Genuine Improvement or Paying Lip Service:

 Conquering the Communication Complexities in Protection Assessments

smith_khan.jpgLaura Smith-Khan,

Research Assistant, Sydney Law School

Two individuals sit facing one another in a small room. Neither speaks the other’s language, and their cultures and personal histories could scarcely be more different. They have never met before, and they can only communicate through an interpreter. These two people have very little time to try to bridge all these differences, and countless others, so that one of them can reach a decision so important it may mean life or death to the other. Welcome to the world of refugee claims.1

Having ratified the UN 1951 Convention Relating to the Status of Refugees and the UN 1967 Protocol Relating to the Status of Refugees (the Convention; and the Protocol),2  Australia has protection obligations to those people found to meet the Convention definition of a refugee. The assessment of refugee claims often involves a great deal of complexity and inequality: the majority of those seeking protection are incapable of speaking the language of the forum effectively or in the same manner as a local speaker, and thus may struggle to make themselves understood.3  Efficiency and demand mean that many otherwise avoidable problems may arise even with the assistance of an interpreter, potentially leading to an inability to effectively communicate information essential to a positive assessment.4  A misunderstanding or oversimplification of cultural or linguistic information on the part of any of the parties may also lead to otherwise unwarranted negative findings in relation to credibility, an often pivotal aspect in decision making in this area.5  Considering the importance of the outcome, often to an applicant’s family group as well as her or himself,7 it is imperative that potential communication breakdowns and misunderstandings are remedied as much as possible. With a principle focus on Australian refugee merits review, this paper will attempt to explore these issues and consider the disparity between the wealth of material dealing with them and the lack of success in addressing them.8

The interpreter: not merely a communication machine

Room for error

As noted by Michael Barnett,9 Australian case law in the past likened the role of interpreters to that of a machine, simply receiving and relaying information in a straightforward, mechanical manner.10 However, more recent cases have recognised the fact that interpreting is an extremely complex process which, to be done successfully, involves a great number of skills, including cultural awareness and, at times, specialist knowledge.11 This is understandably so in an immigration setting where flawed interpreting can lead to a miscommunication of information, potentially damaging the plausibility and internal and external consistency of applicants’ claims.12 Thus, reference to these traditional credibility grounds in such circumstances may prove fatal to applicants’ credibility,13 and often to their cases.14

As Sonja Pollabauer explains, interpreters can undermine communication in a number of ways. Interpreters may subtly control the conversation by making interruptions, either to clarify (especially when they feel they know the questioning procedure well), or because their memory capacity is being strained (as is likely when dealing with large passages of narrative). While, to a certain extent, such practices may be unavoidable, too much interruption and thus control or steering of the conversation can lead to the applicant becoming confused or feeling the need to repeat her or himself.15 For similar reasons, interpreters may also omit parts of responses or direct the applicant to answer more briefly.16 This can reflect badly on the applicant in terms of continuity, and cause what appears to be hesitation. When taken to an extreme, interpreters may appear to be “siding with” the member, rather than acting as a neutral party. This may involve addressing the applicant directly, seemingly independently of the member, or vice versa, and thus becoming an independent participant inthe interaction, rather than acting simply as a “bridge”.17 Finally, communication problems can occur when there is a lack of cultural or linguistic understanding between the parties.18  This can occur for any number of reasons, including potentially great differences between the applicant and interpreter in terms of dialect, religion, and cultural or experiential reasons, such as lack of knowledge about, for example, a particular political party.

The interpreter’s misunderstanding or lack of knowledge can lead to poor interpretation which may lead to a vague response or the use of inaccurate terminology not correctly reflecting the applicant’s meaning (risking major issues of inconsistency),19 which can reflect badly on the applicant if the member is unaware of the interpreter’s lack of knowledge, which will be the case if inaccurate interpreting is being used as a face saving technique (intentionally or not) by the interpreter to cover their lack of understanding or vocabulary.20 Other times, the interpreter’s lack of insight over the significance of choosing one possible meaning over another can lead to significant issues for the applicant in making out their claim. Marco Jacquement gives the disturbing example of the interpretation of the Spanish word “fracas”. The interpreter chose to interpret it as “economic failure” rather than “physical harm” (including torture). Understandably, this one choice had dramatic consequences for the applicant, who was judged as being an economic migrant rather than a refugee seeking protection from politically based violent persecution.21 As will be discussed in greater detail below, such practices can further exacerbate the power imbalances between the parties involved in communication.

Exacerbating factors

Connection between positions of vulnerability and a practice of controlling and limiting participation is particularly noticeable when the applicant is a minor.22  Olga Keselman, Ann-Christian Cederborg and Per Linell observed a number of Swedish asylum hearings involving children and noted the way interpreters often manipulated the child’s involvement in the communication in various ways. Interpreters would sometimes exclude the applicant (refusing to interpret certain conversations between other parties for the applicant, even when specifically requested by the applicant to do so.23 Also, they would sometimes discredit the applicant by purporting to be experts on the applicant’s place of origin and commenting on the accuracy of the applicant’s knowledge, without relaying these comments back to the applicant.24 Finally, they would also guide the conversation by asking multiple questions of their own accord, guiding the applicant to what they consider to be an appropriate response, rather than simply relaying the initial response to the officer’s question straight away.25  

In situations where the applicant has a disability, whether physical or psychosocial, the potential for communication breakdown is also heightened. This may occur due to similar behaviour as that explained above in relation to children, as assumptions may be made about the applicant’s ability to speak on their own behalf or it may have to do with the particular nature of the disability. Applicants with psychosocial disabilities, and especially where these disabilities have not been identified,26 risk having their credibility damaged through a misunderstanding of what could appear to be inconsistencies in their accounts, due to a lack of training and sensitisation for decision makers regarding such disabilities.27 In other situations, inadequate services or facilities may disadvantage those with physical disabilities. For example, in a situation where an applicant is deaf, a sign language interpreter may be required. However, if the only sign language interpreter available is one who knows Auslan (Australian sign language), or an interpreter in the applicant’s sign language is available but does not speak English, this will severely inhibit communication. This is not to mention the situation where, due to a lack of educational opportunities, asylum seekers may not even be familiar with any formal national sign language and may only use an informal sign language within the home, unfamiliar to professional sign language interpreters.28  

Basic knowledge and expert advice: misunderstanding and overreliance

Inappropriate cultural and linguistic assumptions

In all intercultural encounters, participants must confront the tension between their expectations and resulting interpretations – a tension caused by the commonsensical urge to interpret another’s communicative behaviour based on one’s own expectations of what is appropriate.29 

The above explains why even where an interpreter is not needed, there may be barriers to successful communication. Diana Eades explains the problems faced by second dialect speakers, those who speak the same language as the forum language, but using a different version of that language. While it is obvious when someone is speaking a different language that assistance will be required for the parties to communicate, it can be much easier to underestimate communication problems with second dialect speakers.30 Yet there are plenty: these can be related to culture, accent, communicative practices and / or the way a language is used. Cultural context can have a profound impact on communication. For example, an applicant’s inability to give the full name of an important person in their narrative or give a western style numbered street address for a location can often be simple reflections of cultural norms rather than an element to draw doubt on their claims.31 Jacquemet also notes that a simple different emphasis on words in a sentence can lead to a difference in tone and breakdown in communication between the parties (often without them even being aware of what has happened).32 Different communicative practice can lead to fundamental misunderstanding: Eades gives the example of Aboriginal Australians’ use of silence to mean something other than that they have nothing to say, as an Anglo Australian may assume.33

A little knowledge can be a dangerous thing

One arguably unfortunate result of the recognition of the complexity of language factors in is the inconsistency with which this knowledge has been used. While the recognition of specific language issues is a positive step towards the acknowledgement of the many complexities involved in an individual’s manner and ability of communication, the danger is that this knowledge will be relied on in a limited, oversimplified way. Ironically, this can therefore both increase the risk of a mistaken conclusion of fact while giving the decision maker greater confidence and force in their reasoning. Two particular areas where this may occur are in the practice of referring to generic language information sources in determining external consistency and also in specially appointed third party language analysis used for nationality determination of individual applicants.

The Refugee Review Tribunal (RRT) has requested general language related information to assist in a number of reviews. Worryingly, the responses to such requests partly rely on generic information found on undated, non-academic websites.34 It is understandable that decision makers who potentially lack expert sociolinguistic knowledge may commit errors when attempting to apply such general information to specific, diverse individuals, and failing to consider the many possible influences on the way in which someone speaks, especially in the case of an asylum seeker who may have come from a country with “porous borders”, and may have spent several years in various third countries, potentially including Australia, interacting with people from many different parts of the world.35  

Michael Gordon shares the words of one Afghan man in Nauru, explaining the rejection of his family’s review:  

I give my interview in english ... I’m talking and [the case officer] says I am from Pakistan because I am talking like an Australian. “if I am Pakistani, you are a Pakistani ... I learned English here,” he said.”‘I worked in kitchen ... I also learned from Zakir Hussain Jaffari. ... He always told me: you should try to talk like Australians. Now my daughter also talks like you. When we came here she was three and a half years. Now she is [seven and] talking like you. She is also Pakistani? 36

Ironically, this man had presumed learning English would show that he was putting in an effort, but the gesture backfired, actually creating suspicion in the decision maker.

As Tim McNamara points out, “language tests can quickly be transformed from benign means of access into weapons to be used in the pursuit of group goals”.37 External language analysis is increasingly being used in a number of countries,38 mainly in situations where there are suspicions over an applicant’s claimed nationality, thus potentially affecting that basis of their claims, as well as their credibility more generally. The quality of such findings has been heavily criticised by professional linguists, who underline the impossibility of giving definite answers on questions of language,39 and the danger of confusing the influence of origin on language with that of language socialisation.40 While it is noted that reliance on language analysis has decreased in Australia,41 the practice still exists to some extent, with recordings sent to Scandinavian companies to undertake reviews.42 Both McNamara and Eades are extremely critical of the practice, claiming that in some cases, “there is simply not the sociolinguistic information available to determine the issue accurately”.43

Legal responses

In the leading case in Australia, Perera v MIMA,44 the complexity of the interpreting process was appreciated and the related grounds for appeal were set out. Justice Kenny noted that while the Migration Act 1958 only states the RRT Member “may” elect to use an interpreter where the person giving evidence is not proficient in English,45 when this is read in combination with the requirement that the review applicant appear in order to present evidence,46 it is concluded, “that the Tribunal lacks the jurisdiction to continue the hearing before it unless it provides an interpreter”, as otherwise the review applicant would be effectively denied the ability to present evidence, thus also undermining the duty of the Member to act in accordance with substantial justice as set out in section 420(2)(b)of the Migration Act 1958.47 Further, this being a jurisdictional error, it thus does not fall foul of the section 476 privative clause.

The test set out by Justice Kenny in Perera requires there to have been a departure from the appropriate standard of interpreting so substantial as to prevent the applicant from presenting evidence;48 and that this departure related to “matters of significance,” either to the claim or to the decision.49 While subsequent cases disagreed somewhat on whether both parts of this test needed to be satisfied, Perera has been generally recognised as applicable.50 Following the amendment of section 425(1)(a),51 section 425(1) now reads that the “Tribunal must invite the applicant to appear before the tribunal to give evidence and present arguments relating to the issues arising in relation to the decision under review.” However, in subsequent cases, this amendment was considered to have no effect on Justice Kenny’s understanding of the right of the review applicant to present evidence.52 Indeed, the rrt itself continues to recognise the perera test as the applicable one for determining when insufficient interpreting will give rise to jurisdictional error.53

While Perera and subsequent cases focus on the quality of interpreting,54 what can be lost is the significance of elements which go beyond the simple communication of words but are still directly related to the performance of the interpreter and to the ability of the applicant to properly communicate their experiences, as discussed earlier. In some cases, with issues related to the interpreter, the court seems to have placed extra conditions, making meeting the Perera test insufficient.55 In others, there are grey areas in which the effectiveness of communication is undermined in ways that may make it more difficult to make a case which would meet the Perera requirements because the interpreter’s behaviour may produce emotional effects, creating confusion or a loss of confidence in the applicant that could affect their ability to communicate but not in a way as tangible as explicit omissions or mistranslations. Still, there may be strong evidence that the interpreter is acting partially, such as where she or he provides advice regarding the applicant’s accent to a member, as in Sook Rye Son v MIMA.56 Here, the “most significant concern regarding the Tribunal’s use of the interpreter was the fact that the applicant was not made aware, other than in general terms, what the interpreter’s views on the applicant’s accent and language were. This use of the interpreter was said to amount to a ‘fundamental unfairness’ to the applicant”,57 and thus offers another potential ground for appeal in like circumstances which do not fit easily within the Perera framework.

While it appears Perera and subsequent cases have paved the road for a right of appeal based on jurisdictional error where an applicant has been unable to present information significant to their claim or to the member’s decision, it is less clear whether other communication breakdowns could be remedied in such a way. However, there is some prospect of success – with expert assistance – to object to the use of arguably weak language analysis by relying on the right to be made aware of any factor crucial to the case and the right to comment on any adverse information from a third party as set out in MIMA v Lay Lat.58  This case redefines the right to a fair hearing after the addition of section 422b to the Migration Act 1958, following the decision in Re MIMA, ex parte Miah (2001) 206 CLR 57. Section 422B(2) limits the natural justice hearing rule to the contents of sections 416, 437 and 438 and divisions 4 and 7A. Importantly, division 4 includes section 424A, which obliges the member to supply the applicant with any information “which would be the reason, or a part of the reason, for affirming the decision that is under review”,59 make sure they understand its significance,60 and invite them to respond.61 This is limited to information which is personally related to the applicant.62 Arguably, language analysis reports could fall within the ambit of section 424A and thus be a ground for judicial review due to a denial of a fair hearing. This would depend, of course, on whether or not the applicant had been presented the report and given a chance to respond during the review.


This article has explored the ways in which issues and assumptions related to language have the potential to undermine protection applicants’ credibility, and consequently their likelihood of success. It thus emphasises the importance of awareness of such issues among all relevant stakeholders. The RRT itself advises members to be aware of the many factors, including emotional, cultural, linguistic and experiential, that may impact on an applicant’s ability to provide a clear and consistent narrative.63 The best course for adjudicators in regard to communication issues is to identify and treat with caution all information that is not reliable due to its basis in one of the above discussed areas, especially where such information may otherwise have adverse consequences. Likewise, it is essential that the representatives of applicants be aware of potential language and communication issues. Given the great potential for communication breakdown, sufficient time should be set aside to clarify all important information before departmental interviews and RRT hearings. Further, representatives should be alert for any misunderstandings that may affect the credibility, and by extension the success, of their clients during all interactions with departmental officers or adjudicators and raise them where necessary.

As has been seen, given the complexity of communication, it is important to continue developing a better understanding of the issues discussed above and exploring how best they should be addressed. This involves the consideration of more than just the relevant law in the area. As has been discussed, before the law can even be considered, a proper understanding of the social, psychological, linguistic and cultural elements relevant to communication need to be examined by those with the most adept knowledge, meaning that interdisciplinary cooperation is essential. recognising and striving to address the issues discussed in this article is therefore fundamental to ensuring that all applicants have a fair hearing and consequently have their protection claims justly decided. 

The author thanks Daniel ghezelbash and the anonymous reviewer for their suggestions and advice. All errors remain the author’s own.
1  Peter Showler, ‘Bridging the grand Canyon: Deciding refugee claims’ (2007) 114(1) Queen’s Quarterly 29, 30.
2  Convention adopted 28 July 1951 and Australia ratified 22 January 1954. Protocol adopted 31 January 1967 and Australia ratified 13 December 1973.
3  Marco Jacquemet, ‘Crosstalk 2.0: asylum and communicative breakdowns’ (2011) 31(4) Text & Talk 475, 478.
4  Nienke Doornbos, ‘Chapter 6: On being heard in asylum cases, evidentiary assessment through asylum interviews’, in gregor Noll (ed) Proof, evidentiary assessment and credibility in asylum procedures (Martinus Nijhoff Publishing, 2005) 103, 121.
5  Michael Barnett, ‘Mind your language – Interpreters in Australian immigration proceedings’ (2006) 10 UWSLR 109, 112.
6  Sonja Pollabauer, ‘Interpreting in asylum hearings: Issues of role, responsibility and power’ (2004) 6(2) Interpreting 143,143.
7  Steve Norman, ‘Assessing the credibility of refugee applicants: A judicial perspective’ (2007) 19 IJRL 273, 274.
8  Rosemary Byrne, ‘Assessing testimonial evidence in asylum proceedings: guiding standards from the International Criminal Tribunals’ (2007) 19(4) IJLR 609, 612, 629.
9  Barnett, above n5, 114.
10  Ibid. Barnett refers to Menzies J (432-433) and Kitto J (430) in gaio v The Queen (1960) 104 CLR 419.
11 Barnett, above n5, 115.
12 Byrne, above n8, 623.
13 Ibid, 624.
14  See generally guy Coffey, ‘The credibility of credibility evidence at the Refugee Review Tribunal’ (2003) 15 IJLR 377.
15 Pollabauer, above n6, 154-5.
16 Ibid at 159.
17 Ibid at 158.
18 Ibid at 171.
19  Olga Keselman, Ann-Christin Cederborg & Per Linell, ‘“That is not necessary for you to know!” Negotiation of participation status of unaccompanied children in interpreter-mediated asylum hearings’ (2010) 12(1) Interpreting 83, 95-97.
20  Ibid, 93-4, in the sample interview, the interpreter is weak in both target languages and uses inaccurate terms which lead to the officer to view the applicant’s recount as vague and disjointed.
21  Jacquemet, above n3, 487, referring to the work of Inghilleri (2003).
22 Keselman et al, above n19, 101-2.
23 Ibid, 90-1.
24 Ibid at 95-7.
25 Ibid at 99.
26  Niklas Miller, ‘The Case History in a Cross-cultural Milieu’ in Niklas Miller (ed), Bilingual and Language Disability: Assessment and Remediation (Croom Helm, 1984) 169.
27  Mansha Mirza, ‘Disability and Cross-Border Mobility: Comparing Resettlement Experiences of Cambodian and Somali Refugees with Disabilities’ (2011) 26(5) Disability & Society 521, 529.
28  For an example of a similar situation see the case study in Jennifer Harris and Keri Roberts, ‘Challenging Barriers to Participation in Qualitative Research: Involving Disabled Refugees’ in Bogusia Temple and Rhetta Moran (eds) Doing Research with Refugees: Issues and guidelines (Policy Press, 2006) 155, 162.
29 Jacquemet, above n3, 478.
30  Diana Eades, ‘Participation of second language and second dialect speakers in the legal system’ (2003) 23 ARAL 113 124.
31 Jacquemet, above n3, 485-486. 
32 Ibid 477-8.
33 Eades, ‘Participation’ n30, 125.
34  See, for example Refugee Review Tribunal, Country Advice – ghana – gH37464 (6 October 2010); Country Advice – ghana – gH37075 (24 August 2010).
35  Diana Eades, ‘Applied Linguistics and language analysis in asylum seeker cases’ (2005) 26(4) ARAL 503, 510.
36  Michael gordon, Freeing Ali: The Human Face of the Pacific Solution (online excerpt) (UNSW Press, 2005) <>.
37  Tim McNamara, ‘21st Century Shibboleth: Language tests, identity and intergroup conflict’ (2005) 4 Language Policy 351, 369.
38  Diana Eades, ‘Testing the claims of Asylum Seekers: The Role of Language Analysis’ (2009) 6(1) Language Assessment Quarterly 30, 37.
39  Language and National Origin group, ‘guidelines for the use of language analysis in relation to questions of national origin in refugee cases’ (2004) 11(2) IJSLL 261, 265. See also the Language and Asylum Research group <>.
40 Language and National Origin group, above n39, 262.
41 Eades, ‘Testing the claims of Asylum Seekers’, above n38, 37.
42  Eades, ‘Applied Linguistics’ above n35, 506-7; the most recent RRT decision mentioning the use of language analysis undertaken by Swedish company Skandinavisk Sprakanalys A is from October 2010: 0910010 [2010] RRTA 931 (28 October 2010). Other recent published reviews include: 0909999 [2010] RRTA 960 (27 October 2010); 0910048 [2010] RRTA 911 (20 October 2010); 0909931 [2010] RRTA 842 (13 September 2010). While language analysis supported the applicants’ claims in these situations, these reviews nonetheless demonstrate that external language analysis continues to be undertaken and accepted in australia.
43 McNamara, above n37, 363.
44 (1999) 92 FCA 507 (‘Perera’).
45 Migration Act 1958 s427(7)
46 At the time s425(1)(a).
47  perera, above n44, [20]-[21]. Similarly, the need for an interpreter in court is recognised where necessary for ‘due process and procedural fairness’: DZaaa & Ors v Minister for immigration & anor (2011) FLR 423 (Lucev FM) [432].
48 perera, above n44, [38].
49 perera, above n44 [45].
50 Barnett, above n5, 121-122.
51 Amended by n3 of the Migration Legislation Amendment Act (No 1) (1998) (Cth).
52  See, for example Mazhar v MiMa (2000) 64 ALD 395 at 27-31; Singh v MiMa [2001] FCA 1376.
53  See Migration Review Tribunal – Refugee Review Tribunal (MRT-RRT), Legal Issues Paper No.2/2006 – Current Legal Issues arising out of the Tribunals’ use of Interpreters at 6; Interpreters’ Handbook (2007) at 8.
54  For a good summary of the issues in recent cases, see MRT-RRT Legal Issues Paper, above n53
55  Barnett, above n5, 128-129 discusses Szaaj v MiMa [2004] FCA 312 (Hill J) [47], in which it was found that the applicant’s delay in raising a language concern detrimentally affected his ability to rely on the test. barnett rightly cautions against this approach, underlining the inability of many applicants to be able to recognise faults in translation, given their own limited English skills and knowledge of procedural issues.
56 (1999) 86 FCR 584 Moore J, 602.
57 MRT-RRT, Legal Issues Paper, above n26, 9.
58 [2006] FCAFC 61 [46]-[47].
59 Migration Act 1958, s424A(1)(a).
60 ibid at s424a(1)(b).
61 ibid at s424a(1)(c).
62 ibid at s424a(3)(a).
63  Migration Review Tribunal & Refugee Review Tribunal guidance on Assessment of Credibility (August 2008) 4.3.