interpreter during the hearing. In 12 cases, there was evidence of the claimant's experience with trauma, mental health challenges, or reduced mental capacity.

Finally, it is noteworthy that 12 of the claimants were detained at some point during their claim: 11 for immigration-related reasons and 1 for non-immigration reasons. On average, this detention lasted 35 days, though the length of detention ranged from 2 to 92 days.

It is important to emphasize that these observations are grounded entirely in the IRB and lawyer files that were available for each of these cases: the UORAP research team did not conduct interviews with claimants (or their counsel) to explicitly probe for relevant social context factors. As a result, it is almost certain that the attributes presented above do not fully reflect the social context of any of the claimants involved in these cases. Nonetheless, the social factors that are evident in these 40 files are consistent with trends in overall refugee populations and confirm some important collective features about asylum seekers accessing Canada's claims system.

III. EVIDENCE ISSUES

Evidence plays a critical role in refugee hearings. The asylum seeker is required to provide supporting materials to "[establish her] identity and other elements of [her] claim", and must explain the absence of key documents, including "what steps [she] took to obtain them." (58) The claimant must also satisfy the board member that the evidence presented to support her claim is credible and trustworthy. (59)

Frequently, the main evidence on which a protection decision is based is the claimant's story, as told through the written BOC and oral testimony at the hearing itself. It is significant to note that the claimant's testimony must be given the benefit of the doubt, unless she is found to lack credibility or her testimony is contradicted by accepted facts. (60) In general, adverse credibility findings are made on the basis of inconsistency or implausibility, (61) demeanor, (62) or contradictory trustworthy evidence. (63) Where these issues arise, the board member should give the claimant the opportunity to clarify contradictions or inconsistencies during the hearing. (64) Doubts regarding identity can also negatively affect credibility, and a claimant's inability to sufficiently prove identity can be fatal to a claim. (65)

In addition to her personal testimony, a claimant may submit oral testimony by other witnesses and experts, (66) or other types of documentary evidence, such as country condition documents (e.g. reports from reputable human rights organizations), (67) police reports, letters from local organizations, and news articles. The board member must consider the entirety of the evidence before making any determinations, (68) and must especially engage with evidence that appears to contradict her key findings about the case. (69)

In assessing the strength of the claim, the board member is not bound by the same strict evidentiary rules that guide many other kinds of legal or quasi-legal proceedings. (70) In practice, the circumstances surrounding claims for protection frequently lead to deficits in accessing, presenting, and reconciling corroborating records, and claim adjudication must frequently proceed on the basis of imperfect evidence.

The appropriateness of the ways in which Canada's refugee status determination system both produces and responds to these evidentiary challenges has been the subject of preceding and contemporaneous scholarly analysis. (71) For example, in Refugee Law's Fact-Finding Crisis: Truth, Risk, and the Wrong Mistake, Hilary Cameron analyzes a series of Federal Court decisions and concludes that the jurisprudential approach to fact finding in refugee law is not only unclear, but also self-contradictory, and that it ultimately "misguides" decision makers in how to approach their decisions. (72) Other important contributions have been structured around a selected issue, such as a particular source of evidentiary problems (e.g. the limits of memory), a particular area of evidentiary controversy (e.g. identifying country-specific practices of vulnerable groups), or a particular type of evidentiary shortcoming (e.g. lack of adequate evidence).

In contrast to related work, we began our analytical process by broadly surveying and categorizing the nature of the evidentiary shortcomings that were apparent in our specific selection of cases, and then tracing the relationships between those issues and the broad range of potentially applicable social context factors that also appeared on each file. We identified the various features of the decision-making system that appeared to be implicated in creating or exacerbating evidentiary challenges. As such, our study attempts to explore a wider breadth of evidentiary issues than previous studies but does not dive deeply into particular subsets, many of which could be the subject of expanded and dedicated work. It is as a result of both this broad approach and the limits of our underlying data set that we frame our study as exploratory.

The 40 UORAP cases exhibit a wide range of evidence-related challenges which we analyzed and subsequently grouped into three broad categories:

a) those with sufficiency issues, which arose where the file revealed a shortfall regarding the acquisition or presentation of one or more pieces of evidence;

b) those with internal consistency issues, which arose where the file revealed a conflict between two or more pieces of evidence generated by the claimant herself; and

c) those with external consistency issues, which arose where there appeared to be a discrepancy between a piece of evidence the claimant herself had generated and a piece of external information.

It is noteworthy that nearly all of the UORAP cases contained internal consistency issues, and that these issues frequently had a significant impact on the claim. As a result, we have devoted particular space to this category in the analysis that follows.

The survey of evidence issues presented in the following section has three objectives: first, it identifies, explains, and illustrates the primary clusters of evidentiary issues that appear to be most prevalent amongst UORAP cases; second, it illuminates the role that social context may have had in contributing to these issues; and third, it illustrates the varying ways in which decision makers responded to these challenges, including the extent to which they acknowledged the potential relationship between social context and the evidentiary imperfections. As already mentioned, over the course of our analysis four key system features also emerged as particularly significant in relation to evidentiary problems: ministerial interventions, interpretation, legislated timelines, and broad reliance on the discretionary judgment of individual decision makers. Although each of these features operates in an interrelated way, board member discretion plays a uniquely multi-faceted role across a wide range of procedural and substantive questions in any refugee claim. Given the way this particular system feature is embedded, we include some specific observations about its functioning as part of our broader analysis regarding the nature of the evidence issues we encountered. It is also the subject of dedicated consideration in our section on exacerbating system features, which is presented in Part IV.

A. SUFFICIENCY

The UORAP team flagged a sufficiency issue where a piece of evidence deemed relevant to the claim was not presented as part of the hearing process. The resulting gap in the evidentiary record may have been raised indirectly, or explicitly mentioned by the claimant, board member, or intervening minister. In all cases, the board member had to subsequently determine how the absence of the evidence impacted the validity of the overall claim.

Sufficiency issues arose in 34 of the 40 cases examined by the UORAP research team. Amongst these, 23 of the claims for protection were rejected and 11 were accepted. The root of each sufficiency issue was identified as one or more of the following: the missing evidence did not exist (14 cases); the evidence existed but was inaccessible (27 cases); the evidence was accessible but was deemed inauthentic (8 cases); or the evidence was accessible but inadequate because it only partially substantiated the relevant portion of the claim (11 cases). There were also two cases where the cause of the insufficiency was unclear on the face of the files.

The four forms of sufficiency issues identified above manifested themselves in various ways across the UORAP cases. For example, in eight cases the claimant experienced problems obtaining and authenticating original identity documents from her country of origin. This challenge was directly linked to several notable social context factors that are prevalent amongst refugee claimants, such as where the requirements of clandestine travel necessitated the abandonment or destruction of genuine identity documents, or where documents were sought from countries experiencing general instability or applying inconsistent documentation practices. In each of these cases, board members needed to decide whether to accept the claimant's explanation for why evidence was missing and whether efforts to obtain it were reasonable. In four of the eight cases involving missing identity documents, the board member's consideration of the imperfect documentation was explicitly informed by information contained in IRB resources on the relevant country of origin, such as National Documentation Packages or Responses to Information Requests. (73) In all eight of these cases, there appears to have been implicit recognition by the board member that the general and particular social context of refugee claimants can make it difficult to provide perfect evidence of identity, including where efforts were made to investigate an underlying cause for the imperfect evidence. However, none of the board members explicitly named or overtly considered the impact of specific social context factors on the evidentiary challenges, even where imperfect identity evidence was ultimately accepted as sufficient.

Another cluster of sufficiency issues was connected to an absence of police reports to substantiate key elements of the claim. This tended to occur most frequently where a police report for an incident was never produced (six cases) or where the report was inaccessible due to danger or other difficulties in the country of origin (eight cases). Collectively, the UORAP cases suggest that police reports are often treated as very reliable evidence of key incidents, and that their absence is frequently viewed with concern. In all 14 of the cases where a lack of police reports raised a sufficiency issue, the minister's representative or the board member herself expressed an explicit desire for police documentation to substantiate particular elements of the claimant's story. (74) These requests were made even where a variety of social context factors appeared to make it difficult or impossible to access the relevant documentation. For example, in one case the claimant alleged that she was a victim of repeated gender-based violence at the hands of a man with important social status in her home country. She testified that police met her effort to report a particular assault with skepticism and dismissal, and further noted that this response was consistent with their general reputation for viewing domestic violence as a private matter. Thus, a police report was never generated, and the claimant did not reach out to police again, despite further assaults. This made it impossible for her to produce the evidence that was being explicitly sought during her refugee hearing in Canada. (75) In other cases, police reports were either available only in copy rather than in an original version or were not available because the claimant was fearful of approaching the police to seek the report on file. In all of these cases, the accessibility, cooperativeness, or professionalism of police--and the associated actions of the claimant with regard to her willingness to engage with officers--differed substantially from what might be expected in the Canadian context, clearly illuminating the importance of properly considering social context when determining how to treat these kinds of issues.

Our analysis further revealed that the willingness and ability of board members to contextualize their expectations regarding police reports within the country of origin's circumstances, and to assess claimant actions and explanations within the appropriate social context, was inconsistent. In some cases, board members clearly attempted to understand the surrounding circumstances when assessing the implications of a missing document, while in others the insufficiency was deemed fatal without any apparent consideration of the myriad of potential underlying reasons why the documentation was not provided. To the extent that access to justice requires that individual and collective circumstances be properly considered by systems tasked with adjudicating rights, this was not routinely exhibited across the UORAP cases involving imperfect evidence in the form of police reports.

The UORAP cases also revealed sufficiency issues around a variety of other forms of documentary evidence. One particularly complex issue arose in three cases where LGBT claimants were required to provide evidence regarding their sexual orientation. In each case, the claimants testified that their sexual identity had been concealed for most of their lives out of fear of persecution, and as a result they did not have robust documentation to support their claims. These cases all involved the convergence of several social context factors including gender, sexual orientation, cultural differences, and country of origin circumstances, which, cumulatively, made it difficult for claimants to provide the evidentiary materials the board was seeking. (76) Two of these cases also involved written interventions by the minister and language barriers, both of which appeared to compound difficulties of presenting what little evidence was available in a clear and coherent way.

Again, respective board members took divergent approaches when assessing the sufficiency of the documentary evidence before them. In one case, the board member appeared to demonstrate a rich and nuanced understanding of the claimant's social context when considering the sparse evidentiary record supporting the claim. This was evident in various ways, including her recognition that photos the claimant had posted online may have been perceived by members of his home culture as an expression of same-sex orientation. Likewise, she acknowledged that a variety of interpretation problems had occurred over the course of the claim, and that these likely impacted the claimant's ability to share his story in a fulsome way. This member considered these and other factors when weighing the credibility concerns put forward in the minister's written intervention.

In contrast, the board member in a second sufficiency case that also involved an LGBT claimant appeared to give very little consideration to the asylum seeker's complex social context. While acknowledging the prevalence of violence and discrimination on the basis of same-sex orientation in the claimant's country of origin, and the associated need for secrecy in expressing gender identity inside that country, she lamented a lack of evidence to verify the claimant's sexual orientation. Moreover, she gave little weight to a letter from a gay rights organization corroborating the claimant's statements that he identified as gay and dismissed a photo of him participating in a Gay Pride parade because she concluded that "anyone, including heterosexuals, can be part of the organization and take part in the parade." No other consideration of social context was apparent on the file and the claim was rejected. It is noteworthy that on review, the RAD determined that the board member's treatment of the evidence in this case amounted to a reviewable error. While this reversal illustrates the constructive contribution that the RAD can make in promoting appropriate decision making, the contrast between approaches to sufficiency issues at the RPD illuminates both the important interplay between evidentiary imperfections and social context, and the variance in approaches between board members who have been imbued with significant power, flexibility, and discretion with regards to their decision making.

B. INTERNAL INCONSISTENCY

The UORAP team identified internal inconsistency issues where there were discrepancies between or within the items of evidence presented by, or emanating from, the claimant herself. These are in contrast to external inconsistencies, which engaged a piece of evidence that was not claimant generated. For example, we would classify an inconsistency between the narrative provided by the claimant in her BOC and a record of responses provided to a CBSA officer at a port of entry interview as an internal inconsistency, while we would classify an inconsistency between the BOC and a non-governmental organization (NGO) report as an external inconsistency. Our analysis reveals three specific types of evidentiary interactions that generated particularly prevalent or problematic issues relating to internal inconsistency: inconsistency between pre-BOC evidence and the BOC itself; inconsistency between the BOC and subsequent claimant testimony; and inconsistency between the BOC and other claimant-generated evidence, in particular social media profiles. Each of these is elaborated below.

1. INCONSISTENCIES BETWEEN PRE-BOC EVIDENCE AND BOC

Ten of the UORAP's cases contained an internal inconsistency arising from discrepancies between the BOC and some form of pre-BOC evidence. In seven of these files, the relevant pre-BOC material consisted of CBSA interview notes; in two it was a visa application; and in one an application for Humanitarian and Compassionate (H&C) consideration. In all ten cases, the relevant discrepancy was flagged--and relevant pre-BOC evidence introduced to the hearing process--as part of a ministerial intervention. (77)

Our analysis revealed that social context factors were frequently implicated in the inconsistencies between the BOC and pre-BOC material. For example, in one of the UORAP cases the claimant understated to the CBSA interviewer the amount of time that had elapsed between leaving his home country and his arrival in Canada. When confronted with the discrepancy between that erroneous information and the timeframes included on his amended BOC, the asylum seeker explained that he had been advised to lie by a smuggler upon whom he was relying for a false passport and a travel route. According to the claimant, the smuggler advised him to say that only one month had elapsed since his arrival (rather than two) because delay could be used to discredit the claim. In another similar case, the asylum seeker contended that he was misinformed by an informal network of travellers about the ability to claim refugee status at the port of entry, and he thus thought he could only seek protection from inside the country. The claimant in this second case indicated that, as a result of this misinformation, he stated at the port of entry that he planned to return home because he thought this statement would assist with gaining entry to Canada, such that he could subsequently lodge his claim for protection. When he later cited fear of returning home when filing for protection, an inconsistency was flagged.

These cases demonstrate that claimants may have partial or erroneous information about the rights afforded to asylum seekers and the various factors that may disqualify or weaken a claim for protection. This misinformation--coupled with a general mistrust of state authorities, a lack of awareness about the implications of inconsistencies, and any number of other factors--may prompt some asylum seekers to mislead officials in their initial statements, even if they do have genuine grounds for seeking protection.

A comparison between these cases also reveals that Canada's claims system may respond unevenly to inconsistencies between pre-BOC evidence and the BOC itself, even when the underlying causality appears to be based on very similar social factors. For example, the board member in the smuggler case concluded that the claimant's explanation for his misrepresentations was reasonable, noting specifically that his dependence on the smuggler for documents and travel assistance needed to be considered when assessing the likelihood that he would follow the smuggler's advice. In contrast, the board member in the second case rejected the claimant's explanation that the inconsistency resulted from misinformation provided by others. The board member noted that the claimant had explicitly asked whether future applications would be compromised by signing the visitor visa. They also and contended that it was not credible that he would have made this inquiry while at the same time choosing not to reveal that he was fearful of returning home. The board member concluded that the claimant must not have actually feared for his life and rejected the claim. No social context factors were explicitly considered. In this case, the claimant's concern about the impact of the visa application on future applications could be seen as consistent with an intent to subsequently file a claim for refugee protection--particularly when the claimant's apparent confusion about the claims system is taken into account.

In both the visitor visa case and the smuggler case, the files contain indications that an internal inconsistency may be attributable to a lack of independent understanding of the refugee claims process. In the visitor visa case, there is no evidence that as part of the overall credibility assessment the board member grappled with the impact this might have exerted on the claimant's behaviour, while in the smuggler case this constituted a significant part of the board member's assessment. Read together, these cases suggest a lack of consistency in how the social context factors underlying evidence imperfections are identified and considered.

Our files also demonstrate that a lack of language capacity may be implicated in producing inconsistencies between BOC and pre-BOC evidence. For example, in one case a ministerial intervention impugned the credibility of the claimant on the grounds that he had stated to a CBSA officer that he had financed his travel to Canada by selling his car, while the BOC indicated that he had sold some gold items and borrowed money from a sibling. Faced with the discrepancy at the hearing, the claimant contended that the inconsistency arose from difficulties with the telephone interpretation provided at the port of entry: he explained that he had said that he had owned a car, but not that he had sold it. The board member ultimately rejected the minister's arguments against credibility and accepted the explanation relating to an interpretation error as reasonable. This case demonstrated thoughtful consideration of relevant social context.

2. INCONSISTENCIES BETWEEN BOC AND ORAL TESTIMONY

In 17 of our 40 files, there was an issue relating to inconsistencies between the BOC and subsequent oral testimony. In 14 of these files, interpretation difficulties were also evident, and in 4 the hearing audio revealed a high degree of emotion during the oral testimony. Fourteen of these claims were ultimately rejected and interpretation was required in all but two of the rejected cases. Further, issues with the adequacy or accuracy of interpretation were identified in every case in which it was needed. In contrast, interpretation was not required in any of the three claims that were accepted. It is also noteworthy that in 6 of these 17 cases, the claimant did not have assistance from counsel to complete their initial BOC (one was also self-represented at the hearing), and that one of the relevant BOCs was completed while the claimant was detained.

Our analysis of these cases once again reveals that social context is both relevant to the production of inconsistencies and is subject to uneven treatment by decision makers. For example, one claimant sought protection on the basis of domestic violence perpetrated by her husband via a forced marriage. In support of this claim, she produced police and medical reports documenting three violent incidents. Her initial BOC, however, only detailed two incidents. The failure of the claimant to include the third incident in her initial BOC was treated by the board member as a key inconsistency that undermined her credibility. When asked to explain the discrepancy, the claimant testified that she had actually brought the omission of the third assault to the attention of her counsel before the BOC was submitted and requested that it be corrected. She alleged that her counsel refused, noting that amendments could be made at the time of the hearing. The claimant further testified that this and other disagreements led to her retaining different counsel for the hearing, and to her providing her new counsel with a revised BOC which referred to three assaults. The second counsel did not realize, however, that the BOC she received from her new client had been revised and needed to be submitted to the board--a point of confusion that she attributed in part to language barriers. It was only during the hearing itself that the board member and counsel realized they were working from different versions of the BOC: one that described two assaults and one that described three. Despite this confusion and various corroborating documents supporting the process challenges the claimant had experienced, the board member ultimately dismissed the claimant's explanation about the inconsistency as "not logical".

The board member's reasons in this case appear to demonstrate a failure to properly consider her social context. Relevant circumstances include the significant misunderstandings that can arise when claimants and counsel do not share a common language, the challenges of navigating a complex and high stakes legal process in a foreign country, and the confusion that may result from switching counsel part way through any proceeding. This case also illustrates the potential for social context factors to have a compounding detrimental impact on the capacity of asylum seekers to present a clear and compelling claim. On top of the factors already mentioned, the claimant also exhibited high emotion during her testimony when describing the assaults, was only able to secure crucial country of origin evidence at the last minute (and thus needed special permission to present it at the hearing), and encountered a number of challenges with interpretation during the hearing itself. Although the board member in this case opened her decision with reference to the IRB's Guideline on Gender-Related Persecution, the reasoning does not indicate consideration of the numerous and compounding social context factors that likely impacted the way evidence was presented in this case.

Similar issues appear in another of the UORAP's inconsistency cases, this one involving a female asylum seeker whose claim for protection centered on incidents of ethnicity-related violence. The inconsistency at issue related to the specific dates the violence was alleged to have occurred: the BOC stated that the assault happened in January 2013, while the claimant testified it was December 2012. It is once again noteworthy that the claimant appears to have completed the initial BOC without the assistance of counsel. She also appeared at her hearing via videoconference due to pregnancy-related health issues that made it difficult for her to travel; experienced numerous difficulties with her interpreter; did not share a common language with her counsel at the hearing; and was highly emotional throughout her testimony. None of these social context factors were explicitly considered by the board member, who focused exclusively on the internal inconsistency regarding the dates of the alleged assault. He ultimately concluded that, given the significance of this incident to the claim for protection, the discrepancy in dates "could not be easily overlooked". Refugee status was thus refused by the RPD without any consideration of the myriad of social context factors which may have contributed to the inconsistency that was found to be so central to the question of claimant credibility.

It is noteworthy that, on appeal, the RAD found that the board member in this second case had erred. The RAD acknowledged that the RPD was entitled to draw a negative inference as to credibility due to inconsistencies in the evidence but held that the contradiction on dates alone "could not be fatal". The stance taken by the RAD in this case seems to implicitly support the conclusion that decision makers must take into account at least the general social context of refugee claimants when weighing the significance of evidentiary inconsistencies. In a context where evidence can be expected to be imperfect to some degree, the mere existence of a contradiction cannot be fatal to a claim. It is also significant that at the eventual rehearing of this case the claimant presented additional evidence, including a medical report identifying that the assault had occurred on the December date and information about both memory loss in general amongst refugees and the results of a psychological assessment of the claimant in particular. These additional documents suggest that the claimant benefited from more time to prepare an enhanced evidentiary record to support her claim, and thus also highlight some of the challenges associated with legislated timelines. Ultimately, the rehearing resulted in the claimant being granted refugee status in Canada.

3. INCONSISTENCIES BETWEEN BOC AND OTHER INTERNAL EVIDENCE

Internal inconsistencies between the BOC and other claimant-generated evidence arose in seven of the UORAP's files. These included two cases where a ministerial intervention was based on allegedly inconsistent content on the claimant's social media profile (Facebook). The use of social media information as a basis of inconsistency is worth highlighting both because the use of social media platforms is increasingly widespread, and because it represents a relatively new source of evidence that brings a distinct social context associated with the medium itself.

The potential complexities associated with determining inconsistency on the basis of social media postings were highlighted in a case involving a claimant alleging fear of persecution for same-sex orientation. In this case, the claimant's credibility was questioned because he presented himself as heterosexual in various social media postings made while he was in transit from his country of origin. The claimant's explanation for the posts was two-fold: first, that he had suffered persecution and was reluctant to publicly self-identify as gay out of fear of harm; and, second, that his social media profile was in part a ruse to attract a sympathetic Europe-based woman who might eventually help him immigrate to a safe country. The board member dismissed both of these explanations, despite a letter from an NGO with expertise working with gay refugee claimants that attested to the claimant's difficulty in being open about his experiences. The claimant was ultimately found to lack credibility and his refugee claim was rejected. The board member in this case appeared to apply a series of fixed assumptions about how an individual belonging to a particular identity group ought to behave (78) rather than considering either the claimant's particular social context or grappling with the complex norms attached to social media usage. The internal inconsistency was deemed fatal to the claimant's credibility and his claim for protection was denied.

C. EXTERNAL INCONSISTENCIES

Inconsistencies between claimant-generated evidence and other external evidence arose in seven of the UORAP's files. In three of these, the external inconsistency related to the claimant's attempt to establish identity. In the remaining four, it related to evidence on key incidents underlying the substantive claim for protection. The external inconsistencies most frequently engaged general information about practices or circumstances relevant to the claim in the country of origin. For example, where the board member referred to a board-produced National Documentation Package, or where a claimant introduced evidence from a third-party source on country conditions, including, for example, NGO reports detailing the prevalence of different types of persecution. The cases demonstrate two main types of inconsistencies that may arise in relation to the availability and authenticity of external reports, each of which is detailed below.

First, country of origin information relating to the form and substance of official documents, or to the protocols of official processes, may be inconsistent with information provided by a claimant regarding her ability to procure certain types of evidence. This type of problem may arise in relation to, for instance, the availability of identity documents (e.g. a birth certificate) or other official documents to corroborate key incidents (e.g. police or medical reports). For example, one claimant offered a photocopy of an ostensibly official identity card as proof of her identity. The board member sought to test her evidence by asking her to explain the steps she had taken to obtain the card. The claimant's explanation included a procedural step--obtaining a court-issued document proclaiming her nationality--that was not noted in the board-generated country of origin materials. This inconsistency became one basis for the board member's finding that the claimant's identity evidence was not credible. The approach taken by the board member in this case places significant weight on the presumed comprehensiveness and currency of a board-generated document that may be up to seven years old, and which may be oriented to describing only the usual steps in an official process, rather than all possible or exceptional steps. Over-reliance on these documents may lead to a failure to consider the broader social context, including the fact that official procedures may be both inconsistent and vary over time, particularly in countries with under-developed and crisis-affected state infrastructure.

The second type of external inconsistency arose in cases where the outside country of origin information was insufficiently corroborative of (or even contradicted) key substantive elements of the claim. This situation may occur with information relating to the likelihood of persecution, the absence of state protection, or the feasibility of certain specific events. For example, one male claimant from an Arabic-speaking country alleged that he had been subjected to death threats from the relatives of a female cousin. The death threats were allegedly motivated by the discovery that the claimant and his cousin were engaged in a romantic relationship, against the wishes of the cousin's parents. The claimant alleged that he feared a so-called honour killing and presented affidavit evidence from his parents and four siblings, as well as from his town clan's chief, attesting to the threats against him. The board member in this case relied heavily on country of origin information regarding honour killings to conclude that the claimant was not credible. In particular, he found that the idea of an anti-male honour-killing was inconsistent with public documents, which emphasized anti-female honour-killings and were silent on the potential of male victims. The board member also found it difficult to believe that murder would have been threatened, given country of origin documentation that suggested there are a variety of ways in which so-called honour crimes are punished, with murder being the last resort. The board member used this externally generated information to opine that it was more likely that the woman's father would require the claimant to marry her (as he had offered to do), rather than kill him. On the basis of the perceived inconsistencies between general information about honour killings and the claimant's specific circumstances, the board member found the claimant lacked credibility and refused the claim. The reasons in this case do not mention the claimant's corroborating affidavit evidence. Significantly for our purposes, they also do not indicate that the decision maker grappled with how to reconcile inconsistencies between the general country of origin information and specific evidence regarding the claimant's individual social context.

In our view, Canada's refugee claims system must consider both individual and collective social context in order to ensure that claimants are afforded access to justice. Our review of 40 cases containing evidence issues allowed us to cluster and examine the way that these challenges are manifesting themselves and to identify the degree to which a claimant's social context was considered as part of the assessment of evidentiary imperfections. Overall, the degree to which social context was acknowledged and considered by individual board members appeared to be inconsistent, which causes a challenge for the integrity of the overall system. We discuss this point further in the section that follows, which includes dedicated commentary on the system's reliance on the discretionary judgment of individual board members, as well as on each of three other system features that the UORAP cases indicate may be causing or exacerbating access to justice deficits relating to imperfect evidence.

IV. EXACERBATING SYSTEM FEATURES

Our review of issues relating to evidence across the UORAP cases led us to identify four particular system features that appeared repeatedly and seem to demonstrate the potential to create or exacerbate access to justice deficits: ministerial interventions, legislated timelines, interpretation challenges, and board member discretion. In this section, we provide some observations on each of these features.

A. MINISTERIAL INTERVENTIONS

The minister of immigration intervened in 17 of 40 cases and appeared in-person in 5. Concerns about credibility were present in 16 of these cases, and 11 also involved concerns regarding identity. Of the 17 cases where a ministerial intervention occurred, 5 claims were ultimately accepted and 12 were rejected.

Neither publicly available documents nor patterns across the UORAP cases reveal any minimum threshold of significance for triggering a ministerial intervention, let alone any threshold that considers the role that various social context factors might have played in generating the area of concern. The ministerial interventions that we reviewed as part of this study suggest that, regardless of circumstance, any degree of evidence-related weakness or discrepancy may be sufficient to necessitate the production of corroborating evidence, placing significant burdens on both the claimant and the IRB. Moreover, where corroborating evidence cannot be provided, there is a heightened risk that the claim will be rejected on the basis of credibility concerns.

It is noteworthy that some board members appear to be conscious of the potential distorting impact of ministerial interventions where discrepancies might be regarded as either minor or explicable. For example, in an intervention case where the claimant explained a discrepancy over how he funded his journey to Canada as an interpretation error, the board member was willing to still apply a reasonableness standard to the explanation, without seeking corroboration. Similarly, in a case where the claimant completed her H&C application without counsel, the board member not only applied a reasonableness standard to the explanation, but openly questioned the value of the intervention itself: "[T]hose credibility allegations raised by the Minister... haven't been helpful in any form of assessment of this claim at all. I don't find that they, in my mind, have been serious enough to have been raised in the first place." These cases demonstrate the potential of individual board members mitigating the negative impacts of an intervention and represent an intersection between the flexible and rigorous components of the overall system. These cases also demonstrate, however, both the coercive power associated with interventions and the potential they have to exacerbate access to justice deficits for claimants, especially if exercised in an arbitrary manner.

In our view, some of these issues could be structurally mitigated by introducing a social context-informed threshold for triggering ministerial interventions. This threshold could help ensure that government officials reviewing cases consider not only whether there is a credibility issue in the file, but also the severity and materiality of the issue, its cause, and its relationship to the actual circumstances facing the claimant. The threshold could also require consideration of the extent to which corroborating evidence would be helpful, as well as whether it is reasonably available to the claimant. Such an approach would offer a better acknowledgement that the general social context of refugee claims can be expected to produce a variety of inconsistencies and other potential indicators of credibility issues and ensure that heightened scrutiny is reserved for cases where the quality or quantity of issues are sufficiently severe, material, unusual, or unexpected, given the overall circumstances. We also note that there is a significant difference between paper and in-person interventions, from both a resources perspective and when considering their impact on access to justice for the claimant. We would thus suggest the articulation of two thresholds: a lower one to justify ministerial submissions, and a higher one to justify attendance by the Minister at the hearing itself.

B. LEGISLATED TIMELINES

Only 21 of the 40 UORAP cases proceeded according to legislated timelines: In the remaining 19 cases, the time period between filing the BOC and the RPD hearing was longer than 60 days. In addition, there is evidence that 16 claimants struggled to comply with various timelines relating to the submission of evidence. This was most frequently apparent in the form of a formal request to submit additional documents after the submission deadline--a request that was granted in all but three cases. There were also 18 cases where the claimant requested to submit additional documentation at the hearing itself--a request that was granted in all but one case. (79) As indicated in our sufficiency analysis above, there were also a number of cases where evidence deemed important to the case was never presented.

The circumstances leading to delays in the hearing itself varied across the files. In some cases, delays were caused by slow security screening or late ministerial interventions, in other cases it was counsel illness or successful requests for adjournments to obtain crucial evidence. In a number of the longer duration cases, the added time appears to have provided a beneficial opportunity for the claimant to obtain additional supporting evidence.

It is clear that overly aggressive timelines for the production of evidence have the potential to increase challenges associated with a variety of social context factors, including those related to the distance between the claimant and her country of origin, weak documentation systems in certain regions, the inability to travel with authentic documentation, and many others. The potential access to justice deficits created by strict rules that are difficult to meet can be exacerbated by the flexible power given to individual decision makers to override timeline strictures by granting extensions, adjournments, and other relief mechanisms. The uncertainty attached to a system which is dependent on such an override function is detrimental to all actors in the system: it is stressful and time-consuming for claimants and their counsel; unnecessarily resource intensive for board members who must engage these applications; administratively cumbersome for the Board's scheduling processes; and inconvenient and inefficient for other system actors such as interpreters, witnesses, and minister's representatives. Overall, it is our view that, while some form of legislated timelines is likely important to ensure the efficiency of Canada's refugee claims system, the current legislated timelines place unnecessary burdens on the system and relies on ad-hoc solutions. Overall, this exacerbates uneven application and access to justice deficits, and a review of the formal framework--in consultation with all stakeholders in the system--is recommended. (80)

C. INTERPRETATION CHALLENGES

An interpreter was present in 27 of the UORAP's cases. Twenty-two of these interpreters were present in-person and five were present by teleconference. There was explicit evidence that the quality of interpretation was an issue in 17 of the 27 cases. Further, our review of audio recordings indicates that, in all but two of these cases, there were points during the hearing where the interpreter did not understand what was being said, as well as times where the claimant seemed unable to fully understand or express themselves owing to an interpretation problem. Most explicitly, in 11 cases, the interpreter did not speak the same dialect as the claimant, and in 15 cases there was evidence in the form of statements from the claimant or their counsel that the interpreter did not accurately interpret what was being communicated by the claimant. Of the 27 cases in which interpretation was used, 18 claims were rejected by the IRB, (81) and 9 were accepted.

It is noteworthy that, of the 17 claims involving a ministerial intervention, 13 also involved a need for interpretation or translation or both. Moreover, nine of these files evidenced a significant problem with the quality of interpretation, and a further three files indicated at least a difficulty in relation to interpretation or translation. While we do not have sufficient data to assess causality, it is important to underscore the potential concern that inconsistencies triggering ministerial interventions may be a result of the inadequacy of interpretation and translation (especially in circumstances of stress, trauma, detention, and so on) and not as a result of an underlying weakness in the claim itself.

It is trite to note that language differences are likely to be a key social context factor in many claims for refugee protection, and that these differences can have an enormous impact on the ability of the claimant to properly provide critical testimonial evidence in support of her claim. It is essential that the refugee determination process adequately responds to these circumstances by providing interpretation, and the Canadian system should be commended for committing to the provision of these services. It is essential, however, that interpretation is not only available, but that interpretative services are of high quality, including being able to take into account nuances of social and cultural idiom and contextual background. While technically accurate interpretation must be seen as a bare minimum, Robert Gibb and Anthony Good assert compellingly that literal translations of applicant speech in refugee legal proceedings can produce distorted communication between the applicant and the decision maker, "due partly to the fact that words depend for their meaning on how they are combined with other words... [thus] an understanding of this context is required for accurate translation and interpretation to be possible." (82) Gibb and Good acknowledge that the goals of contextual and idiomatic interpretation mean that the responsibilities of a competent interpreter in a refugee hearing are complex, as they have to balance the obligation to translate an applicant's answers honestly, while exercising independent judgment on a range of matters, including "whether or not to intervene to explain a cultural misunderstanding [and] how to negotiate different registers of speech without potentially damaging the perceived credibility of an applicant's... narrative". (83) Yet, despite the challenges of contextual, culturally sensitive, and nuanced interpretative services, there are clear benefits to the refugee determination system (fairness and accuracy) and to the applicant (solace and support) that flow from these services. (84) Further, there are clear risks associated with poor interpretative services, including the potential denial of protection because of linguistic or cultural misunderstanding or both. It is of particular concern that quality issues were evident in the majority of UORAP cases where interpretation was used, and it is incumbent on the system to ensure that adequate processes are in place to ensure sufficient quality.

Further, where issues of language barriers do arise, board members should be required to take them seriously and postpone the hearing if necessary, to ensure accurate communication--it is insufficient to simply continue despite the challenge. The existence of language barriers needs to be considered in other parts of the system as well, including by both government officers and board members when they are considering the presence of apparent inconsistencies on the file. Recognizing the frailties in interpretation and the challenges that are caused by language barriers requires ongoing consideration and it is problematic to assume that the mere provision of an interpreter fully negates these barriers. Sufficient training and an ongoing contextualized approach that fully considers the impact of this critical social context factor must be followed throughout the refugee claims system.

D. RELIANCE ON BOARD MEMBER DISCRETION

The UORAP cases reveal that the refugee claims system relies heavily on a wide-ranging and generous discretion given to individual decision makers: a powerful system feature that can be used to ensure the claims process does not become overly formal or rigid, but which also risks rendering it inconsistent and unprincipled. It is significant to note that, while the discretion we have identified includes power that falls within the formal legal concept recognized and regulated by general administrative law (such as the express discretionary power board members have to admit late evidence), our usage also extends to a less technical aspect. In particular, for the purposes of this piece our use of the term discretion encompasses the power that decision makers have to acknowledge, weigh, interpret, and understand various facts, circumstances, and supporting evidence--including social context factors. The refugee system relies heavily on the judgment of individual decision makers in this regard and gives them significant power--a system feature that was prevalent across many of the UORAP cases. (85)

Our analysis demonstrates that, in the context of the presentation of evidence, flexible board member discretion is frequently used to compensate for--or override--other problematic system features, rather than to enhance and meaningfully complement a principled overall framework. Further, the application of this flexibility appears to be wide-ranging and inconsistently applied, which creates a significant risk of injustice. Our cases demonstrate an unevenness that has also been noted by others. For example, Audrey Macklin, a former decision maker with the RPD, proposes that key determinations in the refugee system are "necessarily and inexorably subjective" and"inconsistent", (86) a reality that she attributes to the internal biases and life experiences of individual decision makers. Hilary Cameron and Sean Rehaag make similar findings, (87) and both favour a normative decision-making framework that considers error preference, arguing in favour of a system that guards against the "worst error": denying a claim that should have been granted. (88)

Our study reinforces these observations, and further suggests that a more concrete and binding framework may be necessary to ensure that board member power is exercised in a way that appropriately considers the social context factors of claimants. Such a framework could include, for example, more robust guidance regarding the presentation of evidence, with clearly articulated principles regarding when social context factors would favour the application of discretionary variation. Ultimately, a proper recognition of the relevance and impact of social context may justify establishing a more substantial "benefit of the doubt" principle for refugee claimants.

Further, ensuring that IRB training materials both provide clear guidance on the application of board member discretion and are publicly available, would provide a benchmark against which decisions based strictly on board judgment could be measured and held to account. Regardless of the strength of these materials, the current practice of protecting their content renders them impotent in terms of their ability to assist with quality control within the system.

We note with particular concern the practice of board member discretion being used as an instrument to compensate for other problematic system features, something we saw most frequently in cases involving unreasonable timelines, but which was also apparent where there were unreasonable ministerial interventions or challenges with interpretation. While flexibility is a useful tool for ensuring that individual claimant's personal circumstances are properly considered, it should not be required to overcome other system features that are themselves fundamentally flawed. We thus favour a system which has clear, reasonable, and transparent rules, including around when and how discretion ought to be applied.

The UORAP cases clearly demonstrated that challenges relating to both insufficiency and inconsistency of evidence are being exacerbated by the four system features discussed above. We note that there are also a number of other system features which, while not highlighted by our specific file set, also have the potential to contribute significantly to access to justice deficits throughout the process. These include the presence or absence of counsel and the treatment of particularly acute vulnerabilities, including through a formal vulnerable persons' application. Our overall recommendation is that system features be carefully considered through a lens that recognizes their potential mitigating or exacerbating effects on the system's ability to respond to the social context factors of the individuals with whom it is engaging.

CONCLUSION

A broad array of both general and individualized social context factors can be expected to raise significant challenges for refugee claimants as they attempt to gather and present evidence to support their claims for protection. While refugee claimants are responsible for demonstrating their case for protection, the system itself is responsible for ensuring access to justice for those attempting to do so. Our analysis of 40 cases decided under the parameters of Canada's reformed refugee status determination system allowed us to identify and categorize the types of evidentiary problems that may be appearing across the system. We also glimpsed the way that underlying social context is being considered by decision makers, as well as how key system features can create or exacerbate these challenges.

Our work revealed a troubling unevenness in the extent to which the system recognizes and responds to social context. It also identified four particular system features with the potential to create or exacerbate challenges: ministerial interventions, legislated timelines, interpretation, and reliance on broad board member judgment. While the UORAP study was by design limited and exploratory, our review raises concerns about the way that evidentiary issues are being treated by the refugee claim system and, ultimately, the extent to which the system as a whole is providing access to justice. Further and deeper examination of the issues our cases have surfaced is strongly recommended: while imperfect evidence is an inherent part of refugee status determination, uncertain justice must not be.

JENNIFER BOND & DAVID WISEMAN ([dagger])

([dagger]) Both authors are Associate Professors in the Faculty of Common Law at the University of Ottawa. This work benefited from generous support from the Law Foundation of Ontario and the Social Sciences and Humanity Research Council (SSHRC). We are grateful for contributions from all members of the University of Ottawa Refugee Assistance Project (UORAP) research team, including JD students who contributed to our research through their participation in the UORAP courses at the University of Ottawa. We also acknowledge the many contributions of UORAP's founding members: Peter Showler, Adam Dodek, and, in particular, Emily Bates, as well as early methodological advice from Janet Cleveland and Catherine Dauvergne. This piece benefited from editorial support from Dr. Eliza Bateman and Laura Macintyre. All errors and omissions are of course our own.

(1) An estimated 55,040 claims for asylum were filed in 2018. This compares with 50,395 claims in 2017, 23,870 claims in 2016, and 16,055 claims in 2015. See Immigration, Refugees and Citizenship Canada, "Asylum Claims by Year" (last modified 2 February 2019), online: Government of Canada <canada.ca/en/immigration-refugees-citizenship/services/refugees/asylum-claims.html>.

(2) There has been increased media attention on the Canadian asylum system over the past twelve months, due in part to a series of high-profile arrivals at Canada's land-border with the United States. See Michelle Zilio, "Number of Asylum Seekers Crossing into Canada from U.S. Continues to Rise, Feds Say" The Globe and Mail (3 May 2018), online: <theglobeandmail.com/politics/article-number-of-asylum-seekers-crossing-into-canada-from-us-continues-to>. At the time of writing, the IRB was under a government-ordered review and had just released the Report of the Independent Review. See Immigration, Refugees and Citizenship Canada, News Release, "Government Launches Review of the Immigration and Refugee Board of Canada" (9 June 2017), online: Government of Canada <canada.ca/en/immigration-refugees-citizenship/news/2017/06/government_launchesreviewoftheimmigrationandrefugeeboardofcanada.html>; Immigration, Refugees and Citizenship Canada, "Report of the Independent Review of the Immigration and Refugee Board: A Systems Management Approach to Asylum" (10 April 2018), online (pdf): Government of Canada <canada.ca/content/dam/ircc/migration/ircc/english/pdf/pub/irb-report-en.pdf>.

(3) Jennifer Bond, David Wiseman & Emily Bates, "Troubling Signs: Mapping Access to Justice in Canada's Refugee System Reform" (2016) 47:1 Ottawa L Rev 1 at 7, n 9 [Bond, Wiseman & Bates, "Troubling Signs"].

(4) For more detail about the formation of the UORAP and its mandate, including its programmatic activities, see Emily Bates, Jennifer Bond & David Wiseman, "The Cost of Uncertainty: Navigating the Boundary Between Legal Information and Legal Services in the Access to Justice Sector" (2016) 25:1 JL & Soc Pol'y 1.

(5) Protecting Canada's Immigration System Act, SC 2012, c 17.

(6) See Bond, Wiseman & Bates, "Troubling Signs", supra note 3. There have been a number of publications and studies addressing the policy objectives motivating the new system. See e.g. Audrey Macklin, "A Safe Country to Emulate? Canada and the European Refugee" in Helene Lambert, Jane McAdam & Maryellen Fullerton, eds, The Global Reach of European Refugee Law (Cambridge: Cambridge University Press, 2013) 99 at 99; Stephanie J Silverman, "In the Wake of Irregular Arrivals: Changes to the Canadian Immigration Detention System" (2014) 30:2 Refuge 27; Amrita Hari, "Temporariness, Rights, and Citizenship: The Latest Chapter in Canada's Exclusionary Migration and Refugee History" (2014) 30:2 Refuge 35; Samantha Bezic, Economic Heroes, 'Bogus' Asylum Claimants, and Genuine Refugees : A Discourse-Historical Analysis of Recent Changes to Immigration Policy, 2010 to 2012 (MA Thesis, Ryerson University, 2011) [unpublished]. More specifically, attention has been drawn to the problematic justificatory focus on supposedly high numbers of so-called "bogus" claims. See Idil Atak, Graham Hudson & Delphine Nakache, "Making Canada's Refugee System Faster and Fairer: Reviewing the Stated Goals and Unintended Consequences of the 2012 Reform" (2017) Canadian Association for Refugee and Forced Migration Studies Working Paper No 2017/3.

(7) Canada also has a complementary form of protection for "person[s] in need of protection". These are statutorily defined under section 97(1) of the Immigration and Refugee Protection Act and are subject to the same procedures as refugee claimants. See Immigration and Refugee Protection Act, SC 2001, c 27, s 97(1) [IRPA].

(8) Before an individual can have their claim heard by the RPD, they must first be referred to the RPD. This means refugee claimants must first be deemed admissible to Canada in order to be eligible to file a claim with the RPD. If a claim is made at a port of entry to Canada, a CBSA officer at the primary inspection line will conduct an interview to determine whether the claimant is eligible to enter Canada and file a claim. The CBSA officer will either determine whether the individual is admissible or will draft a report outlining the alleged grounds of inadmissibility. This report is then forwarded to the Minister's Delegate who reviews the grounds of inadmissibility to determine the validity of the report. Upon reviewing the report, the minister's delegate may decide to: (1) refer the case to an admissibility hearing; (2) allow the person to leave Canada; (3) issue a temporary resident permit; or (4) issue an exclusion order. Subsection 228(1) of the Immigration and Refugee Protection Regulations outlines the grounds of inadmissibility for which the CBSA officer may directly issue a deportation order. These include inadmissibility on the grounds of criminality or serious criminality under subsection 36(1)(a) or (2)(a) of the IRPA, and subsection 40.1(1) of the IRPA on grounds of the cessation of refugee protection. If a claim is made at an inland immigration office, either a CBSA officer or an Immigration, Refugees and Citizenship Canada (IRCC) officer will carry out the admissibility assessment. If a claim is made to an immigration officer, the IRCC officer must determine whether the claimant meets the eligibility criteria within three working days. If no determination is made, the matter is automatically referred to the RPD for hearing. See Immigration and Refugee Protection Regulations, SOR/2002-227, s 228(1) [IRPR].

(9) See Immigration and Refugee Board of Canada, "Assessment of Credibility in Claims for Refugee Protection" (31 January 2004) at 1.1, online: <irb-cisr.gc.ca/en/legal-policy/legal-concepts/Pages/Credib.aspx>. Because board members have the advantage of hearing the claimant's testimony, they are in the best position to gauge the credibility of the claimant and make any required inferences. See Rahal v Canada (Minister of Citizenship and Immigration), 2012 FC 319 at para 42.

(10) See Jamie Chai Yun Liew & Donald Galloway, Immigration Law, 2nd ed (Toronto: Irwin Law, 2015) at 274-75.

(11) See Singh v Minister of Employment and Immigration, [1985] 1 SCR 177, 17 DLR (4th) 422.

(12) Interview with former IRB decision maker (unnamed) confirmed extensive training: interview by Jennifer Bond (3 March 2017). The authors also have a copy of 1346 pages of IRB training materials obtained via an ATIP request. Almost all of this material is redacted.

(13) See Immigration and Refugee Board of Canada, "National Documentation Packages" (28 June 2019), online: <irb-cisr.gc.ca/en/country-information/ndp/Pages/index.aspx>.

(14) See Sean Rehaag, "The Role of Counsel in Canada's Refugee Determination System: An Empirical Assessment" (2011) 49:1 Osgoode Hall LJ 71 at 86 [Rehaag, "The Role of Counsel"].

(15) Canada is currently facing a long-term legal aid crisis that is impacting access to counsel for refugees. See e.g. Legal Aid Ontario, News Release, "LAO to Consult on Suspensions of Refugee & Immigration Services" (19 May 2017), online: <legalaid.on.ca/en/news/newsarchive/2017-05-23_refugee-services-suspensions-consultation.asp>. In Ontario, this crisis has intensified since the provincial government announced a 30% cut to provincial funding for Ontario Legal Aid and a 100% reduction in funding for refugee services in its April 2019 Budget. See Legal Aid Ontario, News Release, "Update Following Province's April 11th Instruction to Use Only Federal Funding for LAO Refugee and Immigration Law Services" (15 April 2019), online: <legalaid.on.ca/en/news/newsarchive/2019-04-15_refugee-services-lawyer.asp>; Ryan Peck, Executive Director of the HIV & AIDS Legal Clinic Ontario, "Update on Legal Aid Ontario Funding Cuts" (Press Statement) (10 July 2019), online: <halco.org/2019/news/lao-cuts-2019july>. See also Ryan Tumilty, "Legal Aid Funding Cut Nearly 30% in Ontario Budget", CBC News (12 April 2019), online: <cbc.ca/news/canada/ottawa/ontario-legal-aid-funding-cut-1.5095058>. While the Ontario government originally announced that the 30% budget cut to Legal Aid Ontario in 2019 would be the first of a series of planned cuts, on 9 December 2019, Ontario Attorney General Doug Downey announced that the previously-announced reductions would remain in place, but there would not be any further reductions to Legal Aid Ontario's budget in 2020. See Allison Jones, "Ontario Government Cancels Future Legal Aid Funding Cuts, But 2019 Reductions Remain", Global News (9 December 2019), online: <globalnews.ca/news/6273787/legal-aid-ontario-cuts>.

(16) See Rehaag, "The Role of Counsel", supra note 14 at 92.

(17) See Immigration and Refugee Board of Canada, "Interpreter Handbook" (October 2017) at 2.1, 2.2, online (pdf): <humaneimmigration.com/wp-content/uploads/2018/12/05-Interpreter-Handbook-Immigration-and-Refugee-Board-of-Canada.pdf> [Interpreter Handbook]. The Interpreter Handbook notes that in "exceptional circumstances" where the claimant speaks a very rare language or dialect, non-accredited interpreters may be used: ibid. See also Robert F Barsky, "The Interpreter and the Canadian Convention Refugee Hearing: Crossing the Potentially Life-Threatening Boundaries Between 'Coccode-e-h,' 'Cluck-cluck,' and 'Cot-cot-cot'" (1993) 6:2 Traduction, Terminologie, Redaction 131 at 146 (the author notes that when the accreditation test was first introduced in 1991, 40% of interpreters who were already working for the IRB failed).

(18) See Interpreter Handbook, supra note 17 at 3.3.

(19) See ibid at 3.4.

(20) See Refugee Protection Division Rules, SOR/2012-256, s 11 (documentary evidence), s 44 (witness testimony) [RPD Rules]. See also Liew & Galloway, supra note 10 at 263-64, 271.

(21) See RDP Rules, supra note 20, s 32(1). In certain jurisdictions, claimants benefiting from legal aid counsel receive some support for translation costs: in British Columbia, the Legal Services Society of BC allows lawyers providing legal representation to refugee claimants to bill up to 10 hours for interpretation services without prior approval and authorized up to $361 in translation fees for each immigration referral issued. See Legal Services Society BC, "Disbursements Tariff" (May 2017) at 8, online (pdf): <lss.bc.ca/sites/default/files/2019-03/disbursementsMay2017.pdf>. In Ontario, lawyers with RPD certificates from Legal Aid can bill Legal Aid online for translation of up to 3500 words. For documents longer than 3500 words, lawyers can submit a request for additional disbursements for translation. See Legal Aid Ontario, News Release, "New Translation Disbursement for Refugee Certificates" (29 May 2015), online: <legalaid.on.ca/en/news/newsarchive/1505-29_translationdisbursement.asp>.

(22) See RPD Rules, supra note 20, ss 26(1), 26(2), 27(1), 27(2), 28(1) (addressing exclusion, integrity issues, and criminality issues that may affect the claim), s 29(1) (setting out the notice requirements for ministerial intervention in a claim).

(23) The board member is required to notify the minister if, before a hearing, they believe there is a possibility of exclusion under Article 1E or F of the Refugee Convention. See RPD Rules, supra note 20, s 26(1). The board member must also notify the minister of potential threats to program integrity, if they have reason to believe that the minister's participation in the hearing may "help in the full and proper hearing of the claim": RPD Rules, supra note 20, s 27(1).

(24) See RPD Rules, supra note 20, s 29(1).

(25) See ibid, ss 29(2)(a)-(b).

(26) On 17 May 2019, the Government of Canada removed all countries from the DCO list. IRCC stated that this decision "effectively suspends the DCO policy, introduced in 2012, until it can be repealed through future legislative changes": Immigration, Refugees and Citizenship Canada, News Release, "Canada Ends the Designated Country of Origin Practice" (17 May 2019), online: <canada.ca/en/immigration-refugees-citizenship/news/2019/05/canada-ends-the-designated-country-of-origin-practice.html>. However, as of time of writing, there has been no legislative amendment to repeal s 109.1(1) of the IRPA, which stipulates that the minister may designate certain countries of origin and limit appeal rights to claimants from these countries (s 110(2)) and, through the IRPR, designate different timeframes for claimants from these countries for document provision, hearings and timeframes for filing and perfecting an appeal (s 111.1). See IRPA, supra note 7 at s 109.1. This means that the minister could designate new "countries of origin" at a future date, and the present legislative scheme (including differentiated timeframes) would apply to claimants from those countries.

(27) See IRPR, supra note 8, ss 159.8(2).

(28) See ibid, ss 159.8(2), 159.9(1). However, in early 2018, the IRB announced it had changed its policy on hearing scheduling and will no longer follow the scheduled timelines. To address the backlog of claims, the IRB began addressing claims in the order they are made. See Media Relations of the Immigration and Refugee Board, "The Immigration and Refugee Board of Canada Changes How Refugee Hearings Are Scheduled" (Press Release) (20 February 2018), online: <https://irbcisr.gc.ca/en/news/2018/Pages/hearing-schedule.aspx>. We discuss this in further detail at note 75, below.

(29) See RPD Rules, supra note 20, s 34(3).

(30) Under section 110(2) of the IRPA, claimants may not appeal RPD decisions to the RAD if: (a) they are designated foreign nationals; (b) their claims for protection have been withdrawn or abandoned; (c) their claims were rejected for having no credible basis or for being manifestly unfounded; (d) their claims were heard as an exception to the Safe Third Country Agreement, or they are from designated countries of origin (DCOs); (e) the RPD's decision caused the cessation of their refugee status; or (f) the RPD's decision caused the vacation of their refugee status. See IRPA, supra note 6, s 110(2). Under section 105 of the IRPA, appeals to the RAD are also restricted if the claimant is subject to an order of surrender under the Extradition Act, SC 1999, c 18. See ibid, s 105. In addition, "[t]he RAD is available only to claimants whose original claim was referred to the IRB after 15 December 2012. Those whose claims were determined under the old refugee determination process and have been ordered back to the RPD after a favourable Federal Court ruling do not have access to the RAD after their second refugee determination decision": Library of Parliament, Refugee Protection in Canada, by Julie Bechard & Sandra Elgersma, Publication No 2011-90-E (Ottawa: Library of Parliament, 2013) at 3.7.2. Since the RAD came into force, the bar to DCO claimants has been successfully challenged in the Federal Court. In YZ v Canada, 2015 FC 892 at para 130 [YZ], the Court found that denying access to the RAD to claimants from DCOs violates their equality rights under section 15 of the Canadian Charter of Rights and Freedoms, Part 1 of the Constitution Act, 1982, being Schedule B to the Canada Act 1982 (UK), 1982, c 11 [Charter]. The Federal Government discontinued its appeal of YZ in 2015. In Kreishan v Canada, 2018 FC 481 [Kreishan], the Federal Court upheld the constitutionality of section 110(2)(d)(i) of the IRPA, which bars appeals to claims heard as an exception to the Safe Third Country Agreement. The Federal Court then referred the following Certified Question to the Federal Court of Appeal (FCA): "Does paragraph 110(2)(d) of the Immigration and Refugee Protection Act infringe section 7 of the Canadian Charter of Rights and Freedoms and, if so, is this infringement justified by section 1?". The FCA answered the first question in the negative, with the Court deciding that paragraph 110(2)(d) of the IRPA did not infringe section 7 of the Charter. See R v Kreishan, 2019 FCA 223, at paras 142-44 [citations omitted].

(31) See Federal Court of Canada, "Statistics (December 31, 2017): Activity Summary--January 1 to December 31, 2017", online: <www.fct-cf.gc.ca/en/pages/about-the-court/reports-and-statistics/statistics-december-31-2017>. These figures suggest a rough approval rate of approximately 22%. This does not, however, account for cases that may not have proceeded for a variety of reasons, or the fact that some cases granted leave in 2017 were commenced in 2016, while others commenced in 2017 were not heard until 2018. Nonetheless, the figures provide some general sense of the frequency of successful leave applications, relative to overall Federal Court activity.

(32) Note that our use of the term discretion extends beyond the formal usage that is familiar in broader Administrative Law. Our usage is discussed further in Part IV.

(33) See especially Bond, Wiseman & Bates, "Troubling Signs", supra note 3.

(34) See Roderick A Macdonald, "Access to Justice in Canada Today: Scope, Scale and Ambitions" in Julia Bass, WA Bogart & Frederick H Zemans, eds, Access to Justice for a New Century--The Way Forward (Toronto: Law Society of Upper Canada, 2005) 19.

(35) Indeed, social institutions and processes play a multi-faceted role, in the sense that they can contribute both to the construction and deconstruction of the tendency to infer positive, negative, or neutral experiences of different social context factors. See Donna Hackett & Richard Devlin, "Constitutionalized Law Reform: Equality Rights and Social Context Education for Judges" (2005) 4:2 JL & Equality 157; Nancy J Burke et al, "Theorizing Social Context: Rethinking Behavioral Theory" (2009) 36:5 Health Education & Behavior 55S; T Brettel Dawson, "Judicial Education on Social Context and Gender in Canada: Principles, Process and Lessons Learned" (2014) 21:3 Intl J Leg Profession 259 for more on the development of the social context approach.

(36) See Macdonald, supra note 34. Macdonald refers to the need for access to justice strategies that are multi-dimensional and legally-pluralistic. Strategies should be multidimensional in the sense that different people are differently situated in ways that may require specialized and tailored strategies for improving their access to justice. Macdonald identifies four key dimensions of difference in this context: geography (e.g. urban or rural or remote), socio-demographics (e.g. "gender, age, employment status and education"), social disempowerment (e.g. Aboriginal peoples), and problem-perception (i.e. diversity of perceptions in terms of experiences of problems and self-assessment of solution-needs). Strategies should be legally-pluralistic in the sense that a variety of sites of law, including formal and informal processes and public and private institutions, need to be steered towards access to justice. See ibid at 24-25. Macdonald also refers to a range of barriers to justice: physical and material, objective, subjective, sociological and psychological. See ibid at 26-29.

(37) In including the contested identity marker of "race" as one social context factor relevant to refugee processes, we acknowledge the fraught history of this term, as well as the similarly fraught histories of other identity markers listed. With respect to the term race specifically (and also illustrating our approach to such identity markers more generally), we note that we understand race as a socially constructed identity, rather than an essentialist description of something fixed, objective, and measurable. In this regard, sociologists Michael Omi and Howard Winant describe race as "a concept which signifies and symbolizes social conflicts and interests by referring to different types of human bodies", and suggest that effort must be made to "understand race as an unstable and decentered complex of social meanings constantly being transformed by political struggle": Michael Omi & Howard Winant, R acial Formation in the United States From the 1960s to the 1990s, 2nd ed (New York: Routledge, 1994) at 55. This description of race as a social identity that is contested and constructed by social power relations (among other things) is particularly apt when considering discussions of race and racial identity that arise in refugee determinations, where claiming or denying allegiance to particular family unit, social life, or cultural background can be central to a finding--or denial--of protection. See e.g. Aliya Saperstein, Andrew M Penner & Ryan Light, "Racial Formation in Perspective: Connecting Individuals, Institutions and Power Relations" (2013) 39:1 Annual Rev of Sociology 359 (for the development of social constructivist theory within sociology and tensions in the field as to the definitions of race and ethnicity); Kitty Calavita, "Immigration Law, Race and Identity" (2007) 3 Annual Rev of L & Soc Science 1 (for the significance of social constructs of race in immigration law in particular).

(38) See Bond, Wiseman & Bates, "Troubling Signs", supra note 3 at 11-14 (for further discussion of general and specific social contexts that are relevant to refugee claimants).

(39) This is also consistent with works examining the impact of the new system on particular populations. See Cynthia Levine-Rasky, Julianna Beaudoin & Paul St Clair, "The Exclusion of Roma Claimants in Canadian Refugee Policy" (2014) 48:1 Patterns of Prejudice 67; Judit Toth, "Czech and Hungarian Roma Exodus to Canada: How to Distinguish Between Unbearable Destitution and Unbearable Persecution" in Didier Bigo, Elspeth Guild & Sergio Carrera, eds, Foreigners, Refugees or Minorities? Rethinking People in the Context of Border Controls and Visas, 1st ed (Surrey, UK: Ashgate, 2013) 39; Marina Caparini, "State Protection of the Czech Roma and the Canadian Refugee System" in Bigo, Guild & Carrera, supra note 39, at 131; Julianna Beaudoin, Sean Rehaag & Jennifer Danch, "No Refuge: Hungarian Romani Refugee Claimants in Canada" (2015) 52:3 Osgoode Hall LJ 705; Evangelia Tastsoglou et al, "(En)Gendering Vulnerability: Immigrant Service Providers' Perceptions of Needs, Policies, and Practices Related to Gender and Women Refugee Claimants in Atlantic Canada" (2014) 30:2 Refuge 67. For work on particular types of claims, see Lobat Sadrehashemi, "Gender Persecution and Refugee Law Reform in Canada: Legislative Comment on The Balanced Refugee Reform Act (Bill C-11)" (2011), online (pdf): Battered Women's Support Services <bwss.org/wp-content/uploads/GENDER-PERSECUTION-and-REFUGEE-LAW-REFORM-IN-CANADA_2.pdf>; Rohan Sajnani, "Envisioning LGBT Refugee Rights in Canada: The Impact of Canada's New Immigration Regime" (June 2014), online (pdf): Affiliation of Multicultural Societies and Service Agencies <amssa.org/wp-content/uploads/2015/06/Report-Immigration-Regime-Jan-20151.pdf>; Nicole LaViolette, "Sexual Orientation, Gender Identity and the Refugee Determination Process in Canada" (2014) 4:2 J Research in Gender Studies 68. For work on particular discourses, see Emily Field, Whose Safety Matters? Exaltation, Risky Refugees, and Canadian Safe Country Practices (Master of Arts Thesis, University of Ottawa, Institute of Women's Studies, 2013) [unpublished]; Rivera Perez, Redefining Refugees in Canada: Comparing Policy Frames for Refugee Health Care Policy in Canada and United States (Master of Arts Major Research Paper, University of Ottawa, 2014) [unpublished]; Basia D Ellis, "The Production of Irregular Migration in Canada" (2015) 47:2 Can Ethnic Studies 93.

(40) Cecile Rousseau et al, "The Complexity of Determining Refugeehood: A Multidisciplinary Analysis of the Decision-Making Process of the Canadian Immigration and Refugee Board" (2002) 15:1 J Refugee Studies 43.

(41) Ibid at 57-60 (regarding the difficulties Board Members experienced in assessing the complex psychological context of refugee claims), 60-64 (regarding the numerous types of cultural misunderstanding or miscommunication that were found in those RPD cases evaluated in the study).

(42) See e.g. Nicole LaViolette, "Independent Human Rights Documentation and Sexual Minorities: an Ongoing Challenge for the Canadian Refugee Determination Process" (2009) 13:2-3 Intl JHR 437 [LaViolette, "Documentation"]; Catherine Dauvergne & Jenni Millbank, "Forced Marriage as a Harm in Domestic and International Law" (2010) 73:1 Mod L Rev 57; Sean Rehaag, "Patrolling the Borders of Sexual Orientation: Bisexual Refugee Claims in Canada" (2008) 53:1 McGill LJ 59. It is noteworthy that the IRB has developed targeted guidelines and training for board members that acknowledge the need to consider these particular social context areas. See Immigration and Refugee Board of Canada, Chairperson's Guideline 9: Proceedings Before the IRB Involving Sexual Orientation and Gender Identity and Expression (Guideline) (Ottawa: Immigration and Refugee Board of Canada, 2017), online: <irbcisr.gc.ca/en/legal-policy/policies/Pages/GuideDir09.aspx> [IRB Guideline 9]; Immigration and Refugee Board of Canada, Chairperson Guidelines 4: Women Refugee Cl aimants Fearing Gender-Related Persecution (Guideline) (Ottawa: Immigration and Refugee Board of Canada, 1996), online: <irb-cisr.gc.ca/en/legal-policy/policies/Pages/GuideDir04.aspx>.

(43) Sherene Razack, "What Is to Be Gained by Looking White People in the Eye? Culture, Race, and Gender in Cases of Sexual Violence" (1994) 19:4 Feminism & Law 894 at 897-98.

(44) Ibid. See also Sherene Razack, "'Simple Logic': Race, the Identity Documents Rule and the Story of a Nation Besieged and Betrayed" (2000) 15 JL & Soc Pol'y 181; Sherene Razack, R ace, Space and the Law: Unmapping a White Settler Society (Toronto: Between the Lines, 2002).

(45) See Rehaag, "The Role of Counsel", supra note 14 (on counsel).

(46) See Tess Acton, Understanding Refugee Stories: Lawyers, Interpreters, and Refugee Claims in Canada (Master of Law Thesis, University of Victoria, 2015) [unpublished] (on interpretation).

(47) See Hilary Evans Cameron, "Refugee Status Determinations and the Limits of Memory" (2010) 22:4 Intl J Refugee L 469 [Cameron, "Limits of Memory"] (on memory).

(48) A small but growing number of studies, beyond those already cited above, address specific aspects of the modified system, including: (a) limitations on appeal rights (see Angus Grant & Sean Rehaag, "Unappealing: An Assessment of the Limits on Appeal Rights in Canada's New Refugee Determination System" (2016) 49:1 UBC L Rev 203); (b) quantitative analysis of access to counsel and success rates (see Sean Rehaag, "2018 Refugee Claim Data and IRB Member Recognition Rates" (19 June 2019), online: <https://ccrweb.ca/en/2018-refugee-claim-data>); (c) accessibility and the cost of refugee healthcare (see Cecile Rousseau et al, "Encouraging Understanding or Increasing Prejudices: A Cross-Sectional Survey of Institutional Influence on Health Personnel Attitudes about Refugee Claimants' Access to Health Care" (2017) 12:2 PLOS ONE, online: <https://doi.org/10.1371/journal.pone.0170910>); and (d) the role and availability of legal counsel (see Lobat Sadrehashemi, Peter Edelmann & Suzanne Baustad, "Refugee Reform and Access to Counsel in British Columbia" (26 August 2015), online: British Columbia Public Interest Advocacy Centre <bcpiac.com/report-refugee-reform-access-to-counsel-in-british-columbia>). Other studies of the pre-reform system have focused on specific elements, including access to counsel. See Michael Barutciski, "The Impact of the Lack of Legal Representation in the Canadian Asylum Process: Report Researched and Written for the UNHCR" (6 November 2012), online (pdf): The UN Refugee Agency <unhcr.ca/wpcontent/uploads/2014/10/RPT-2012-06-legal_representation-e.pdf>. See Martin Jones, "Abandoning Refugees? An Analysis of the Legal Framework Governing Non-Compliant Claimants in Canada" (2008) 25:2 Refuge 132 for access to recourse for failed claims. Studies have also identified factors affecting: (a) decision maker independence and bias (see Jacqueline Bonisteel, "Ministerial Influence at the Canadian Immigration and Refugee Board: the Case for Institutional Bias" (2010) 27:1 Refuge 103; Sean Rehaag, "2012 Refugee Claim Date and IRB Member Recognition Rates" (13 May 2013), online: Canadian Council for Refugees <ccrweb.ca/en/2012-refugee-claim-data>; Innessa Colaiacovo, "Not Just the Facts: Adjudicator Bias and Decisions of the Immigration and Refugee Board of Canada (2006-2011)" (2013) 1:4 J Migration & Human Security 122); (b) procedural safeguards for vulnerable claimants (see Janet Cleveland, "The Guideline on Procedures with Respect to Vulnerable Persons Appearing Before the Immigration and Refugee Board of Canada: A Critical Overview" (2008) 25:2 Refuge 119); (c) the role of expert psychological evidence in refugee hearings (see Janet Cleveland & Monica Ruiz-Casares, "Clinical Assessment of Asylum Seekers: Balancing Human Rights Protection, Patient Well-Being, and Professional Integrity" (2013) 13:7 American J Bioethics 13); and (d) the role of international human rights law in deciding refugee claims (see Catherine Dauvergne, "How the Charter Has Failed Non-citizens in Canada: Reviewing Thirty Years of Supreme Court of Canada Jurisprudence" (2013) 58:3 McGill LJ 663). See e.g. United Nations High Commissioner for Refugees (UNHCR), "UNHCR Statement Relating to Bill C-31, Protecting Canada's Immigration System Act: Senate Standing Committee on Social Affairs, Science and Technology" (18 June 2012), online (pdf): <unhcr.ca/wpcontent/uploads/2014/10/Canada_Bill_C-31_Oral_Presentation.pdf>; Naomi Alboim & Kareen Cohl, "Shaping the Future: Canada's Rapidly Changing Immigration Policies" (October 2012), online (pdf): Maytree Foundation <maytree.com/wpcontent/uploads/shaping-the-future.pdf> (for non-academic reports and comments).

(49) See IRPA, supra note 7, ss 3(2)(a)-(b), (e), (g) (listing the objectives as being: (a) to recognize that the refugee program is in the first instance about saving lives and offering protection to the displaced and persecuted; (b) to fulfil Canada's international legal obligations with respect to refugees and affirm Canada's commitment to international efforts to provide assistance for those in need of resettlement; (e) to establish fair and efficient procedures that will maintain the integrity of the Canadian refugee protection system, while upholding Canada's respect for the human rights and fundamental freedoms of all human beings; (g) to protect the health and safety of Canadians and to maintain the security of Canadian society).

(50) See Bond, Wiseman & Bates, "Troubling Signs", supra note 3 at 67-68 (for an in-depth explanation of these methods).

(51) This research was conducted in compliance with Tri-Council ethics requirements. See Jennifer Bond and David Wiseman, "Access to Justice in Canada's Refugee System: the Use of Evidence in the Refugee Status Determination Process", Application to the Social Sciences and Humanities Research Council (Insight Development Program) (6 February 2014) at 3.

(52) In some cases, counsel providing the case did not represent the claimant at the original RPD hearing, but rather was retained for the purposes of an appeal or re-hearing.

(53) Twenty-seven (67.5%) of the claimants were between the ages of 26 and 45, and 15 (37.5%) were between the ages of 26 and 35.

(54) The term "Palestine/Jordan" is used to reflect the complexities of determining the nationality (and citizenship rights) of refugee claimants of Palestinian origin who have lived in Jordan or have arrived via Jordan. The Embassy of Canada to Jordan has

provided the following information to the Research Directorate of the IRB on this issue: "Jordan issues different types of passport that gives its holders controlled access to different government services, including citizenship and right of re-entry. The categories differ depending on the origin of the citizenship which is linked to refugee movements from Palestine": Research Directorate of the IRB, "Response to Information Request (Report on Country Conditions), Jordan/Palestine: Whether A Person of Palestinian Origin Who Lives in Jordan Can Return to Palestine and Obtain a Palestinian Passport" (21 December 2010), online: <https://irb-cisr.gc.ca/en/country-information/rir/Pages/index.aspx?doc=453294>.

(55) For reference, the top five countries of origin for asylum seekers to Canada during the relevant period were: China, Nigeria, Pakistan, Somalia, and Syria. See Sean Rehaag, "2015 Refugee Claim Data and IRB Recognition Rates" (30 March 2016), online: Canadian Council for Refugees <ccrweb.ca/en/2015-refugee-claim-data>.

(56) In 67.5% (27) of cases, the harm feared was based on membership in a particular social group, followed by political opinion (50% [20]), religion (20% [8]), race (10% [4]), and nationality (2.5% [1]).

(57) These acceptance rates are lower than the overall Canadian average, where 59% of all refugee claims at the RPD in 2016 received a positive decision, and 26.6% of all claimant appeals to the RAD were allowed. See Sean Rehaag, "2016 Refugee Claim Data and IRB Member Recognition Rates" (8 March 2017), online: Canadian Council for Refugees <ccrweb.ca/en/2016-refugee-claim-data>.

(58) RPD Rules, supra note 20, s 11.

(59) See Orelien v Canada (Minister of Employment and Immigration), [1992] 1 FC 592 at paras 17-21, [1991] FCJ No 1158 (FCA).

(60) See Maldonaldo v Canada (Minister of Employment and Immigration) (1979), [1980] 2 FC 302 at para 5, [1979] FCJ No 248 (FCA); Chan v Canada (Minister of Employment and Immigration), [1995] 3 SCR 593 at para 142, [1995] SCJ No 78; Sedigheh v Canada (Minister of Citizenship and Immigration), 2003 FCT 147 at paras 54-55.

(61) See Ansong v Canada (Ministry of Employment and Immigration), [1989] FCJ No 728, 9 Imm LR (2d) 94 (FCA). See Cameron, "Limits of Memory", supra note 47 (for a critique of basing credibility findings in refugee cases on memory).

(62) See Wen v Canada (Minister of Employment and Immigration), [1994] FCJ No 907, [1994] ACF No 907 (FCA). See also Michael Kagan, "Is Truth in the Eye of the Beholder? Objective Credibility Assessment in Refugee Status Determination" (2003) 17:3 Geo Immigr LJ 367 (for a critique of the use of demeanor in the refugee claim context).

(63) See Armson v Canada (Minister of Employment and Immigration), [1989] FCJ No 800, 101 NR 372 [1989] (FCA); Leung v Canada (Minister of Employment and Immigration) (1990), 74 DLR (4th) 313, [1990] FCJ No 908 (FCA).

(64) See Gracielome v Canada (Minister of Employment and Immigration), [1989] FCJ No 463, 9 Imm LR (2d) 237 (FCA); Virk v Canada (Minister of Employment and Immigration), [1992] FCJ No 199, 140 NR 290 (FCA); Rajaratnam v Canada (Minister of Employment and Immigration), [1991] FCJ No 1271, 135 NR 300 (FCA).

(65) See IRPA, supra note 7, s 106. Failure to prove identity is fatal to a claim. See e.g. Zheng v Canada (Citizenship and Immigration), 2008 FC 877 at para 15.

(66) It is up to the IRB to determine the credibility of such testimony. See Bula v Canada (Secretary of State), 1994 FCJ No 937 at para 6, 1994 CarswellNat 2726 (FC). See also Rebecca MM Wallace & Karen Wylie, "The Reception of Expert Medical Evidence in Refugee Status Determination" (2013) 25:4 Intl J Refugee L 749.

(67) The IRB keeps a documents package for each country, which are accessible online at: <www.irb-cisr.gc.ca/Eng/ResRec/NdpCnd/Pages/ndpcnd.aspx>. However, in some circumstances this information can be non-existent or difficult to find. See e.g. LaViolette, "Documentation", supra note 42.

(68) See Tung v Canada (Minister of Employment and Immigration), [1991] FCJ No 292, 124 NR 388 (FCA).

(69) See Cepeda-Gutierrez v Canada (Minister of Citizenship and Immigration), 157 FTR 35, [1998] FCJ No 1425 (FC). See also France Houle, "Le Fonctionnement du Regime de Preuve Libre dans un Systeme Non-Expert: Le Traitement Symptomatique des Preuves par la Section de la Protection des Refugies" (2004) 38:2 RJT 263 (for a general critique of the IRB evidentiary assessment).

(70) See IRPA, supra note 7, s 170(g).

(71) For research on the pre-reform refugee system identifying issues relating to evidentiary challenges, see Cameron, "Limits of Memory", supra note 47; LaViolette, "Documentation", supra note 42; Francois Crepeau & Delphine Nakache, "Critical Spaces in the Canadian Refugee Determination System: 1989-2002" (2008) 20:1 Intl J Refugee L 50. The crucial role that a lack of adequate evidence plays as a basis for negative credibility findings has been analyzed in Hilary Evans Cameron, "Risk Theory and 'Subjective Fear': The Role of Risk Perception, Assessment, and Management in Refugee Status Determinations" (2008) 20:4 Intl J Refugee L 567.

(72) See Hilary Evans Cameron, Refugee Law's Fact-Finding Crisis: Truth, Risk, and the Wrong Mistake (Cambridge: Cambridge University Press, 2018) [Cameron, Fact-Finding Crisis]. To support her conclusion, Cameron examines a number of refugee law cases before the Federal Court of Canada, and finds that the Court disagrees with itself on almost every aspect of how fact-finding operates in practice, and "vacillates between preferring to err in the claimant's favour and preferring to err against her, and this lack of a stable preference is reflected in a law of fact-finding that tries to favour both types of mistake at once. As a result, Canadian refugee status decision makers are often able to justify either conclusion on the same evidence": ibid at 24.

(73) Responses to Information Requests (RIR) are reports requested by IRB decision makers on country conditions. The database contains an archive of seven years of RIRs. Individuals can search using key words in a full-text search, by title of the report, country, language, or date. See "Country of Origin Information" (last modified 3 July 2018), online: Immigration and Refugee Board of Canada <irb-cisr.gc.ca/Eng/ResRec/RirRdi/Pages/index.aspx>.

(74) Four of these cases involved a formal ministerial intervention.

(75) In another, similar case, the claimant's evidence was that he and his family approached police when their land was illegally seized by a local militia leader, but that the perpetrators' far-reaching influence, combined with police corruption, meant that no investigation was pursued, and no reports of the event were generated. Again, this made it impossible for him to produce the materials being sought.

(76) The difficulty for claimants to provide the evidentiary materials sought by the IRB prompted the IRB to develop Guidelines to promote greater understanding of cases involving sexual orientation and gender identity and expression (SOGIE) and to help guide these proceedings. The Guidelines became effective on 1 May 2017. They apply to all four divisions of the IRB: the Immigration Division (ID), the Immigration Appeal Division (IAD), the Refugee Protection Division (RPD), and the Refugee Appeal Division (RAD). See IRB Guideline 9, supra note 42.

(77) In one case, the ministerial intervention did not occur until an appeal was lodged at the Federal Court.

(78) Assumptions of this nature have been identified as being particularly prominent in relation to claimants who fear gender-based violence (i.e. violence against women) and sexual-orientation violence (e.g. violence against people of gay or lesbian sexual orientation). See Nicole LaViolette, "Coming out to Canada: The Immigration of Same-Sex Couples Under the Immigration and Refugee Protection Act" (2004) 49:4 McGill LJ 969; Efrat Arbel, Catherine Dauvergne & Jenni Millbank, "Introduction: Gender in Refugee Law: From the Margins to the Centre" in Efrat Arbel, Catherine Dauvergne & Jenni Millbank, eds, Gender in Refugee Law: From the Margins to the Centre (London: Routledge, 2014).

(79) Some of these cases overlap with the previous 16 cases where the claimant requested to submit additional evidence after the document submission deadline but before the hearing.

(80) In 2012, the government introduced legislated timelines for hearings into the IRPR, which resulted in new cases being resolved before older claims in certain circumstances. This created a significant backlog, particularly for "legacy cases" which were consistently being deprioritized to accommodate newer cases which needed to be heard in a timely manner in order to meet the IRPR's requirements. In addition, the number of asylum claims has been steadily rising since 2014. On these two points, see "Report of the Independent Review of the Immigration and Refugee Board: A Systems Management Approach to Asylum" (10 April 2018) at 10-11, 13-14, online: <canada.ca/content/dam/ircc/migration/ircc/english/pdf/pub/irb-report-en.pdf>. As a result, the IRB announced in 2018 that it would implement new scheduling practices and schedule claims primarily in the order they were received. The IRB stated that the expected wait time for status determination under the new schedule was expected to be approximately 20 months. See Media Relations of the Immigration and Refugee Board, supra note 28. The Regulations contemplate an exception to the timelines as a result of operational limits under section 159.9(3)(c) of the IRPR, thus giving the IRB authority to implement these changes without further legislative amendment. See IRPR, supra note 8, s 159.9(3)(c).

(81) Two of those rejections were then overturned on appeal.

(82) Robert Gibb & Anthony Good, "Interpretation, Translation and Intercultural Communication in Refugee Status Determination Procedures in the UK and France" (2014) 14:3 Language & Intercultural Communication 385 at 389, citing Joan Colin & Ruth Morris, Interpreters and the Legal Process (Winchester: Waterside Press, 1996) at 17.

(83) Gibb & Good, supra note 82 at 396.

(84) Between July 2013 and December 2014, Nick Gill, Rebecca Rotter, Andrew Burridge, Jennifer Allsop, and Melanie Griffiths (University of Exeter) attended 390 asylum appeals to assess the (dis)advantages of the interpreter-mediated communication in the context of the asylum determination process in the United Kingdom. They acknowledge the serious risks of holding an asylum determination process in a language which the appellant does not speak, noting that these "pitfalls largely arise because interpretation, done well, is not simply a matter of mechanically matching words in one language with words in another." Nonetheless, they conclude that nuanced and sensitive interpretation services provide "forms of resilience and solace for the appellant in their encounters with a hostile, monolithic system. The services offered by professional, qualified, ethical and appropriately remunerated and resourced interpreters constitute modest consolations for the structural injustices of border control": Nick Gill et al, "Linguistic Incomprehension in British Asylum Appeal Hearings" (2016) 32:2 Anthropology Today 18 at 18, 21.

(85) This second aspect of decision maker discretion is an equal part of the day-to-day operation of the refugee system and interacts intimately with the formal variant. It is, however, less explicitly addressed in administrative law, and sits uneasily at the boundary between the tasks of finding facts and applying law. Other authors that we refer to and reference in this section use different terms to refer to a similar aspect of the system. For instance, Audrey Macklin refers to the difficult "subjective nature" of a decision maker's task of drawing inferences from evidence, concluding that it is ultimately about "making choices... in the face of empirical uncertainty": Audrey Macklin, "Truth and Consequences: Credibility in Refugee Determination" in International Association of Refugee Law Judges, eds, Realities of Refugee Determination on the Eve of a New Millennium (Haarlem: International Association of Refugee Law Judges, 1999) at 140.

(86) Ibid. Macklin suggests that, in recognition of these biases, decision makers must ask difficult inward questions of themselves when weighing a credibility question, such as "[d]o I mistrust those around me... Do I feel like I have experienced and overcome oppression in my own life? Does that make me identify with the victim, or harden my attitude towards those who have not overcome adversity the way I have?": ibid at 140.

(87) Cameron argues that, while there is an assumption that the law of fact-finding is a neutral process, in reality, adjudicators effectively "choose" to view refugee claimants either as vulnerable litigants who risk grave harm if their claim is denied, or as ordinary claimants who try to benefit from manipulating the system: Cameron, Fact-Finding Crisis, supra note 72 at 166-67. In a series of illuminating articles that rely on statistical analysis, Rehaag identifies a luck of the draw feature in refugee decision making, with the identity of the decision maker being a key factor in the likely outcome of the claim. See Sean Rehaag, "Judicial Review of Refugee Determinations: The Luck of the Draw?" (2012) 38:1 Queen's LJ 1; Sean Rehaag, "I Simply Do Not Believe...: A Case Study of Credibility Determinations in Canadian Refugee Adjudication" (2017) 38 Windsor Rev Leg Soc Issues 38 [Rehaag, "I Simply Do Not Believe"]. In "I Simply Do Not Believe", Rehaag argues that IRB decision makers are charged with determining when refugee complainants are telling the truth, while in reality "[t]here is simply no way to do this. That is, there are no procedures, no instructions, and no formulas that would allow decision makers... to reliably detect truth and falsehood": ibid at 67.

(88) Cameron explains that all decision-making systems are guided by underlying normative frameworks that are related to "error preference": Cameron, Fact-Finding Crisis, supra note 72 at 16. In the context of refugee law, such a guiding normative framework must be premised on an identification of which "mistake" is worse: granting a claim that should have been denied or denying a claim that should have been granted. She argues that, unlike criminal law or civil law which are clearly anchored by norms that have assessed risk, refugee decision makers currently have the option of choosing one approach over the other, and are therefore free to "err as they choose"--a feature which exacerbates inconsistent decision making: ibid at 167.

Article Details

Printer friendly Cite/link Email Feedback
Title Annotation:	ARTICLES
Author:	Bond, Jennifer; Wiseman, David
Publication:	University of British Columbia Law Review
Geographic Code:	1CONT
Date:	Jan 1, 2020
Words:	19795
Previous Article:	THE DEIFICATION OF PROCESS IN CANADA'S DUTY TO CONSULT: TSLEIL-WAUTUTH NATION V CANADA (ATTORNEY GENERAL).
Next Article:	THINLY CONSTRUING THE NATURE OF THE ACT LEGALLY CONSENTED TO: THE CORROSIVE IMPACT OF R V HUTCHINSON ON THE LAW OF CONSENT.
Topics:	Refugees