Curricular design and implementation of a training course for interpreters in an asylum context

Asylum settings increasingly require interpreting in language combinations for which few formally-trained interpreters are available. The consequences of this have been comprehensively discussed in the literature. This paper describes a project entitled QUADA: Qualitätsvolles Dolmetschen im Asylverfahren (High-Quality Interpreting in Asylum Proceedings), the aim of which was to develop a viable approach to improving interpreting quality within asylum settings in Austria. The project was initiated by UNHCR Austria and co-financed by the European Refugee Fund and the Austrian Federal Ministry of the Interior. It involved the development of a training curriculum as well as the production of a handbook for trainees and trainers. The article commences with a brief overview of research on interpreting in asylum proceedings and the challenges associated with designing training programmes for community interpreting in general, and for asylum settings in particular. It then describes in detail the project and various project phases, addressing theoretical, pedagogical and organisational aspects.


Introduction
This paper focuses on the design of a specific training curriculum for interpreters in an asylum context (not interpreter or translator training in general or court interpreter training) and outlines its implementation in a national European context.
Quality enhancement stands at the core of the project outlined in this contribution.To achieve and enhance quality in interpreting, a broad range of factors come into play: diverse stakeholders and interest groups need to be involved in the process and, ultimately, changes often boil down to political will and money.
With respect to quality in interpreting it may be assumed that, irrespective of the type of setting, all participants in an interpreted interaction are (or at least should be) equally committed to ensuring that high-quality interpreting can be achieved -if for no other reason, one might think, than the potential extra costs that a flawed interpretation may entail.
The "costliness" of interpreting can be illustrated with an example based on a national (Austrian) context.For Austria, as for many other European nations, the civil war in Syria resulted in a massive increase in the number of refugees seeking entry.In 2015, the country received a total of 88,340 applications for international protection (Austrian Federal Ministry of the Interior, 2015).Of the countries that keep records, Austria was the sixthlargest recipient of applications for international protection in 2015 (UNHCR, 2016, p. 39), behind Germany, the United States of America, Sweden, the Russian Federation, and Turkey.Although numbers fell again in 2016 and 2017, the OECD International Migration Outlook 2017 shows that "permanent migration flows to OECD countries are on the rise" (OECD, 2018).The indications are that in future, countries will also have to provide interpretation for public service institutions such as asylum and migration departments.
Because of national legal requirements, refugees are generally interviewed twice in Austria -first by the police in the initial questioning after lodgement of the asylum application, then in a detailed assessment under the in-merit procedure if the case is admitted.It seems reasonable to assume that many of the 88,340 applicants would have undergone more than one interview, putting the total of "interpreted events" well into six figures.Based on current interpreter pay rates, the costs incurred by the national asylum administration in providing interpreters for these cases would run to several millions of Euros.Accordingly, one would expect that it is in the interest of the nation -and not merely of the applicants -that the authorities work with qualified interpreters.In view of the lack of comprehensive training for interpreters in an asylum context, however, the reality seems to fall short of this ideal, in Austria as in other countries.
This somewhat absurd situation, where considerable sums of money are spent on the provision of a service that seems to be neither well organised nor accompanied by quality assurance mechanisms, prompted the project team to focus comprehensively on the training of interpreters for an asylum context and national training needs.
Despite the complexity of asylum interpreting (see Section 2), few countries have devised strategic plans that can safeguard high-quality interpreting in asylum procedures (UNHCR, 2010, p. 34).Many countries do not have a standardised approach to training and accreditation, or mechanisms to assure satisfactory interpreting quality (for an overview see Tipton & Furmanek, 2016, p. 86-87).In 2010, a report by UNCHR summarising the findings of a study conducted in several EU countries concluded that "(a)cross the Member States in this research, the provision of training for interpreters is, at best, limited, and in many cases non-existent" (UNHCR, 2010, p. 33).
The consequences of this lack of a structured approach became acutely apparent in 2015.The dramatic increase in the number of asylum seekers revealed serious deficiencies in interpreter provision and highlighted the shortage of trained interpreters, specifically in languages that are less widely spoken or taught (languages of limited diffusion, LLDs) in the host countries.
For some national administrations, these developments represented a wake-up call.Over the last few years, several countries have implemented training courses or scaled up available programmes.However, there is still a shortage of trained interpreters for the rarer languages, forcing caseworkers to employ interpreters with little or no training at all.Moreover, the current training programmes and quality schemes differ in scope and content, and are not internationally coordinated (as yet, no comprehensive overview of recent national training and quality assurance initiatives for the asylum context in the EU is available).One indicator of the global importance of the issue and the corresponding attempts to enhance international collaboration, is the decision of the European Asylum Support Office (EASO) to develop an online training course for interpreters in asylum settings (for details see Section 5) (EASO, 2018b(EASO, , 2018c)).
In the following sections, we will first give a brief overview of the specifics of interpreting in an asylum context, and the research thus far into that field (Section 2).In Section 3, we will review the literature on the difficulties associated with developing and organising training programmes for community interpreting in general and asylum settings in particular.Finally, we will describe the training programme Qualitätsvolles Dolmetschen im Asylverfahren (QUADA, literally "High-Quality Interpreting in Asylum Proceedings"), which was initiated by UNHCR Austria and developed by an expert group comprised of relevant stakeholders in the country (Section 4).

Interpreting in asylum settings
Interpreters in asylum procedures bear an enormous responsibility.The majority of asylum applicants do not speak the language of the recipient country and depend on the interpreter to relay the information they present accurately and completely.Similarly, the asylum authorities -whether caseworkers or adjudicators -must be able to trust the interpreter to provide a rendering that allows them to effectively and fairly assess the applicant's claim (as to "plausibility" and "material truth") and personal "credibility" (UNHCR, 2013).Because applicants can rarely provide written evidence to corroborate their claims, the oral accounts of their experiences generally form the sole basis for the officials' decision (Pöllabauer, 2015, p. 203).Errors, misunderstandings and faulty renditions of speakers' utterances by interpreters may put the welfare and even lives of asylum applicants at risk.Interpreting in an asylum context involves specific challenges that make it a recognisably distinct field; however, for many years it was a neglected area within Interpreting Studies (Pöllabauer, 2006), having been described as a "grey zone" of legal interpreting (Bancroft et al., 2013).Some publications from the 1980s mention asylum interpreting, yet these are mostly personal, anecdotal accounts by interpreters with experience in asylum procedures.From the 1990s, the field began to receive more attention, with a sharp increase in the publication of empirical studies after 2000.Most of the research was qualitative in nature and drew for its analytical framework on a variety of disciplines -including communication studies, linguistics, comparative literature, sociology and law (Pöllabauer, 2006(Pöllabauer, , 2008) ) -with data typically collected through a triangulation of methods including the analysis of "authentic" data (i.e.recordings of real-life interpreting situations).Although the studies typically made reference to general aspects of the asylum adjudication process and global issues, few of the publications that were analysed in a scientometric study in the mid-2000s (Pöllabauer, 2006) went beyond a narrow national perspective (for a review of research within Interpreting Studies (IS) and related disciplines see Pöllabauer, 2015, andTipton &Furmanek, 2016).
Many of these publications -especially the data-based empirical studies -identified and discussed the same or very similar "problems" (Tipton & Furmanek, 2016, p. 86; for a comprehensive overview of research in this field see Pöllabauer, 2015).Interestingly, the same "problems" are still being addressed in recent publications (Lee, 2013;Tryuk, 2017), suggesting that problem-awareness within the scientific community has not led to major improvements in practice and service provision.Among the topics discussed in the literature we find: issues of role and role conflict(s) (often associated with the asymmetrical communication constellation and power differentials present in asylum proceedings), aspects of conversation management and turn-taking, intercultural aspects, aspects of register and style, face and politeness, the caseworkers' investigative strategies, questions related to the participation framework and footing, and problems associated with the use of untrained interpreters (see for instance, Barsky, 1994;Blommaert, 2001;Inghilleri, 2005;Jacquement, 2011;Keselman, Cederborg & Linell, 2010;Lee, 2013;Maryns, 2006;Merlini, 2009;Pöchhacker & Kolb, 2008;Pöllabauer, 2005Pöllabauer, , 2006;;Rienzner, 2011;Scheffer, 2001;Tipton, 2008;Tipton & Furmanek, 2016;Tryuk, 2017).The persistence of these themes was confirmed by the results of a small-scale study that was conducted in 2014 in preparation for the training course in Austria (for details see Section 4).
A further characteristic of asylum settings is the emotionally-charged nature of many interviews (Barsky, 1994, p. 41;Tipton & Furmanek, 2016, p. 82-83).Interpreters need to develop strategies for coping with the emotional impact of having to interpret and "re-tell" stories of victimisation, trauma and torture (called "trauma-informed interpreting" by Tipton & Furmanek (2016, p. 104)), the need to develop heightened resilience, the risk of burnout and compassion fatigue and even vicarious traumatisation (Harvey, 2015;Ndongo-Keller, 2015;Tipton & Furmanek, 2016, p. 104-108) and the challenge of interpreting for vulnerable applicants and unaccompanied minors (Keselman, Cederborg & Linell, 2010;UNHCR Austria, 2018;Wedam, 2018), as well as the absence or near-absence of support structures such as supervision or intervision (peer counselling) for interpreters.
These complex challenges that need to be tackled by interpreters in an asylum context make it evident that (specific) training for interpreters in such a field is desirable.From his perspective as a professional conference interpreter and scholar, Daniel Gile even contends that community interpreting, which he deems "socially far more important than conference interpreting" (2017, p. 246), demands specific interpersonal skills that conference interpreters do not generally require -a view clearly not shared by all, as community interpreting is often associated with low or even negative symbolic capital (Prunč, 2017, p. 25).

Training community interpreters and the special challenges of training interpreters for asylum settings
Training for interpreters in asylum settings falls under the wider scope of community interpreter (CI) training (Bancroft, 2015;Mikkelson, 2014, p. 13-18).The diversity of available CI training programmes, as well as their absence in some countries, has been discussed by Ertl & Pöllabauer (2010, p.167), amongst others -see also Bancroft (2015, p. 225), Hale & Ozolins (2014, p. 218-224), Mikkelson (2014, p. 13), andPöllabauer (2013, p. 5).According to Hale (2007, p. 163), CI confronts organisers of training programmes with a multitude of challenges that can be grouped into four categories: a) a general lack of recognition for the need of training, b) a lack of compulsory pre-service training for practitioners, c) a lack of adequate training programmes and d) considerable differences regarding the quality and effectiveness of training measures.Not surprisingly, Hale concluded that training is "one of the most complicated and problematic aspects of Community Interpreting" (Hale, 2007, p. 162).
The complexity associated with the organisation of CI training programmes is amply demonstrated by the following representative list of problems and the corresponding studies:

•
The high cost of training programmes typically prevents the organisation of full-scale language-specific training, especially for LLDs.A frequent compromise is to offer non-language-specific training which, however, cannot always fully address the participants' (and users') needs and expectations (see also Bancroft, 2015, p. 228;Lai & Mulayim, 2010;Mikkelson, 2014, p. 17;Rudvin & Tomassini, 2011, p. 81-85).
• Where interpreter training for LLDs is to be organised, providers are frequently confronted with the problem that these languages are not taught at traditional interpreting schools, so that few qualified trainers are available (see also Bancroft, 2015, p. 228;Kalina, 2011, p. 55;Lai & Mulayim, 2010), and specific teaching methods may be required or specific group dynamics may be in force (see Hale & Ozolins, 2014;Hlavac, Orlando & Tobias, 2012;and Rudvin & Tomassini, 2011, p. 81-84; for a discussion of the problems related to the provision of training for interpreters for rare or emerging languages).

•
CI training is often also expensive for participants whose remuneration is frequently too low to "justify spending much time and money on professional development" (Bancroft, 2015, p. 228).

•
Potential training candidates may not have an adequate entrance level of language proficiency in all of their working languages (Hlavac, Orlando & Tobias, 2012) and a sufficient degree of cultural awareness.
Valero-Garcés ascribes such "internal" problems to the insufficient "involvement of both higher education institutions and public service institutions and interpreting agencies" (2011, p. 127); she also identifies "external" difficulties, such as long distances to course venues, adverse climate conditions, lack of resources (computers, technical equipment, ICT, adequate classrooms) and so forth.
All these problems also apply to the organisation of training for asylum interpreters.As mentioned above, it is essential to pitch training at the right level and select candidates appropriately.
Currently, interpreters working in asylum settings vary greatly in their cultural and linguistic background, level of formal training, prior professional qualifications, and certifications.A broad distinction can be made between four different groups of interpreters and degrees of training (Pöllabauer, 2015, p. 209).The grouping is based on the situation in Austria, but similar observations have been made in other countries (e.g.Maryns, 2012, p. 309-310): 1. Interpreters with a degree in interpreting from a third-level interpreter training institution, such as universities or colleges.Mostly, these individuals have been trained in languages that are traditionally offered at a tertiary level, which in turn depend on the needs of the labour market, student numbers, the availability of teaching staff, university policy, budget restrictions and so forth (see for instance Rudvin & Tomassini, 2011, p. 81-82).2. Sworn and court-certified interpreters.The extent and level of training received depend on national frameworks for court interpreting and the national degree of professionalisation of court interpreting.Some have undergone full-scale courses, others have participated in shorter courses; some courses offer general training, others court-or asylum-specific content.Some interpreters may not have undergone any training, having simply passed a court interpreter examination or been certified within an official accreditation scheme.In many countries and jurisdictions, court interpreter certification is still not linked to formal training (Lee, 6 2015, p. 192).

Interpreters who have completed shorter pathways, such as partially
institutionalised training (e.g.in-house training), condensed specialist or generalist interpreter training courses offered outside academic institutions and which may have been non languagespecific.These alternatives offer greater flexibility and costefficiency (Rudvin & Tomassini, 2011, pp. 82-83) but have the drawback that language screening, if incorporated, can often only be provided for one language, that is to say the host country's language (Mikkelson, 2014, p. 17).Moreover, time and/or financial constraints may also lead to "compromises in the curricula" (Hale, 2007, p. 169), "creating a sense of complacency in governments and policy-makers who may be led to believe that such courses are sufficient to ensure quality in interpreting services" (Hale, 2007, p. 169; for an overview of a number of monolingual short training courses and possible limitations of such courses, see also Hale & Ozolins, 2014).4. Interpreters with no training in interpreting.This is frequently the case where LLDs are concerned.
Training programmes for asylum interpreting will need to cater to all these groups and their diverse needs.
Which category is employed by asylum authorities frequently depends on how aware governments are of the challenges posed by a complex setting such as asylum interpreting, and the need for some form of quality assurance (Tipton & Furmanek, 2016, p. 84).Some countries screen their interpreters before recruitment (e.g. for political activism, Tipton & Furmanek, 2016, p. 90, or to make sure they have no criminal record) or employ sworn or trained interpreters in preference to the untrained where this is possible.In Austria, for instance, the asylum authorities' internal regulations allow for a "ranking" with regard to the recruitment of interpreters.The first choice is sworn court interpreters; if none are available, interpreters with a university degree in interpreting should be recruited; if no court or trained interpreters are available, any other "language-competent" individuals can be called upon to interpret2 .No official system is yet in place, however, to document the criteria upon which interpreters are recruited (personal communication, Federal Ministry of the Interior, December 2017).
It is difficult to assess how many training courses for asylum interpreting are currently available, as precise data on asylum-specific training measures are scarce (see Apostolou, 2012, on the situation in Greece).The following overview presents the results of a desk-research analysis conducted in 2014, which confirmed the 2010 UNHCR study that identified a lack of training in many EU member states (UNHCR, 2010).As was pointed out above, since 2015, several countries have begun to design and implement interpreter training courses.Details, however, are still difficult to obtain and may at times seem imprecise or obscure.In particular, information about the curricula and pedagogical underpinnings of the courses is generally not available, making it impossible to compare them with the Austrian approach outlined in Section 4. The following overview is, therefore, of necessity incomplete3 .
Information on available training courses is mostly found on the websites of private organisations, charities and intergovernmental bodies.UNHCR, for instance, which has a specific mandate as an intergovernmental global refugee institution with a clear non-political, humanitarian and social mission, has been offering a range of self-study modules (UNHCR, 2009) and booklets (UNHCR, 2005) giving an introduction to interpreting in a refugee context as well as in-house, short-term, non language-specific training workshops for interpreters (internal UNHCR sources).In recent years, a small number of other non-profit and tertiary education institutions have also been providing training for the field, some in cooperation with UNHCR.Among the most salient initiatives we find the following: Some of these initiatives (inZone, Cross-Cultural Communications and Voice of Love) also offer (or have offered) train-the-trainer programmes and/or train-the-user programmes (VOL).Both approaches are stressed as key elements of comprehensive community interpreter training (see e.g.Lai & Mulayim, 2010, p. 59;Rudvin & Tomassini, 2011, p. 23).
an interesting overview of training approaches in the legal realm (mostly at the tertiary level), including online courses (Balogh, Salaets & van Schoor, 2016).

The "QUADA" training design
The project Qualitätsvolles Dolmetschen im Asylverfahren (literally, "Highquality interpreting in the asylum procedure") was initiated by UNHCR Austria and conducted in cooperation with experts in the field (from law, translation and interpreting studies, linguistics, African studies, intercultural psychotherapy) between January and December 2014.The curriculum and content were developed in close collaboration with trainers and researchers at the Department of Translation Studies of the University of Graz.The project was co-financed by the European Refugee Fund and the Austrian Ministry of the Interior.
The main aim of the project was to develop both short-term and longterm measures that could improve the quality of interpreting in Austria's asylum procedures.The urgent need for quality improvements had been identified in a UNHCR project that monitored the national asylum procedure and included areas such as interpreting provision and quality.The findings were confirmed by a needs analysis and a participant observation study (see 4.1).The project scope and activities will be outlined under the following headings: 1) identification of social needs, 2) formulation of outcomes and identification of student profile and needs, 3) design of course content and activities, 4) identification/acquisition of resources, 5) implementation, 6) (design of) assessment and course evaluation, and 7) quality enhancement.The discussion will reference Kelly's model of curricular design for translator training (2005, p. 3), which has also been used for interpreting studies (Abril-Martí, 2006) and was slightly adapted here.

Identification of social needs
In a preparatory investigative and needs-assessment phase, UNHCR Austria conducted desk research to establish the existing types of curricula for asylum interpreter training in Europe and worldwide, and the quality assurance mechanisms in place.These data provided the base for the subsequent needs analysis, which included in-depth interviews with asylum applicants and beneficiaries of international protection as well as interpreters in Austria.
As part of the needs analysis, twelve first-instance asylum interviews (seven initial interrogations and five personal interviews on the substance of the respective applications) were observed (using participant observation) to obtain information on the conduct of the interpreters and other participants.The checklist included predefined quality criteria such as: compliance with professional ethics; interpreting techniques and note taking; and communication, language and completeness of interpretation.
Findings from the participant observation showed numerous problems, including: distortion of information and translation errors; misunderstandings with regard to culturally determined concepts or phrases; role conflicts and role shifting whereby interpreters tried to perform multiple, incompatible activities -either presenting themselves as "cultural experts", "cointerviewers", "neutral language conveyors" and/or "expert witnesses", or being viewed as "helpers" and/or "collaborators" or "traitors".Often, side conversations were not translated, resulting in a lack of transparency for the other participants.Interpreters lacked the necessary (legal) terminology and knowledge of the asylum system, reformulated the original utterances (e.g.change of register, use of a less or more authoritative, bureaucratic or simple code), or adapted the language style to the requirements of the (written) record of the interview.There were also instances of unprofessional demeanour (rudeness, lack of respect, biased behaviour), emotional involvement (e.g.interpreters with a migration background), and problems associated with the complex interdependencies inherent to the communicative situation (interpreters are contracted and paid by the authorities).Even well-trained 9 interpreters seemed to find it difficult to put into practice what interpreting theory and professional codes of conduct expect them to do.
These findings largely coincide with the results of previous studies (see Section 2).The role of interpreters in asylum settings appears to be unclear.Contrary to what is generally assumed, interpreters do not always adopt an impartial, invisible or neutral role (see also Pöllabauer, 2015, p. 207) and can influence the outcome of an interview.
Part of this phase was also the involvement of relevant stakeholders in the field, who were invited to help to identify the social needs for training for the asylum context and the profile and needs of potential trainees to ensure the long-term sustainability of the programme and obtain support for training.These stakeholders included representatives of the Federal Office for Immigration and Asylum, the Directorate General for Public Security of the Ministry of the Interior, the Federal Administrative Court, the Austrian Association of Certified Court Interpreters, and the Austrian Interpreters' and Translators' Association as well as trainers and researchers from the Department of Translation Studies of the University of Graz and the Centre for Translation Studies of the University of Vienna.They were invited to join an expert reference group, which met twice in a face-to-face roundtable and was also regularly updated on the project's progress.
The outcome of this research and needs-assessment phase helped to identify the social needs for an interpreter training programme in an asylum context, and was used to formulate training outcomes, identify student profile and needs, and decide on the course structure and topics to be tackled.

Formulation of training outcomes and identification of student profile and needs
The overarching outcome of the training, which was identified on the basis of the social needs assessment, was to provide basic training for interpreters working in an Austrian asylum context.Specific learning outcomes were identified for the different units (modules) of the programme and outlined in a handbook (see 4.3.) at the start of each unit, basically following Bloom's (revised) taxonomy of learning objectives that is often used for the specification of learning outcomes (Armstrong, 2018).
The needs analysis also indicated that the student profile was diverse and that the training needed to cater to the needs of the four different groups of interpreters outlined under Section 3 (interpreters with a degree and courtcertified interpreters with no specific training in interpreting; interpreters who had completed some kind of short training and interpreters with no training at all).Interpreters with only brief or no training were identified as the groups with the highest needs both for training and for basic information on interpreting and asylum-specific content.Interpreters with a degree and courtcertified interpreters were identified as the two groups most in need of asylum-specific content only.
The analysis of the trainees' profile also indicated that their educational and professional backgrounds and language combinations were highly diverse -two aspects that needed to be taken into account regarding course design.

Design of course content and activities
The findings of the monitoring observations and the results of the needs analysis provided the main foundation for the design of the course content.The developers were also able to draw on the outcomes of two earlier collaborative projects between UNHCR Austria and key stakeholders.One was the Handbuch Dolmetschen im Asylverfahren, a brief manual providing key information for interpreters and caseworkers in asylum settings (Österreichisches Bundesministerium für Inneres (BMI), 2006), and the other Prozedurale Mindeststandards für den Einsatz von DolmetscherInnen im Asylverfahren (Netzwerk Sprachenrechte, 2005), a document setting out minimum standards for the use of interpreters in asylum interviews (see www.sprachenrechte.at).
A modular approach was chosen to give the target group members (who, as mentioned, are highly diverse) greater flexibility and allow them to choose content according to their individual requirements.With regard to the latter, we followed Knowles' (1980) assumption that adults who attend further development courses are self-directed individuals who are driven by internal motivation to obtain new knowledge and skills, and have very specific awareness of why they want to learn (see also Merriam & Bierema, 2014, p. 47) and what knowledge gaps they want to fill.
Based on these considerations, it was decided to produce a curriculum in the form of a training handbook, with suggestions for teaching activities and materials that could be used variously as a trainer resource in a classroom environment, as a self-study tool, and for input into 'train-the-users' workshops -the latter being strongly recommended by many of the interpreters who participated in the pilot and initial training workshops (see also Bahadir, 2017, pp. 138-139 on the benefits of "constructive cooperation" between all participants in an interpreted situation).
The handbook comprises twelve modules which set out the general framework: introductions to legal aspects of asylum and refugee protection, the interviewing techniques used by caseworkers, the interpreter's role in asylum settings, the specific challenges of interpreting for vulnerable applicants, the characteristics of multicultural and transcultural communication in general, and emotional and psychological aspects, etc (see also Annex for detail); it then provides practical guidance on the different modes of interpreting, interpreting techniques, note-taking, and sight translation (for authors see UNCHR, 2015 and UNHCR, 2017).
In line with recent recommendations in interpreting didactics literature, the training design is underpinned by an approach that builds on experiential and situated learning (e.g.González-Davies & Enríquez-Raído, 2016;Perez & Wilson 2011), using authentic scenarios and activities and immersion in professional situations (Perez & Wilson, 2011, p. 251) as well as service learning (Lesch, 2011) to allow problem-based self-reflection and collective reflection (cf.Perez & Wilson, 2011, p. 250; see also Merriam & Bierema's concept of the "circle of teaching" (2014, p. 125)) and collaborative knowledge construction (see Mulayim & Lai, 2015, for an interesting approach to online learning by using the community-of-inquiry framework).
Each module includes a varied range of tasks and activities (for details see UNHCR, 2017) catering to different learning styles and user needs.These can be grouped into four types (following Merriam & Bierema, 2014, p. 125): concrete experience (e.g.role plays, case studies, films, self tests), active experimentation (e.g.role plays, group work, problem-solving activities), reflective observation (e.g.structured discussions, world cafés, films) and abstract conceptualisation (e.g.individual reading, lectures, documents, (flow) charts).
Role plays, in particular, were chosen because they are "hands-on" activities that allow trainers to emulate the dialogic nature of interpreting in community settings (on the use of role-plays for community interpreter training see, for instance, Bahadir, 2011;Kadric's theatre-pedagogical approach, 2011 and2014;Wadensjö, 2014).Learners are invited to actively contribute and learn from each other by critically reflecting on the challenges of interpreting in asylum settings.To support trainers, a chapter was included in the handbook (Kadric, 2018) providing guidance on how to conduct role plays in training sessions and recommendations for how these can be scripted.All twelve learning modules follow the same structure and comprise four major sections: 11 a) Theoretical framework and background: this section provides a concise introduction to the topic.The information is presented in a way that can be easily comprehended even by readers with little background knowledge, with explanations of key concepts and specialist terms in plain language.Some modules provide more detailed information in a separate box ("Compact information/ knowledge").b) Literature and links: this section presents bibliographic references and recommendations for further reading as well as a list of relevant Internet sites.

Identification and acquisition of resources
Kelly's model also allows for the identification and acquisition of resources, for example, through trainer training.
In the first stages of the implementation (see Section 4.5), no trainers' training was provided.Trainers were selected on the basis of their professional expertise (most of them were authors of the handbook modules) and their teaching experience (most of them, though not all, had previous teaching experience, either in an academic or an extra-university context, specifically in the field of community interpreter training).
One fact that soon proved relevant to the identification and selection of competent trainers was that the (national) pool of available trainers was rather small and that institutionalisation of the training would require the recruitment (and training) of additional trainers to guarantee their availability in sufficient numbers and permit more flexibility in planning.
As a first follow-up measure, "trainers' seminars" were initiated; these have been held twice so far and allowed the course trainers (see Section 4.5) to streamline and harmonise content and activities, and exchange experiences and ideas.A special trainers' section was also set up on the dedicated learning platform (see Section 4.5) to allow more exchange between trainers.However, a more thorough trainers' training would be desirable, and is indeed envisaged for subsequent improvements of the training.

Implementation
In the autumn of 2014, a selected number of modules from the pre-print version of the training handbook were piloted at two face-to-face training workshops organised by UNHCR Austria, with the authors of the modules serving as trainers.The workshops were held in Salzburg and Vienna to enable interpreters from all parts of Austria to attend without overly long journeys.Of the participants (33 in total) the majority had no interpreter training but did possess experience as interpreters in an asylum context (therefore representing the "little or no training" classification).They were asked to complete and submit written feedback forms concerning items such as organisational aspects, content (validity and usability), trainer competence, teaching concepts and training methods, and content applicability for interpreting practice.Feedback was very positive overall and was used by the project team to fine-tune the curriculum and handbook.The main aim of these pilots was to assess the validity and usability of the content of the handbook, with the aim of offering full-scale training covering all the modules once the handbook was completed.
One aspect that was considered important from the project's outset was to ensure its sustainability: it was evident that an implementation model needed to be developed that would guarantee the continued delivery of courses after the end of the project period.Since few of the prospective participants fulfil the entry requirements for admission to university programmes, it was decided that face-to-face training would need to be offered in an extra-university context to reach as many potential trainees as possible and to offer lowthreshold access to training.(Informal talks suggested that university-located training also appeared to be "intimidating" to some prospective trainees).
To institutionalise the training, UNHCR sought contact with the Verband Österreichischer Volkshochschulen (Austrian Adult Education Centres; VÖV/VHS4 ), a well-established adult education institution, which agreed to pilot three of the twelve training modules at one of their training centres in Vienna to assess the training's acceptance among the target group.
The pilot seminars, which had more than 60 participants in total, were again assessed by using a written feedback form (individual evaluation of organisational aspects, content (validity and usability), trainers, teaching concepts and training methods and manageability of online training).In addition, qualitative phone interviews were conducted with a selected number (7) of participants.Aspects that were negatively mentioned by a small number of participants were the manageability of the learning platform that was used, and the group's heterogeneity and problems with regard to its dynamics.Neither of these aspects is surprising, as a lack of computer literacy and the challenges of group dynamics have been documented in the literature as challenging factors for CI training (e.g.Mulayim & Lai, 2015;Valero-Garcés, 2011, p. 127).Based on the overall positive evaluation results and the fact that the training was much sought-after by prospective trainees, VÖV decided to continue its cooperation with UNHCR and institutionalise the training (VÖV, 2018).Since 2016, the full training programme has been offered three times in Vienna and once in Salzburg, with plans to offer the course in other Austrian cities.
The course is taught in German and uses a blended-learning format with both distance-learning and on-site sessions where attendance is compulsory; the online phase comprises three teaching units (each 50 minutes), with five teaching units for the face-to-face mode.
The programme is delivered in three thematic blocks (Asylum Procedure, Role and Ethics, and Interpreting Skills and Techniques); these can be taken individually to enhance certain competencies or skills.Each block consists of four modules and an introduction to Moodle, the learning platform that is in standard use by VÖV/VHS.The first face-to-face session is preceded by a two-to three-week online phase during which a variety of activities and tasks have to be completed, including contributions to forum discussions, answering questions on prescribed texts, comments on video scenarios, the compilation of glossaries, and self-reflection.
The course fees in 2017 were €590 per block, which may seem rather high.However, a lower figure would not be economically viable due to the number of modules offered and the number of trainers involved.Scholarships or financial support for trainees is sometimes available from NGOs or employers, though the project team does not have access to information on scope of support provided to trainees by external bodies.Those with language combinations that are currently much sought after (e.g.Arabic-German, Dari-German, Farsi-German) and who are being recruited on a regular basis, can presumably recover some of the course costs through their freelance work for the asylum authorities at fixed rates.For the remainder, who have fewer work opportunities and attend the training principally out of interest, the costs are high and are sometimes seen as a financial strain.
The student profiles for the courses offered thus far have been as expected, with trainees having very different professional and educational backgrounds.The majority have little previous training in interpreting but work in an asylum context on a regular basis.A smaller number have already completed training in interpreting (some have a university degree, some are court-certified interpreters), and complete the course with the hope of future recruitment in asylum settings.Some do the training only out of interest, without hoping or wanting to work in an asylum context.
As of 2016/2017, applicants were required to provide official attestation of having attained a minimum level of B2 in German under the European Framework of Languages.Those unable to produce a certificate were offered additional German language classes at the VHS, custom-tailored to the requirements of QUADA trainees (class attendance was necessary for obtaining the course completion certificate).The German classes run in parallel with the training course.For those trainees who showed sufficient language skill but could not provide qualifications, the decision as to whether they were required to attend the German courses was made individually, based on their professional and educational backgrounds.Overall, the trainees' language skills were highly divergent: some had been living in Austria for a very long time, had perhaps even studied there, and were highly proficient in both written and spoken German; others had only 2 years of experience in learning German but already had experience as interpreters, usually for "exotic" languages that were currently much sought after.

Assessment and course evaluation
Evaluation took place in stages.The handbook (and curriculum contained therein) was subjected to a process of internal review (through the project team) and external review (through external review of each of the modules by at least two reviewers competent in the respective subject area).The validity and applicability of the chosen instructional design were also evaluated by an external expert on adult learning and didactics, who gave a positive report with suggestions for change which were taken up in the final version of the handbook.
The two pilot phases (offered through UNHCR and VÖV) were evaluated via trainee responses, based on written feedback forms (pilot phase 1 and 2) and phone interviews (pilot phase 2, VÖV) (see Section 4.6).
The full-scale programme offered through VÖV as of 2016 is evaluated on a regular basis through written feedback forms.So far, evaluation has been mostly positive.When negative aspects are reported, the project team and trainers are informed at regular intervals and at the trainers' seminar.Suggestions for change on the part of the trainers and training provider were discussed at the trainer seminars (see Section 4.4), which were introduced to permit more exchange between the trainers.
Currently, the effectiveness of the chosen blended learning format is being evaluated as part of a Master's project at the Department of Translation Studies in Graz.Some participants were lacking in IT literacy skills and found the e-learning phase extremely challenging (see also Mulayim & Lai, 2015).
One drawback regarding evaluation of course outcomes is that there is presently no formal assessment of trainee performance in either the online or face-to-face sessions (only attendance and submission of assignments are registered).After completion of the modules, trainees receive a certificate of attendance, without grades or other performance indicators.
Experience in many countries has shown that the prospect of receiving a recognised certificate based on performance assessment can serve as a powerful incentive for prospective trainees to enrol in professional development courses.A standardised performance assessment for each module (including the tasks submitted for the online sessions) would thus be highly desirable and is a future ambition.
One of the project's longer-term objectives is, therefore, to develop an assessment and certification scheme (see also Corsellis, 2009, p. 60;Salaets & Vermeerbergen, 2011 for approaches to certification).An initial step was taken in 2017, when an optional end-of-course examination was offered to all trainees who had successfully completed all the modules.The examination comprises a longer interval of consecutive interpreting, with note-taking, and an oral examination on the content of the modules and handbook.The examiner panel is composed of interpreter-trainers and asylum experts.So far, the certification exam has been offered twice, with a pass rate of more than 50% (five failed candidates from a total of 12).
One positive development with respect to the certification exam is that UNHCR and VÖV have arrived at an agreement with the federal asylum authorities to establish an incentive system.The authorities have agreed that interpreters who have both completed the full QUADA programme and passed the final examination will be preferred over untrained interpreters.It is to be hoped that this scheme will help give participants a clear signal that their personal and financial investment in training is being officially recognisedeven if completion of the course is, unfortunately, not a job guarantee.

Quality enhancement
Following the feedback provided by the trainees and the VÖV team and trainers, it was decided that some modifications would be introduced in subsequent editions of the course to enhance the quality of its design.
Regarding German language proficiency for example, in future the required level will be C1.Participants that have attained level B2 will be admitted, but will have to attend the accompanying German language courses.(If they do not succeed, they will not receive the certificate of attendance.) Another aim is to increase the overall number of teaching units (one unit being 50 minutes) from the current 8 to 12 (including an additional online phase after the face-to-face training).Based on the European Credit Transfer and Accumulation System (ECTS), which can be used to compare the volume of learning based on learning outcomes and associated workload for trainees, the new system would amount to roughly 6 ECTS, including also the certification exams (1 ECTS credit point corresponds to approximately 25-30 hours (60 minutes) of individual student workload).
One development that can also be viewed as a quality enhancement is that the handbook, which initially existed only as a PDF file, is now available in print in both German and English versions.In 2015, following numerous enquiries, Trauner Publishing decided to publish the German handbook as a paperback (UNHCR, 2015), and this has helped to promote its use and also support the sustainability of the project.
Subsequent interest resulted in an initially unplanned spin-off -"localisation" of the German version for an international (or at least European) audience in 2017.As Kelly (2005) stresses, an important aspect of course design is that training does not take place in a vacuum and always has to be adapted to particular local (regional, national) needs.The course design presented in the previous section is clearly focused on specific national needs.After enquiries from different countries in 2017, the project team decided that it was possible to "export" central ideas and elements to other locales.UNHCR provided funding for the translation of the handbook into English and the adaptation of the contents to an international (European) locale.In the English version, all country-specific content has been revised to ensure that the book can be used in different national contexts.In 2018, a paperback version of the English handbook was published by Frank & Timme (UNHCR, 2018) and it is to be hoped that it will further support the dissemination of the content.
UNHCR has also since been contacted by several organisations which have inquired about further translations or adaptations into their local languages.Currently, a French version of the handbook is being finalised by UNHCR Brussels.
An unexpected outcome of the project was the use of the handbook as a reference tool by the European Asylum Support Office (EASO).The handbook, along with its underpinning didactics, was presented at an EASO meeting in 2016.An international project team worked for over a year to develop a comprehensive e-learning training course for interpreters in asylum settings as well as training materials for face-to-face training workshops (EASO, 2018c).A first pilot of the e-learning course will be launched in 2018.We are hopeful that in the future it will be possible to integrate the EASO materials into the national QUADA programme.(The word "hopeful" is used purposely, given that such changes, and general organisational and financial support for enhancing the quality of interpreter provision and training, largely depend on national asylum authorities and the overall political climate and will to act.)

Conclusions
When reviewing the project's evolution from its small-scale beginnings, it can certainly be said to have come a long way: from the original idea of simply drafting a brief curriculum for interpreter training in an asylum context, it ultimately grew into the implementation of a comprehensive course offered at a certified adult education institution.
Other positive outgrowths include the establishment of a certification exam, a first agreement with the asylum authorities on a system of prioritising trained interpreters over the untrained, the English adaptation of the handbook as a spin-off product, and the print publication of the German and English handbook versions.
Against this backdrop, the project can be viewed as a first but important step towards quality enhancement for interpreting in asylum settings.Nonetheless, the present review has also indicated several drawbacks that merit further attention and should be addressed in the future.
One important milestone would be the introduction of a full-scale performance assessment for all of the modules offered in the training (both online and on-site).Since this would entail expanding the pool of trainers and examiners, a train-the-trainers programme would also be a valuable improvement and contribute to enhancing didactic quality.
To achieve full institutionalisation of the training, it would also be important to increase lobbying for its long-term establishment and for support among political decision-makers.The prerequisites for lasting change and improvements are political will and an acknowledgement of the necessity for interpreting quality enhancement: the prioritisation of trained interpreters over the untrained will conceivably incentivise attendance of a training course.If all sides concede that training is an important quality indicator and should be a necessary basis for interpreter recruitment (our experience is that more awareness-raising is certainly necessary in this regard), those who complete it could recoup their costs more easily (and more would be willing to invest in training and continuous professional development generally).
Another milestone, which would need to be preceded by a political decision and the clear commitment to quality enhancement, would be the integration of current European developments.Given the necessary means, existing high-quality resources (such as provided by agencies like EASO) could be integrated into a national training context.Such a pan-European approach, which could be tailored to specific national needs, would potentially be another contributory factor toward quality enhancement.
What must also be examined more fully is the effectiveness of languageindependent training.Feedback that the project team has received from trainers and stakeholders partly confirms the findings of previous studies, and should be researched more thoroughly.Suggested themes include: whether attendance of interpreter training courses could improve the social integration of trainees who are migrants (Lai & Mulayim, 2010); the different dimensions of interpreter identity (Bahadir, 2017, p. 126); the effect and organisation of team teaching; better content provision for trainees from aural or oral learning cultures (Lai & Mulayim, 2010); the advantages and drawbacks of assessment (Salaets & Vermeerbergen, 2011) and accreditation systems (Hlavac, 2015); and trainee expectations vis-à-vis the limitations of short(er) training courses (Hale & Ozolins, 2014, p. 232).
We are well aware that language-specific training is to be preferred where possible, and that the non-language-specific format of the course discussed in this paper is not ideal.Nonetheless, it can definitely address important situational issues, such as the following described by one of the interpreters interviewed for the QUADA project: The authority always expects the interpreter to be on its side.They want me to be as emotional [as they are].But I don't do that.A police officer once asked me to translate: 'Tell him he should not take me for an arsehole.'I didn't want that.So I asked if I could at least say 'jerk'.

5
The feedback and evaluations received from trainees indicate that they have become more aware through the course of the complexity of asylum settings and of their own role within them.When dilemmas of the above kind arise, it is to be hoped that this will help interpreters to recognise and deal with them in an assured and professional manner that befits the process and the gravity of its eventual outcomes.
c) Activities and tasks: this section includes various training activities and tasks.It also includes worksheets that trainers can print out for use in class.d) Individual reflection on learning objectives: this section encourages trainees to reflect critically on what they have learned in the module.It uses a blend of both open and/or closed questions, as well as examples and scenarios that trainees can analyse and reflect upon.