Abstract
To secure protection in the global North, asylum-seekers must overcome restrictive government policies and present a convincing refugee narrative. Their credibility becomes their main asset and must survive the multiple challenges arising from intercultural communication and interactions involving multiple institutional actors. Aiming to explore the impact institutional understandings of refugee narrative creation have on credibility assessment, I present the findings of an analysis of a corpus of documents from the Australian tribunal responsible for the administrative review of asylum decisions. I critically analyse these texts to identify how the tribunal and its agents discursively present the various actors involved in asylum appeals. I argue that despite the cautions of existing scholarship, these texts present the asylum-seeker as the sole author of the final refugee narrative, regardless of the role that decision-makers and other actors, such as lawyers and interpreters, play in its co-construction. Thus, the institution places disproportionate responsibility on the asylum-seeker for communication outcomes, creating significant challenges for their credibility.
Keywords
Introduction
The small portion of the world’s displaced people who reach the global North to seek asylum must contend with increasingly exclusionary government policies (Crock and Ghezelbash, 2010; Maryns, 2004: 243). Australian policy is no exception, with suspicion of those who seek protection underlying public and institutional discourse. Indeed, this has led to an almost complete ban on asylum applications from those arriving by boat without authorisation (Department of Immigration and Border Protection, 2016). Those who arrive in Australia by regular means with a visa are allowed to seek asylum but still face challenges. Given that they often lack documentary evidence of the persecution they claim to fear, developing a credible refugee narrative becomes a crucial step in the quest for recognition as a refugee (Crock and Martin, 2013; Jacquemet, 2011: 478).
The 1951 United Nations Convention Relating to the Status of Refugees (‘Refugee Convention’) is the central instrument governing the rights of refugees, defining them as anyone who
owing to well-founded fear of being persecuted for reasons of race, religion, nationality, membership of a particular social group or political opinion, is outside the country of his nationality and is unable or, owing to such fear, is unwilling to avail himself of the protection of that country. (Article 1A(2))
The Convention allows states to create their own assessment procedures, and credibility has become an important element of these in many receiving countries (Kagan, 2003). However, research identifies many challenges to how credibility is assessed. The purportedly objective credibility indicators create unrealistic expectations of the way credible applicants communicate. These expectations partly stem from language ideologies – beliefs about the nature of language.
For example, McKinnon (2009) explains how credible asylum-seekers are expected to tell ‘logical, linear, truthful stories’ (p. 213). Yet various psychological factors mean it is rare for people to recount stories consistently over time (Cohen, 2001). Cultural and individual differences mean that they may remember details differently from institutional expectations (Evans Cameron, 2008; Jacquemet, 2011).
While asylum-seekers may be held responsible for the final refugee narrative (and thus its credibility), in reality, other participants – such as the decision-maker, lawyers and interpreters – all play a role in its construction. Decision-makers control questioning and create the final record (Blommaert, 2001; Jacquemet, 2011; Maryns, 2006). Lawyers help draft statements and influence their clients’ communication-related decisions, such as deciding whether to use an interpreter. Whereas interpreters may be considered to simply reproduce a message from one language to another, they must choose between sometimes competing translation norms (see Inghilleri, 2003; Tipton, 2008; Van der Kleij, 2015). Their attitudes and personal attributes may also influence the asylum-seeker’s communication (as discussed by Merlini, 2009; Pöllabauer, 2004).
Therefore, I seek to uncover the institutional discourse that shapes credibility assessments. Focusing on administrative appeals, I analyse the key Australian guidelines for assessing credibility, along with a corpus of recent published decisions dealing with credibility concerns. Guided by Van Leeuwen’s (1996) ‘sociosemantic’ model for Critical Discourse Analysis, I consider how these texts represent the various actors involved in asylum processes and explore the ideologies underpinning these representations.
In the following I first provide an overview of the Australian asylum system, the research data set and analytical approach. I then present my analysis of the key credibility guidelines and the corpus of published appeal decisions, respectively. This article concludes with a critical discussion of the impact of institutional discourse on credibility assessment, identifying ongoing challenges, in light of the existing literature. I conclude by suggesting that highlighting the multi-party interactional construction of the refugee narrative may constitute a challenge to the official discourse and indeed to the claim that credibility is assessed objectively.
Australian refugee reviews and policy guidance on credibility as discourse on seeking asylum
In this section I introduce the Australian asylum process and the research data. I then explain the analytical approach I adopt to uncover the institutional discursive construction of the key participants involved in these appeals and the resulting challenges.
The institutional setting and the data set
Asylum-seekers who travel to Australia with a valid temporary visa (e.g. as students or tourists) may apply for a permanent protection visa from the Department of Immigration and Border Protection (‘Immigration Department’). This involves lodging a written application on prescribed forms and attending an interview.
If unsuccessful, asylum-seekers may then apply to the Migration and Refugee Division of the Administrative Appeals Tribunal (AAT) to reconsider the merits of their case (the AAT took over from the separate Refugee Review Tribunal (RRT) in July 2015). Applicants prepare a written submission, including any new information or evidence, and attend a hearing. Usually, a single tribunal member conducts the hearing and makes a decision. They ask questions and present adverse information – including any credibility issues – to the applicant for a response. If the applicant is again unsuccessful, they have limited grounds for seeking judicial review, making the tribunal stage crucial.
The primary text aimed at guiding credibility assessment in these reviews is the Migration and Refugee Division Guidelines on the Assessment of Credibility (AAT, 2015). These ‘Guidelines’ are an official policy document, prepared by the AAT, drawing on other Immigration Department instructions and citing relevant case law. They are aimed at AAT decision-makers but are publicly available. Consisting of 51 paragraphs of instructions and advice, plus endnotes, they are effectively identical to the previous RRT Guidelines.
AAT decision-makers must prepare a written decision for each appeal based on the evidence they collect, including the Immigration Department application and interview, and the tribunal submissions and hearing (in which they ask questions and ‘test’ the evidence; AAT, 2015: para 16).
In addition to the Guidelines, this study includes a corpus of 27 decisions, selected from the body of anonymised tribunal decisions publicly available online via the Australasian Legal Information Institute website. I selected the decisions by running a keyword search of ‘credib*’, including the top 10 search results from the AAT and the top 20 from the RRT (sorted by the search engine by relevance). I eliminated three decisions from 1995 to 1996, published before the introduction of the Guidelines. In total, 18 different individuals made these decisions, with one joint decision by two people. Three were successful appeals, and applicants came from 12 different countries (Table 1).
Decisions corpus.
Analysis
Existing research suggests serious concerns with the way credibility assessments operate in asylum decision-making. While the various actors involved in the refugee application process each contribute to constructing the official narrative (Inghilleri, 2003; Smith-Khan, 2017; Tipton, 2008), institutions and their agents may not recognise these processes, due to conflicting language ideologies – that is to say, the beliefs that they have about language.
These ideologies are represented discursively in the institution’s texts. They influence how decision-makers interact with applicants (and other participants) and how they respond to the applicant’s language and communicative behaviour when reaching their decision. Such beliefs and choices undoubtedly affect credibility assessment.
Therefore, I focus on uncovering institutional beliefs about the role that each participant plays in this institutional setting. Van Leeuwen (1996: 35) proposes a ‘sociosemantic’ approach to critical discourse analysis. This involves identifying how linguistic choices shape the way actors and their actions are represented and exploring the beliefs or values underlying these representations. In the present analysis I apply this approach to identify how the actors involved in asylum claims are represented by the official texts outlined earlier and the underlying beliefs influencing these representations. For example, I analyse the authors’ choices in naming the various participants and consider phrase structure and the assignment of agency.
Furthermore, I explore how the texts (and appeal participants) explicitly engage with communication and language-related issues, drawing on examples from the tribunal decisions and comparing these with instructions given in the Guidelines. I identify the ideologies underlying these approaches. Finally, I consider the consequences of the identified discourses for individual decisions, in light of the challenges identified in existing research, and the broader implications for the operation of credibility assessments.
The Guidelines
As the primary institutional document governing credibility assessment, the Guidelines offer key insight into institutional discourse on the Australian asylum application and appeal process and the roles of its various participants. They advise consideration of ‘the overall consistency and coherence of an applicant’s account’ against ‘contradictions, inconsistencies and omissions’ (paras 28–29) and other factors echoing common credibility indicators identified in the existing literature (e.g. Coffey, 2003; Hunter et al., 2013; Millbank, 2009; Sweeney, 2009).
The decision-maker and applicant are the main participants discussed in the Guidelines, with some mention of interpreters, legal representatives and expert and other witnesses. Next I examine how the Guidelines present each of these participants by identifying some key discursive choices. I explain how these choices reflect certain ideologies about these participants and the nature of the process itself, as well as how these relate to the challenges raised in existing research.
The ‘applicant’: Person, cultural group member and evidence giver
The Guidelines refer to the asylum-seeker as ‘applicant’, thus classifying them according to their relevant function: they are applying for a visa (Van Leeuwen, 1996: 54). By calling them applicants, we understand that they are attempting to obtain a visa: they are only potential refugees – their credibility is unconfirmed.
Throughout, the Guidelines assign these applicants an active role in constructing the narrative, by using possessive constructions and making them the subject in transitive sentences relating to narrative construction, as set out in Table 2. The ‘applicant’s claims’ are explored, and the applicant is the one who presents evidence.
Thus, the Guidelines discursively construct the applicant as the principal, if not sole, creator of the official refugee narrative.
Count of possessive and subject positioning of the asylum-seeker in Credibility Guidelines.
There are multiple references (17 of the 51 paragraphs) to the applicant’s subjectivity, addressing some of the challenges raised by existing research. They advise that individuals may remember things in different ways and that trauma can affect behaviour and memory (concerns raised in Byrne, 2007 and Cohen, 2001). They require the decision-maker to assess each case in its individual circumstances and context, with references made to cultural and social background. This is another caution presented by existing research: that, for example, while the institution may value proper names and time indicators such as dates, these may not be culturally significant to the applicant, so they may not remember or express them in expected ways (Jacquemet, 2011).
Paragraph 34 allows negative credibility findings based on demeanour, but warns that ‘[t]he tribunal should also be aware of the effect of cultural differences on demeanour and oral communication’. This is a mild caution, compared with the criticism demeanour receives in the literature (Coffey, 2003; Millbank, 2009). Even without considering cultural difference, visual cues have been found to be unreliable indicators of deception (Nolan and Goodman-Delahunty, 2015), and decision-makers may have trouble setting aside their ‘human response’ to such cues, even when advised to do so (see UK decision-makers’ accounts in Thomas, 2011: 150).
The guidance presents the applicant as a ‘person’ whose actions, beliefs and choices are inherently linked to their cultural and social background. For example, it emphasises accommodating intercultural communication and accounting for psychological factors that may affect recall and behaviour. As I explore next, however, this differs significantly from the discursive presentation of the decision-maker. Indeed, I argue that in many ways, the applicant’s role construction results from what the other participants are not: the applicant’s responsibility is reinforced by the lack of recognition of other actors’ roles in narrative construction.
The reasonable decision-maker: Institution over individual
In contrast to the agency assigned to the applicant regarding narrative construction and their characterisation as a cultural and social being, the Guidelines emphasise the decision-maker’s objectivity or neutrality. Rather than being assigned a function-based classification like the ‘applicants’, they are referred to as ‘members’ (of the tribunal) 10 times in the text. This affiliative identity indicates their position as institutional and societal insiders: they belong. This contrasts with applicants, who are outsiders, applying for acceptance into a new country and society.
However, even more commonly, the text refers to decision-makers as ‘the tribunal’ (75 times in total). Thus, these actors become ‘spatialised’ (Van Leeuwen, 1996: 59): they are classified according to the place in which they carry out their activities, literally representing the institution. Hence ‘the tribunal’ is an actor that ‘may doubt’ and ‘have regard’ and keep an ‘open mind’. This naming choice is likely influenced by legal linguistic conventions (e.g. references to ‘the court’ rather than judges; Tiersma, 1999). The effect is that the decision-maker is impersonalised. The individual making the decision is effectively backgrounded, along with their personal attributes, life experiences and beliefs. They are not just part of the institution, they are the institution: a uniform actor, with few personal characteristics. They are further referred to as ‘it’, suggesting the ultimate removal of their personhood, but without diminishing their status. Rather, this strips them of their humanness in a way that makes them nearly superhuman. It reinforces their position as objective and neutral, rather than as an active participant involved in co-constructing the official narrative through their (inevitably) subjective interpretations.
The decision-maker has different roles and responsibilities in the text compared to the applicant. The vast majority of the time decision-makers are discursively constructed as thinkers. There are a total 89 instances in which verbs or verbal nouns indicate cognitive processes decision-makers should or may undertake. These include 35 relating to the process of reaching a decision, 36 regarding conclusions or opinions and finally 18 relating to expectations and awareness (Table 3).
The decision-maker’s cognitive processes.
The remaining acts assigned to the decision-maker relate to collecting information: questioning and taking evidence (six mentions), listening and hearing (three mentions) and challenging evidence: raising, directing attention to, making aware and confronting (five mentions). They are also responsible for conducting the hearing (three mentions) and accommodating other participants (four mentions). Finally, in reaching their findings, they should provide reasons and clear explanations (four mentions).
Thus, decision-makers are constructed primarily as receivers and testers of information that others present, rather than being involved in creating that information themselves. Reference to the decision-maker’s subjectivity as a social actor is very limited. Where this occurs, unlike the applicant whose behaviour derives from their cultural group membership or background, decision-makers are individuals divorced from such deterministic group attributes. Paragraph 9 reminds them that
Findings made by the tribunal on credibility should be based on relevant and material facts. What is capable of being believed is not to be determined according to the Member’s subjective belief or gut feeling about whether an applicant is telling the truth or not. A Member should focus on what is objectively or reasonably believable in the circumstances. (Throughout, italics indicate my emphasis)
Decision-makers must also ‘maintain, and be seen to have, an open mind’ (para 18). While these instructions recognise potential subjectivity, they equally imply that decision-makers can actually achieve objectivity. There is no suggestion they should practise self-reflexivity and consider how their own social and cultural context and experiences influence their evaluations.
This belief in the potential for objectivity is reinforced by instructions calling for the decision-maker to apply their expectations based on their knowledge (informed by government-issued country information) of people from particular countries of origin. For example,
[t]he tribunal may doubt part of a person’s evidence if a person’s testimony is incoherent or vague or lacks the detail or knowledge where greater detail or knowledge might be expected of a person in the person’s claimed position or from the person’s social or cultural background. For example, the tribunal is entitled to have regard to an applicant’s level of knowledge of matters about which the applicant would reasonably be expected to know if his or her claims were truthful. (Para 32)
Thus, the decision-maker is trusted to determine the reasonable level and type of knowledge an individual from a certain background should have. This presents obvious challenges for people who come from socially stigmatised groups that fall outside institutional or stereotypical expectations of what is reasonable for certain people (see discussion in Shuman and Bohmer, 2014). It also reinforces the difference between applicants and decision-makers: applicants’ behaviour and knowledge are inherently linked to their cultural or social group, whereas decision-makers are individuals.
Rather than explicitly identifying the decision-makers as agents who must decide whether the person has presented sufficient detail or knowledge, the expression is abstracted: ‘where greater detail or knowledge might be expected … matters about which the applicant would reasonably be expected to know’. This neutralises the individual decision maker’s socially, culturally and experientially informed evaluative processes, implying that ‘what is expected’ is somehow neutral or standard (resembling observations made in Noll, 2005 and Tipton, 2008). This aligns with the discursive strategy of identifying the decision-maker as the tribunal: they assume authority and uniformity in the discourse, and are therefore construed as capable of applying purportedly objective tests.
Other participants
While the Guidelines primarily deal with the decision-maker and applicant, they also mention other potential participants, including legal representatives, interpreters and expert and other witnesses. However, these participants receive much less attention, thus reinforcing the applicant’s role as primary creator of the refugee narrative. Legal representatives are mentioned directly only once: ‘It is hoped that the contents of this paper will be well understood and made use of by members, applicants and representatives’ (para 51). Apart from this, there is a single acknowledgement that
There may be good reasons why new information or claims are presented by applicants at a later stage in the application process. These reasons may include stress, anxiety, inadequate immigration advice and uncertainty about the relevance of certain information … (Para 12)
Here, the Guidelines do not explicitly mention legal representatives as actors, but rather acknowledge that advice (presumably something they produce) may be lacking. While an important concession, this structure makes the applicant the sole actor, responsible for the provision of information. This downplays the significance of the involvement and assistance of legal representatives in the process, which can make a substantial difference to credibility outcomes (a key finding in the pioneering article by Anker, 1992). Legal assistance has been found to increase the likelihood of a visa grant (Schoenholtz et al., 2014: ch. 7), with lawyers performing an important role in presenting a ‘parallel narrative’, using their legal linguistic skills and contextualising the claim (Kjelsvik, 2014; Zagor, 2014: 334). Lawyers may also interject to overcome misunderstandings, when applicants feel powerless to do so (Smith-Khan, 2017).
The Guidelines also only briefly mention interpreters, with three instructions: members should ‘be mindful of the difficulties of assessing oral evidence provided through an interpreter’ (para 21); they should ‘consider who is present at the time the evidence is to be given and whether it would be appropriate for an interpreter of a particular gender to assist’ (para 24); and finally, paragraph 34 instructs them to ‘exercise particular care if it relied on demeanour in circumstances where a person provides oral evidence through an interpreter’. While these instructions recognise that interpreting adds challenges to communication, all three still frame the applicant as the giver of evidence. The idea of evidence being provided ‘through an interpreter’ evokes the ideology of interpreters as machines or instruments for translating information. This downplays their role as active social participants who make choices about how to translate (Pöllabauer, 2004; Van der Kleij, 2015) and are influenced by their beliefs about what constitutes good or institutionally valued interpreting (Inghilleri, 2003; Jacquemet, 2011). Thus, even though the Guidelines recognise basic communication challenges, the applicant remains the primary creator of the evidence.
Finally, there are multiple references to ‘other persons’ and ‘experts’ as witnesses. Nine paragraphs mention experts, expert evidence or opinion. Experts have a valued status, referred to as ‘appropriately qualified’ and having ‘expertise’ (para 39). Similar to tribunal members, experts are assumed capable of giving an ‘objective, unbiased opinion’ (para 40), assigning them a higher level of authority or capability than normal people. Finally, the Guidelines note that expert evidence is ‘generally submitted in the form of written reports’ (para 39). The remaining six references are to ‘other persons’, which could include experts, but also encompasses lay witnesses. Here, references reflect (and sometimes accompany) the role construction of the applicant. For example, paragraphs 14 and 16 note that members ask questions and take evidence from the applicant and other persons, whose role is to give evidence. Decision-makers may assess the credibility of ‘other persons’ (para 8) and both ‘applicants and other persons giving evidence swear an oath …’ (para 14). Finally, the Guidelines note that evidence provided by two or more persons may differ and that this ‘may be due to an individual’s ability to recall an event and the emphasis and perspective placed on particular aspects of an event’ (para 30). While this acknowledges individual difference between witnesses, the influence of other participants remains invisible. Thus, these individuals’ differences in emphasis or memory alone are responsible for the shape the narrative takes, rather than the way the decision-maker emphasises certain events or features through their questioning or indeed the involvement of interpreters and legal representatives.
Tribunal decisions
By denying decision-makers’ individuality and subjectivity and instructing them to make assessments based on objectivity and reasonableness, the institution demonstrates that it values a standard approach. This is premised on the potential for the objective application of credibility indicators, expert information and advice (paras 38–42). Undoubtedly, this reflects a broader tradition within legal institutional discourse. However, the objectivity of legal fact-finding processes is not without challengers. Within the legal realist tradition, for example, fact sceptics question reliance on objective reasonableness in judicial fact-finding by arguing that such processes conceal ‘a plethora of motivations such as policies, principles, personal idiosyncracies, ideologies, and standards of morality’ (Saltman, 1991: 71). Such criticisms stimulate critical reflection of the dominant discourse underlying judicial processes. Within the asylum context, objectivity is also challenged by statistics that demonstrate large variances in success rates between different immigration offices and individual decision-makers (Luker, 2013; Schoenholtz et al., 2014: ch. 9).
Therefore, while applicants are subjective, the system’s fairness is premised on uniform decision-making processes, measured against standard, purportedly objective information. The asylum-seeker’s subjectivity must align with standardised institutional expectations of the typical person from a particular country of origin and background.
Administrative law requires that decision-makers make all the legally required relevant considerations before reaching a decision. It also requires procedural fairness for applicants: they must generally be informed of any issues or material that may be adverse to their case and be given the chance to respond to this (in Australia, the decision in Kioa v West (1985) 159 CLR 550 sets out this principle). The member’s written decision constitutes the official record of both the hearing and the decision-making process and is used to seek judicial review. Naturally, this legal process influences the way these decisions are written, in terms of both form and substance.
The decisions follow a uniform structure, with set headings. While the level of detail varies somewhat between decision-makers, all decisions mention the relevant law, often in verbatim standard paragraphs. They outline the evidence included, paraphrasing or copying sections of applicant statements, first instance decisions and parts of the tribunal hearing. They include particular issues raised and the applicant’s responses, and usually also refer to official country information.
Next I present an analysis of published decisions. I identify the way the decision-makers construct the participants’ roles and the implications this has for outcomes. I deal with the applicant last in order to contrast their role with those of the other participants as well as to identify ways in which applicants challenge the discourse, and the results of such challenges.
The decision-maker
Decision-makers often follow the institutional naming convention, referring to themselves as ‘the Tribunal’. Of the 18 decision-makers, 13 consistently refer to themselves as ‘the Tribunal’. Two others alternate between ‘the Tribunal’ and personal pronoun ‘I’. The remaining three use only ‘I’. All the decision-makers follow their choices across all their decisions in the corpus, demonstrating that they value uniformity. The effect of impersonal language is significant:
It appears as an objective and powerful finding, made not by one frail human being, but endorsed by a venerable and powerful institution … This usage thus helps legitimate the judicial system by making it appear above the fray of human emotions and biases. (Tiersma, 1999: 68)
Indeed, Western tradition favours third-person narration for this reason. Williams (2005) argues that third-person use helps to ‘reinforce the idea of impartiality and authoritativeness in legal texts’ (p. 37). It is also popular in academic writing. For example, Verschueren (2012: 84) notes that dominant third-person usage helps to background author subjectivity. This choice creates the impression that the text is written by an omniscient, neutral bystander, observing the interaction and describing the applicant’s and decision-maker’s actions, rather than it being written by and from the perspective of one of these participants themselves.
Various other choices contribute to the decision-maker’s discursive representation. These often reflect constructions in the Guidelines; however, there is variety between different decision-makers. While a majority (12 of 18) refer to themselves in active forms when recounting the hearing, the types of actions they associate with themselves reflect the roles assigned to them in the Guidelines. Many mention acts around questioning, for example ‘The Tribunal asked about her son’ (Murphy, para 67), and confronting the applicant, for example ‘I put to the applicant’ (used frequently throughout Short1). However, even the minority who background their questioning and confronting roles by using passive constructions still tend to use active language when describing their cognitive processes. So, for example in Bray’s decision, backgrounded questioning contrasts heavily with decision-making, with statements such as ‘I attach no weight to …’ (para 60) and ‘I do not accept uncritically …’ (para 63). Henry takes a similar approach, making statements such as ‘The Tribunal has also considered’ (para 71) and ‘the Tribunal is not satisfied’ (para 80). Thus the decision-makers present their roles in similar ways to the Guidelines, most universally emphasising their roles as thinkers and deciders. For those who make some mention of their interactive role in the hearing, the verbs chosen centre around questioning and confronting, but these occur less frequently and not across all decision-makers – just as they occur less frequently in the Guidelines.
These decisions highlight the influence of the institution’s ideologies, present in the Guidelines: the decision-maker’s primary role is as receiver and processor of information, rather than as an active participant in its creation. Still, approaches vary, in terms of both self-referencing choices and backgrounding practices. This indicates that while all decision-makers demonstrate that they are influenced by the institution, practice can vary between individuals. This variation (and variations in how they present other participants) exemplifies the way in which – contrary to the discourse and as raised in the literature – decision-makers are indeed diverse, drawing on their own idiosyncrasies and beliefs, evident in how each chooses to structure and present their written decisions. It is not difficult to imagine that each would equally have their own preferences and styles for oral communication and interaction in the hearings they conduct, making their interactions unique rather than uniform.
Other participants
Like in the Guidelines, the decisions infrequently mention other participants. Just as decision-makers vary in portraying the applicant’s role, there are different approaches to acknowledging and describing the contributions of other actors. In some decisions, it is not even clear whether a legal representative or interpreter was present at any stage of the application. In others, the decision-maker mentions them in an introductory sentence but does not acknowledge their involvement thereafter, despite a strong focus in the written decisions on describing what was communicated in the hearing. Some describe their involvement in some detail. Finally, the credibility of experts and other witnesses or evidence appears to depend on applicant credibility, rather than being independently assessed.
Interpreters
Similar to the Guidelines, there is very little mention of interpreter participation in the decisions corpus (Figure 1).
Type of mentions of interpreters (number of decisions employing each approach).
Four decisions make no mention of the language the applicant used for the hearing or whether an interpreter was present (Cullen, Short1, Short2 and Short3 – this included one applicant from Ghana who perhaps may be presumed to speak English). A further three mention (either explicitly or indirectly) that they were conducted in English (Murphy, Hely1 and Grau).
In total, 11 of the 27 decisions include only a pro-forma style reference to interpreting (Carlton1, Carlton2, Carlton3, Henry, Joliffe1, Joliffe2, Syme&E, Webb, Cosentino, Godfrey and Younes2), for example, the frequently occurring structure:
The applicant appeared before the Tribunal on 24 March 2015 to give evidence and present arguments. The Tribunal hearing was conducted with the assistance of an interpreter in the Sinhala and English languages. (Webb, para 3)
A further six decisions mention the interpreter beyond the initial pro-forma statements, but only in response to issues raised by the applicant (or their representative) (Millar1, Millar2, Hely2, Hely3, Derewlany and Younes1). For example,
In this regard the Tribunal notes the first applicant’s comments in response to its letter under s.424A that the interpreter did not properly interpret the husband’s query to the Tribunal. However the Tribunal is satisfied that the husband’s query was satisfactorily interpreted. (Derewlany, para 103)
The remaining three decisions do describe some interpreter participation beyond basic responses to explicit applicant concerns (Bray, Syme and Wilson). Wilson provides the most extensive descriptions, including by noting, ‘The interpreter said she was not sure of the translation of the specific movements. The applicant said the first one, something, means “putting holding the hands”’ (para 30). Later, he mentions asking the interpreter for a translation:
the Tribunal asked the interpreter what [Company A] meant in English. She said that she needed the Chinese characters written down as she cannot really tell. The applicant was asked to do so. (Para 64)
This provides some details of the interaction between the decision-maker and interpreter, including meta-communication and negotiation around challenges with meaning. It also gives information about at least one personal attribute of the interpreter – her gender.
While on their own these examples may seem banal, they highlight what is missing in the majority of the other decisions. In the latter, details of the actual participation of the interpreter – their personal attributes, their approach to interpreting and the way they interacted with and influenced other participants – are effectively invisible. The exception is the few cases in which the applicant or migration agent raises interpreting issues themselves (I elaborate on these below). The result is that, overall, the decisions reinforce the Guidelines’ construction of the interpreter: they are not quite a fully fledged participant; rather they act as an invisible channel of communication through which the applicant presents information to the receptive decision-maker.
Legal representatives
Legal representatives also receive scant attention from most decision-makers. In some cases, some of their contributions are acknowledged. A popular approach is describing how they supplied an initial or follow-up written submission. Even then, however, the decisions differentiate between the applicant’s primary role and the legal representatives’ more neutral participation. While some decisions describe how the legal representatives claim or explain something, in others (Carlton1, Joliffe1, Joliffe2, Murphy and Webb), the submission itself becomes the actor. For example,
The applicant’s representative provided written submissions … those submissions referred to additional information provided by the applicant … The submissions seek to address the delegate’s concerns … The submissions, as indicated, also provided a copy of an International Committee of the Red Cross document … The submissions state that the delegate made an error. (Joliffe1, para 25)
Van Leeuwen (1996) calls this ‘utterance autonomisation’ (p. 60): the actor is represented by referring to their medium of communication. This lends ‘a kind of impersonal authority to the utterances’ and is often used to reflect official or high status.
In contrast, four decisions assign agency to the representative (Millar1, Millar2, Short1, and Short2). For example,
In the submissions [in] May 2013 the representative recites the various claims and evidence put forward by the applicant about his protection claims and submits that he is at risk in Bangladesh because of that. (Millar2, para 122)
In one case, where the applicant wrote his own submission, the language assigns him agency (Hely1). Finally, despite the representative having drafted the submission, another decision assigns agency to the applicant:
By letter dated [in] June 2011, the applicant provided the following reply to the Tribunal’s s 424A letter: First, I would like to thank you for giving us the chance to put in writing all issues you have raised in your letter. In relation to [the applicant]’s application for protection visa … (Hely3, para 64)
Therefore, in a majority of decisions mentioning a written submission, the representative’s personal involvement is either backgrounded, by assigning agency to the document, or else it is reassigned to the applicant. This reinforces the applicant’s responsibility for shaping the narrative while neutralising the representative’s role: they are hidden behind the (often) written form of their acknowledged contributions.
Another choice is whether to mention the existence of a legal representative at all. In seven decisions it seems the applicants are unrepresented – that they had no professional assistance in their appeal. Yet, in only one decision is this clearly stated: ‘The applicant, who was not represented, did not raise a claim that future persecution would be for a Refugee Convention reason’ (Syme&E, para 28). Syme and Eteuati thus acknowledge that lack of assistance affected the applicant’s ability to make a claim. This contrasts with the other six decisions in which it would appear that the applicants are unrepresented, but where this fact remains completely unacknowledged, despite at times the quality or focus of their arguments or their lack of legal knowledge obviously having an impact on the success of their appeal (Bray, Cosentino, Cullen, Derewlany, Hely1 and Short3). For example, for an applicant who claimed to have been abused by her ex-husband, Cullen differentiates between her past struggles, or ‘tough life’, and future persecution:
the Tribunal is not satisfied that the applicant faces a real chance of persecution involving serious harm were she to return to China in the reasonably foreseeable future for any of the reasons she claims associated with having a tough life. (Cullen, para 100)
In this and the other cases of apparently unrepresented applicants, there was often a lack of strong third-party evidence and submissions constructed around refugee law concepts, or what Zagor (2014: 334) calls the ‘parallel narrative’, essential for a successful claim and hard to construct without professional assistance.
Both the backgrounding of representatives when describing their written submissions and their complete absence from the text in other cases mean the decisions tend to reinforce the Guidelines’ discursive position on the insignificance of legal representatives. This minimises the vital role that legal assistance can play in preparing applicants for their interactions with decision-makers and supporting them with a parallel legal narrative to help focus and legitimise their experiences for this specialised setting, thus maximising their institutional credibility.
Witnesses and experts
In 12 decisions, applicants included some type of written witness evidence in their application. Five included written expert statements to support claims related to injuries or mental illness (Webb, Bray, Millar1, Hely2 and Hely3). In 10 cases, written statements came from a variety of other witnesses, including relatives, friends, political party members, business associates or fellow religious adherents (Carlton2, Syme&E, Bray, Millar1, Millar2, Murphy, Hely2, Derewlany, Younes1 and Younes2).
Responses to the applicants’ witnesses vary, but the common trend is a clear alignment of the witnesses or evidence with the applicants. In the most extreme form, for example, some of the written statements are simply disregarded as suspected fraud. In other words, the applicant has created the document themselves, perhaps inventing the person named in the document, for example an alleged political party official (Murphy). Alternatively, statements provided by friends or relatives are construed as being written simply to support the applicant’s claims, under the applicant’s direction. In other cases, such as for medical and psychological reports, while the decision-maker may trust the writer, they limit the document’s significance by arguing that it relies on the applicant’s self-reports to the expert, meaning such documents offer no additional support to the claims the applicant has already made.
In none of the reviews did expert witnesses attend in person, and in only three cases did lay witnesses attend the hearing to support the applicants. In one of these, where practising Falun Gong was central to the applicant’s claim, the witness (‘Ms A’) was a fellow adherent (Derewlany). While the applicant’s explanations of her history with Ms A are presented in some detail, Ms A’s participation in the review is summarised in two short paragraphs:
[Ms A] confirmed that she had provided a letter of support for the first applicant and had stated in the letter that she knew the first applicant in China, and that the first applicant had been persecuted because of her Falun Gong practice. She confirmed that the first applicant worked in the children’s hospital in China because her relative’s child had been hospitalised and cared for by the first applicant. She had phone contact with the first applicant (from China) who told her she was scared because she had been threatened by the police. The first applicant submitted a letter of support from [Ms A] dated [in] January 2010, stating that she was a Falun Gong practitioner and met the first applicant at the Children’s Hospital in [Location 1], adding that the first applicant had been responsible for the care of a relative in the hospital. The letter stated that the first applicant commenced her study and practice of Falun Gong in the summer of 2005. It was stated that the first applicant was persecuted in China because of her practice and came to Australia to seek protection. (Paras 83–84)
No further mention is made until the ‘Findings and Reasons’, where Derewlany finds ‘that [Ms A]’s evidence does not overcome the significant problems of the evidence relating to the first applicant’s and concludes it has been contrived for the purposes of boosting the first applicant’s refugee claims’ (para 114).
In another case, a friend of the applicant, now living in Australia, explained how he had witnessed the applicant being attacked. The decision-maker states,
The Tribunal also found [Mr A] to be an unconvincing witness and does not accept that his evidence was truthful. From his own evidence, he had met [Applicant 1] only once or twice in person before he claimed that [Applicant 1] rang him to tell him that he was married in February 2005. He claimed that he travelled to visit [Applicant 1] at his home in the Punjab and arrived just as he was being attacked by his wife’s relatives. The Tribunal does not believe that this incident ever occurred. (Henry, para 80)
In the third case, the witness claimed to have been in a relationship with the applicant, whose claim centred on being homosexual (Millar1). Despite the fact that ‘the Tribunal found him to be sincere in discussing his own sexuality and accepts that he himself is homosexual’ (para 58), Millar takes issue with various inconsistencies in the applicant’s case and rejects the applicant’s claim that he is homosexual. He then uses this conclusion to reject as not credible the claims made by the witness (and written claims of another witness) attesting to the applicant’s homosexuality, rather than considering this evidence separately. Effectively, this means that the applicant’s own credibility appears to predetermine his witnesses’ credibility, rather than each person’s credibility being assessed separately. In this representation, the applicant remains the primary narrative constructor, who may present other persons or evidence simply as props or agents in the applicant’s construction activity.
The reception of witness evidence in the corpus demonstrates how witnesses, as participants in asylum appeals, are discursively very closely connected with applicants. This is similar to the way witnesses are linked to applicants in the Guidelines. This has serious implications for the ability of applicants to provide evidence to support their claims and boost their credibility: if the decision-maker finds inconsistencies or implausibility that affect an individual’s credibility, there appears a risk of circular reasoning in which additional evidence that may support credibility is rejected because credibility is already tainted.
The applicant
Reflecting the Guidelines, decision-makers generally refer to ‘the applicant’, which is unsurprising since published decisions anonymise applicant details. Other language also reflects the Guidelines, with frequent expressions across the corpus such as ‘the applicant’s claims’, ‘the applicant’s account’ and ‘the applicant gave evidence’. Some decision-makers more frequently use neutral verbs such as ‘the applicant said’ and ‘the applicant stated’, while others prefer marked language such as ‘she has now claimed’ and ‘allegedly’. Certainly, the latter are useful devices to communicate disbelief in particular aspects of an applicant’s account, thus supporting negative conclusions. Regardless of whether they use value-laden terms, the language reflects the belief, as presented in the Guidelines, that the applicant is primarily responsible for narrative construction. This belief is reinforced when contextualised against the role construction of the other participants, as explored earlier. However, this does not prevent applicants from challenging this assumption.
Challenging the discourse
In some cases, the applicants or their representatives challenge the decision-maker’s (and institution’s) discursive construction of participant roles in order to respond to credibility concerns. These challenges mostly relate to redistributing responsibility for narrative construction across the various participants, thereby challenging the discursive representation of the applicant as sole author. Their arguments (presumably unwittingly) echo those made in the literature: that the narrative is co-constructed and that this official version is entextualised through the series of interactions that make up the application and appeal process (for detailed discussions on entextualisation, see Blommaert, 2001; Jacquemet, 2009, 2011; Maryns, 2006: 340). Thus, while the Guidelines acknowledge the need to accommodate intercultural communication, their focus on cultural difference may actually obscure the interactional and linguistic factors impacting applicants’ participation (see Piller, 2011 and Smith-Khan, in press for arguments regarding this practice in intercultural communication discourse).
Numerous responses to credibility concerns (such as inconsistency or missing detail) draw attention to the role of the interpreter or translator. For example,
The Tribunal indicated to the applicant that there appears to be inconsistencies in the evidence, namely that in oral evidence he had said that he was giving the customer some material whereas in writing he has claimed that he was sending the material. The applicant stated that the mistake had been made by the translator. The Tribunal indicated that the inconsistency could raise doubts about the veracity of his claims and his credibility generally. (Younes1, para 74)
In this case, Younes not only holds the applicant responsible for the inconsistency, but also disregards the fact that the inconsistency she identifies is between two terms in the translated rendition, rather than those used in the original Mandarin utterances. This approach reflects the ideology, described above, that the interpreter is a mere machine, capable of faithfully reproducing verbatim the utterances of the speaker, and overlooks in this case the possibility that there may not be a perfect alignment of terms across languages or that different interpreters or translators may make different lexical choices.
Responding to a concern that a certain detail had not been mentioned in the first instance interview, another applicant noted that the interpreter had been ‘slow’, spent time taking notes and perhaps had not mentioned this specific detail in his translation. Again, the decision-maker assigns responsibility to the applicant: ‘The Tribunal stated that she had not raised any issue regarding the interpreter with the Department’ (Derewlany, para 38). This reflects an expectation that the applicant should know that a detail was missed in the English rendition of her speech in her first interview, even though she is not made aware of this until the Tribunal hearing, linguistically not having access to the English rendition at the interview. The expectation that the applicant would make a general complaint about the interpreter’s aptitude, especially without being aware of the specific details of any errors or oversights, overlooks the impact of power dynamics in such settings, as well as the basic linguistic barriers (see discussion and examples in Smith-Khan, 2017).
The applicant who claimed to face persecution based on being homosexual highlighted the interpreter’s role as a social actor. During the hearing he gave details of his first relationship and sexual encounter, which appeared inconsistent with an earlier written statement. Responding to this concern, the applicant described his ‘stress, shyness and confusion’ in sharing this information with a female Indian interpreter. The decision-maker does not accept that the social dynamics or the sensitive and personal nature of the subject matter explained the way the applicant had ‘prevaricated and put forward the vague responses’ (Millar1, paras 13–14).
Another example involved both linguistic issues and access to legal assistance: to address inconsistencies arising from his initial written statement, the applicant explained that there had been no interpreter to facilitate communication between the applicant and lawyer when the latter prepared the statement. Yet the influence of other credibility concerns meant the decision-maker rejected this point, stating that he ‘placed some reliance on this inconsistency in undermining the applicant’s overall credibility’ (Hely3, para 95). Thus he dismissed the real concern that linguistic (and presumably financial) barriers had materially undermined the applicant’s access to quality legal assistance and consequently the impact this had on the development of a credible (consistent) narrative – a narrative he is held responsible for constructing despite these barriers.
The same applicant also drew attention to the structure of the application and appeals process and the role that different decision-makers play in developing the narrative through their particular style of questioning:
The Tribunal noted its concern that his accounts of what happened in respect of who attacked his brothers did not appear to have been consistent. The applicant responded that it was possible the Department officer did not ask the question in the same way as was asked by the Tribunal. (Para 55)
He went on to explain the initial absence of certain elements of his claim with a similar focus on the questioner – ‘the Tribunal had only asked who opened the door but he had not been asked if someone else had spoken before the door was opened’ (para 59), and ‘he was not asked in his interview if he went to the police station’ (para 60). Hely rejected these explanations, leading to multiple apparent inconsistencies, which cumulatively discredited the applicant.
The examples described above deal with explanations made by applicants that challenged the institutional discourse uncovered in the Guidelines and decisions: that the applicant produces and owns the final refugee narrative, with minimal impact from other actors, linguistic factors or the structures of the process itself. The reception of their arguments suggests two conclusions: first, that while asylum-seekers may face power asymmetries, they are not completely without agency in terms of challenging institutional processes. Second, however – and very importantly, when these challenges conflict with core institutional ideologies, it is – perhaps unsurprisingly – improbable that the decision-makers will accept their arguments.
Fair go? Reining in credibility assessments
In this article I have critically analysed the official text guiding credibility assessment in Australian asylum reviews and a corpus of published decisions dealing explicitly with credibility. By identifying the roles assigned to each participant in these texts, I have uncovered key institutional ideologies about asylum procedures. First, I identified a belief that the applicant constructs the official narrative that comes out of the application and appeals process. While the decision-maker is represented as playing an important role in thinking and questioning, their role as narrative co-constructor is minimised. Next, there is a contrast between the belief that asylum-seekers (and other witnesses) are subjective beings and the assumption that decision-makers are capable of objectivity and neutrality when assessing credibility. Finally, when considering the question of credibility, the institution pays little attention to the influence of other participants, such as legal representatives and interpreters.
These assumptions overlook significant factors highlighted by existing research. The nature of the setting creates the parameters for what story should be told and how: asylum-seekers must perform a limited version of themselves – the part that faces persecution for Refugee Convention reasons (see Vogl, 2013 for a discussion on specialised refugee narratives). Rather than the applicant constructing a narrative in a vacuum, in reality the process is a group effort: each participant involved in the application process has a role to play in its creation. Furthermore, the power imbalances in this setting mean that instead of being able to lead or drive narrative construction, applicants have to comply with structural limitations (answering questions and filling out forms) and may often defer to others with higher status: decision-makers, lawyers and interpreters. Far from being capable of perfect objectivity, decision-makers draw on their own experiences and perspectives in the evaluative process of applying credibility indicators. The varying approaches taken by different decision-makers in the corpus exemplify this point. Furthermore, the narrative is officially presented from their perspective, rather than the applicants’. This means that even when applicants do challenge the discursive, structural and linguistic barriers they face, the decision-maker makes the final decision on the legitimacy of their arguments.
In the decisions analysed here, while seeking to address credibility concerns, applicants and their advisers – perhaps often inadvertently – presented challenges that reflect the findings and arguments of research on credibility. This entailed highlighting the significant role interpreters and legal representatives play in shaping the narrative, as well as pointing to the structure of the interactions involved in applying for asylum and appealing a negative decision, including the way decision-makers steer the narrative through their questioning. Yet when applicants (and other participants) challenge credibility concerns by drawing attention to such factors, they face a formidable challenge: they must confront and overcome institutional discourse – as uncovered earlier – that ignores or contradicts these phenomena. In the current analysis, these attempts are therefore unsuccessful.
It is altogether possible that in the cases examined, the conclusions reached were just: not all applicants have genuine claims. However, uncovering the ideologies that underlie the representation of the various actors raises serious questions. Can credibility assessments based on problematic assumptions about communication and participation ensure fair outcomes in all cases? While not surprising, the fact that the explanations applicants presented in response to credibility concerns were rejected because they clash with dominant language ideology is troubling. If the discourse was different, issues framed as significantly affecting credibility, for which applicants are held responsible to their detriment, may not be considered issues at all. The discourse around the objectivity of applying credibility indicators only helps to further obfuscate the matter. Credibility concerns that are considered a matter of objective fact rather than arising out of evaluative processes influenced by institutional ideologies and individual idiosyncrasies are much less open to scrutiny or self-reflection and thus much harder to effectively challenge.
However, uncovering this troubling discourse does not lead to a dead end. Rather, identifying and drawing attention to this discourse highlights a need for critical reflection and action. As a former Canadian decision-maker suggests, this involves ‘looking inwards – at our own values, prejudices, orientation and perspective’ (Macklin, 1998). In part, this could entail a stronger emphasis on professional development and sensitisation, drawing on the rich and ever-growing body of interdisciplinary scholarship on narrative construction in refugee procedures. This is a valuable resource for starting a conversation, challenging and reformulating dominant institutional discourse, and ultimately improving the chances of fair credibility processes.
Footnotes
Acknowledgements
I am grateful for the ongoing encouragement and assistance of my supervisor, Professor Ingrid Piller, associate supervisor, Dr Daniel Ghezelbash, and an anonymous reviewer, and the comments and support from my Language on the Move research group peers. Helpful feedback was also received in early stages of data analysis presented at the Australian Linguistics Society Conference (2015), the Association for the Study of Ethnicity and Nationalism conference at the London School of Economics and Political Science (2016) and a seminar at Leicester University (2016).
Declaration of conflicting interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Research for this article was undertaken as part of a Doctor of Philosophy degree, supported by a Research Training Programme Scholarship (formerly APA).