Corps de l’article

1. Introduction

The increasing mobility of labour and people in skilled occupations around the world has increased the necessity of establishing sound regimes for translating and receiving the official documents which accredit or support an individual’s identity, nationality, civil or legal status, history, relationships, qualifications and so forth. For receiving immigration authorities, personal official documents are clearly essential resources when assessing and, depending on outcome, subsequently settling new arrivals in new homes and new work; for intending migrants and foreign workers, they are a ‘make-or-break’ item for their lives and aspirations. Yet, compared with other robustly researched areas of translation theory and practice, the translation of official documents has received rather minimal scrutiny and research. Among the few scattered exceptions, we have the important contributions of Mayoral Asensio (2014), Lambert-Tierrafria (2007) and Źralka (2007), as well as the work of Nord (1997) on theoretical issues of documentary translations.

In this paper, we report part of the findings of a study carried out in Australia on current norms, practices as well as quality and authenticity issues in the translation of personal official documents. The part of the study reported here consisted of eliciting and comparing the views of three main stakeholders: translators, language service providers (LSPs) and public service staff. Before outlining the research methods and findings (Sections 4 and 5 respectively), it is necessary to provide a brief theoretical introduction to the translation of personal official documents and the key norms associated with it (Section 2) and background information about the Australian context (Section 3).

2. Translation of official documents: nature and norms

Official translation, as Mayoral Asensio (2014: 1) notes, is not clearly defined and overlaps with other fields such as legal translation, court translation and community interpreting. While a simple mode-based distinction can easily be made between official document translation and court or community interpreting (one is written, the other oral or signed), fuzzy boundaries remain, particularly between legal translation and the translation of official documents. Mayoral Asensio himself acknowledges that official translation can apply to any text that “falls within a judicial process or a request of acknowledgement of rights before any kind of administrative body” (2014: 4). As such, documents that may require official translation can include genres as diverse as personal letters and text messages or legal contracts. Civil registry certificates, driver licences, police clearances, financial statements and academic qualifications exemplify the typically encountered documents.

One main distinctive feature is that the translation of personal official documents can be characterised as ‘documentary,’ a type-term that Nord (1997) uses to refer to a heavily source-focused kind of translation which involves a significant amount of meta-text description – “the target text, in this case, is a text about a text” (Nord 1997: 47) – which manifests itself in translation practices such as the flagging of official features (e.g. coat of arms, logo, stamp, signature, etc.) or in-text notes or footnotes about aspects which might have significance for the authorities (e.g. presence of handwriting, corrections or erasures, signalling of source errors or inconsistencies with ‘[sic],’ etc.). The documentary approach required for the translation of official documents means that translation quality in this area needs to be scrutinised in terms not only of propositional content accuracy and language appropriateness, but of authenticity and integrity as well. Authenticity and integrity in this context are related: while the former relates mainly to the genuineness of the source document (e.g., whether there are signs of fraud), it is also relevant to translator approach and translation integrity, i.e. whether the translator renders the source text (and meta-text) sufficiently and appropriately for the relevant authorities to assess both authenticity and contents with confidence.

As Toury (1995) noted decades ago, all types of translation are governed by norms, but norms vary from one socio-cultural context to another and from one translation subfield to another. The translation of official documents is often associated with stricter norms, but these too vary from country to country and their status can range from legal norms to conventions and customs (Mayoral Asensio 2014: 1, 4). One clear example is that in some countries (e.g., Spain and Morocco) one norm is that translations of personal official documents should be full translations, while in others (e.g., Australia) extract translations are more common. Toury’s (1995: 56-57) ‘initial norm’ draws a blurry line between contexts and translation approaches where priority is given to the textual relations and norms of the source text (adequacy) and those that adhere to the norms of the target language and culture (acceptability). As mentioned earlier, official translation tends more towards the source text (adequacy). However, this does not mean that translators in this area do not comply with the sociocultural and professional norms applicable in their national or organisational context (acceptability).

In Taibi and Ozolins (2023: 891) we propose a rubric that encompasses the key norms and quality criteria in this area of translation practice. The rubric consists of four main criteria: accuracy, integrity, language and style, and presentation. Accuracy in this context refers to an accurate rendition of the propositional content and intent of the source document, including names of persons, places and organisations. Integrity refers to the (meta-text) description of official and other features in the source document (e.g., stamps, signatures, handwritten amendments), as well as the translator’s in-text notes and footnotes. The criterion of language and style refers to correct language use, which is applicable to all translations, and the use of register and phraseology appropriate for official documents. Finally, presentation relates to presenting the contents of the source document in a clear, orderly and user-friendly manner.

The extent to which the above criteria (especially the last one) can be considered norms will vary from one country to another, and even from one organisation to another. In this paper we are concerned with the Australian professional context of the translation of official documents through the eyes of practitioners, language service providers (LSPs) and receiving institutions.

3. Australian institutional particularities and their translation needs

Australian institutions have long dictated the means of processing official documents so as to meet two main areas of needs – a) immigration and associated documents of settlers and b) business-related regulatory information (tax, business regulation, excise, quarantine, etc.). While business-related norms have been largely met by insisting that clients and companies submit English-language information, immigration-related documents have presented a larger issue. Over the decades and the millions of official documents that settlers have brought with them, Australian policy has gradually changed from refusing many document categories to openly encouraging translating documents for settlement purposes. Martin (1978) recounts the historical reluctance of Australian institutions and professions to recognise foreign qualifications and foreign documentation generally during the period of post-war immigration (from 1947 to her time of writing). This was partly because the migration program had begun as a refugee program, with many settlers lacking clear documentation from their home countries, but was even more influenced by the reluctance of Australian professions to accept qualifications from non-English speaking countries, whether documented or not (Kunz 1988). This blanket disqualification slowly led to critiques of ignoring migrant talent (Jupp 1966), and recognition of the need for government action, which resulted in the establishment of the Committee on Overseas Professional Qualifications (COPQ) in 1968 and the slow but eventually successful persuasion of many Australian professions to accept non-British overseas qualifications.

Translations of official documents had begun to be seen as an area of necessary government involvement in 1960 with the establishment of a translation unit in the Department of Immigration (Hlavac 2022: 69, Martin 1978: 39) which offered a translation service for newly arrived migrants. Part of its brief was to provide translations from English into LOTE (Languages Other than English) for migrants, but it also provided translations of migrants’ official documents; however, there were no recognised standards for translators within the Department. The Department of Social Security also translated client documents for its own needs (Ozolins 1998: 68). At the same time, there were official document translation services offered by small translation agencies, but again without recognised standards for their translators either. In both cases, translations were full renderings of the source documents whose acceptance by the relevant authorities was haphazard and uncontrolled.

As migration boomed over the late 1960s and early 1970s, a push to establish standards for interpreters and translators grew, with active participation from organisations such as the Australian Council of Social Service (ACOSS) and nascent migrant organisations. This impetus culminated in a 1974 report by the COPQ entitled The Language Barrier (Martin 1978: 42), which recommended a national body to stipulate interpreting and translation qualifications and standards, and training programs for the profession. This eventually resulted in the establishment in 1977 of the National Accreditation Authority for Translators and Interpreters (NAATI). This was a pivotal moment for the translation of official documents in Australia, as it introduced the requirement that translations for any government department be carried out by NAATI-accredited (now NAATI-certified) translators. This has since been extended to all public institutions and, increasingly, to non-government and private organisations too. The Immigration Department’s translation unit was later expanded in the 1970s to include a world-first telephone interpreting service, reflected in a name change to the Translation and Interpreting Service (TIS), now TIS Multilingual.

Another significant initiative that affected official document translation in Australia came from the possibilities afforded by the growth of computing from the 1990s onward. Ozolins records that in the 1990s TIS translations evinced “a greater demand for correct formatting and higher standards of presentation from practitioners” (1998: 53). Electronic word processing facilitated the use of standardised templates or extract translation formats, which became increasingly used for certain categories of documents – essentially those related to identity, civil status, employment and educational qualifications (European Union 2016,[1] Lambert-Tierrafria 2007, DiSalvo 1999[2]). This was arguably a significant Australian innovation, as in other OECD countries the use of sworn translators and often highly legal requirements for translation usually demanded the full translation of documents. This practice persisted until only very recent initiatives such as the European Union (2016) agreement on standard formats for a large number of documents related to identity, work, education, civil status and other important documents for labour mobility. In the USA, full translations, certified by individual translators, are still the norm (Comech 2021).[3]

Two institutions in Australia have thus set the pattern for the handling of official documents: NAATI as the unique certification agency for translators and TIS (now TIS Multilingual) as the flagship translation provider. Interestingly, TIS has retreated from its early role in providing public information translations from English into LOTE, and now restricts itself to personal official document translations. Through TIS Multilingual and Home Affairs, migrants can access a Free Translating Service for the translation of their documents during their first two years of settlement in Australia. The now widespread acceptance of NAATI qualifications and the increased number of alternative suppliers and translators has enabled these moves (Hlavac 2022).

To date, no research has been conducted on the practice of translating official documents in Australia, the views held by different stakeholders on quality and best practice or the relationships between those stakeholders. The current study aimed, among other things, to elicit and compare the views of official document translators, language service companies and public service staff regarding various aspects of this area of translation activity, especially in relation to translation quality and integrity.

4. Research methods

In this study, we explored how different stakeholders viewed quality, integrity and authenticity in the translation of personal official documents, and whether the practices of professional translators and language services are consistent with the expectations of institutional end users. Research data were collected through:

  • a survey of freelance practitioners who actively translate these documents (115 translator questionnaires and 16 translator interviews);

  • a survey of language service providers who provide the substantial quantum of such translations (7 respondent organisations);

  • a survey of end users (9 respondent organisations) who process translated official documents as part of their business; and

  • desk research on the institutions which receive such translations.

The three surveys were designed and completed online using Qualtrics. The translator survey consisted of a total of 31 (fixed-choice and open-ended) questions, including demographic questions as well as questions about working languages, types of official documents translated, features of documents and relations with translation commissioners (direct clients and translation agencies). In the three surveys, the main focus was on two broad areas: a) translation quality and b) document authenticity and integrity. In terms of quality, the respondents were asked, among other aspects, about the procedures they follow to ensure the quality of translations and the challenges associated with personal official documents. In terms of authenticity, they were asked to provide non-identifying information about issues relating to the authenticity of translated personal documents. Interviews were conducted with translators only (a sample total of 16) and covered similar questions, with special emphasis on language-specific issues such as language variation, diversity of documentation and document origins, handwritten documents, national practices in terms of document structure and contents, and so on. While the questionnaires covered all working languages, the interviews focused on Arabic, Chinese and Spanish in particular, as these were the languages chosen to be included in the translation assessment phase of the project (Taibi and Ozolins 2023). This choice was based on the fact that these are major community languages in Australia and that we were only able to cover a limited number of languages in the translation assessment.

The three stakeholder groups were recruited by voluntary sampling. Potential translator participants were informed about the project through major language services and other relevant organisations; an information sheet and a consent form were shared with organisation contact persons and translators interested in participating made initial contact with the research team and provided their signed consent form. The only inclusion criteria were a) NAATI translator certification or accreditation, and b) experience in the translation of personal official documents. The LSP group was recruited by identifying the main language services in Australia, and then emailing project information to the principal contact address with a request to nominate or share the information with staff in charge of quality assurance (managers, quality managers, team leaders, etc.). For the end-user group, the focus of the research project required that we identify organisations likely to process translated personal documents: universities, civil registries, the Department of Immigration, the Department of Foreign Affairs and Trade, Australian Federal Police, passport offices, Roads and Maritime Services, Centrelink (social security) and so on. More than 400 organisations were sent information sheets and consent forms by email or post. A total of 9 participated in the survey, mainly from the education sector, but also from a health regulatory agency and a service under the Department of Foreign Affairs and Trade.

The data collected from the three respondent groups (translators, language service managers, and end users) were analysed both quantitatively and qualitatively. For the qualitative analysis, we codified open-ended responses and compared them within the same group and across groups. With the main focus on quality, integrity and authenticity issues and strategies, we identified recurrent themes to have an understanding of how different stakeholders viewed and approached the translation of official personal documents.

5. Findings

5.1. Stakeholder views on quality

5.1.1. Quality criteria

Translators and LSPs concurred in defining what constitutes quality official document translation, differing only in how detailed their responses were (translator responses generally being more detailed and specific). Accuracy and complete reflection of the vital information in the source text were stressed by virtually all respondents; readability and fitness-for-purpose complemented this to stress the purpose and accessibility of the document to those who will need to read it. As one translator put it, “A quality translation is an accurate, precise and transparent translation.” Many technical and aesthetic details were also mentioned: presentation, neatness, layout, and clarity:

In a nutshell, to me quality in translation is basically about the professional presentation of translations and their linguistic accuracy. The final translation product should be of a high standard, and not requiring further attention to improve its quality and presentation.

Translator#77

A great level of detail was specified by some translator respondents:

Official documents when translated from one language to another must reflect the total information held in the document that is translated. Nothing other than essential explanation needed to understand the document should be added. No information whatsoever should be left out. “Lost in Translation” should not happen.

Translator#47

In addition to information accuracy and completeness, the translators stressed the purpose and readership of the translation, as the following example shows:

1. Accuracy of information on the source document i.e. spelling of names, places names, dates, identification numbers, etc. 2. Contains all the necessary information. 3. Meets the requirements of the departments who use the translation in terms of format and protocol.

Translator#89

Another participant expressed this more succinctly:

Concise, Formal, Informative, and Details Accuracy

Translator#114

LSPs concurred that the main quality criteria are “factual accuracy and natural English [or other target language].” They also mentioned other aspects such as “consistent fonts and formatting” and correct spelling and punctuation. One LSP, however, seemed to have an a posteriori, business-minded approach to quality, gauging it by the number of complaints or correction requests from clients: “Quality is defined by the amount of correction requests.”

There was above all a very clear understanding of the documentary approach that is required in official translations, and that this differed sharply from the translation of more narrative texts. Essentially, this means recognising that the constraints of the source text match the constraints of the translation’s intended purpose – for example, translators must give clear descriptions of qualifications, but without attributing any equivalence to them.

5.1.2. Quality assurance strategies

The translators who mentioned quality assurance strategies in their responses focused on three main points: research (into linguistic and extra-linguistic aspects), communication with clients (directly or through the translation agency) and rigorous revision, including by someone else. One of these respondents put it as follows:

I never release the final product until I’m completely satisfied that it meets my quality standards. As part of the process, among other things, in cases where a source document (and its translation) contains a significant amount of detailed information (many personal names, dates, numbers, digits, etc.), I usually get a fresh pair of eyes to do the checking for me.

Translator#86

In relation to quality assurance strategies, some LSPs mentioned they have guidelines specifying the requirements for translators, as well as a quality control department or project managers to check and oversee quality. Others trusted that translators would follow NAATI guidelines, although – it must be noted – NAATI does not have specific guidelines for the translation of official documents. A couple of LSPs also mentioned that they have an internal rating system for contractor translators taking into consideration both their quality standards and cost effectiveness. Based on the said rating system, they “find the most cost-efficient and high-quality translators available.”

One aspect all the LSP respondents agree about is the need for translation checking. As one of the respondents put it, “[p]ersonal document translations undergo a stricter proofreading process because we believe that an error (misspellings and misinformation) in these types of documents can delay our clients’ processing of these documents on their end.” Another one detailed the steps taken:

Prepare draft. Return draft to client for proofing, asking client to particularly check spelling of names, addresses, dates and numbers. For work by translators other than myself, I also check the translation, and correct or recommend changes as necessary, before submitting to the client for proofing.

LSP#2

In relation to this checking process, it is worth noting that, as reported in Taibi and Ozolins (2022), some agencies might not follow an efficient strategy and might not have staff with relevant language skills. For example, they would initiate the revision process after the translator has printed, stamped, signed and scanned the source document and the translation. When there is an error to be corrected or a note to be added, “then the translator has to go back to it, print, stamp, scan, etc. all over again which is a waste of time and money, after all.”

For the institutions that process translations of official documents as part of their business, quality assurance revolves mainly around setting eligibility criteria for translators whose translations can be accepted. As one such end user put it, “[t]ranslations prepared by persons familiar with the language of origin, including relatives, friends, acquaintances or other volunteer agencies, are not accepted.” Inside Australia, the main, if not only, eligibility criterion is NAATI certification at a professional level or above. Translations completed overseas, as one institutional end user said, “will only be accepted if completed by an officially recognised translator and is endorsed with their full name, address, telephone number, job title and details of their qualifications.” When vetting the credentials of Australia-based translators, end users often consult the online NAATI directory of certified practitioners. Although it is difficult sometimes to establish credentials of translators based overseas, there are resources like the International Federation of Translators (FIT) which some local end users resort to.

One organisation mentioned that their employees who process translations “are trained to understand the quality requirements for translated documents as well as to identify those that do not meet those requirements.” Another referred to “proof-reading, search reference in original language,” which obviously can only be used provided there are qualified staff with relevant language skills. Education organisations added the requirement of full rather than extract translations as another quality assurance measure:

Extract translations of degrees, diplomas, certificates and transcripts are not accepted. All translations must be prepared in accordance with our guidelines for certifying documents. We require the original certified copy of the document used to prepare the translation, the translation, and the original signed statement in English. Translators must provide their full details which will be verified online (NAATI website).

End user#1

Another one, on the other hand, acknowledged they did not have any quality checking procedures, except when bilingual staff are available: “However, for languages where we do not have native speakers there is no procedure to check for quality.”

5.1.3. Challenges in ensuring quality

Just under 65% of translator respondents replied that they did not have difficulty meeting the requisite quality standards as outlined. Among the third who did indicate difficulty, the overwhelming response attributed this to source document quality: poor handwriting, illegible passages or official stamps, overwriting, erasures, deteriorated physical condition and poor copies were the features most commonly reported as making quality translations difficult, especially given that documents might date back half a century or in some cases even more. A few of the translator respondents also recognised challenges relating to their own extralinguistic and terminological knowledge.

In regard to linguistic and cultural translation issues, there were far fewer concerns about quality, though there was some mention of incompatible structures or administrative arrangements between Australia and the country, region or institution issuing the document, including religious organisations. This area also included technical aspects such as incompatible military ranks, education structure, civil registry forms or other certificates. With modern enhanced research capacities through the internet, these issues could for the most part be solved, but not always. Again, translators would determine potential issues by reference to how local (Australian) officials – who will presumably lack the knowledge implicitly assumed of readers in the native context – might comprehend or assess obscure institutions or practices referenced in translated documents. The actual consumer of translated texts is almost invariably an unknown quantity too, or as one respondent put it, “you don’t know who would be reading your translation.”

The demands of clients also received some mention and had convinced some translators not to take on direct commissions (from individual clients), though for the majority this continued to be the major source of their work in official documents. For a very few, agency practices could affect quality – in some cases because the clerical staff employed by agencies to manage translation projects could not respond to translation issues; in others, because agencies had particular house styles (which differed from agency to agency):

Agency editors sometimes make minor changes to my work, some of which I disagree with. One example is that they say I should not write ‘female’ for the gender if it is not explicitly stated, even if the surname ends in -a in Russian, which is a female ending. I accept changes unless they distort the meaning and do not argue.

Translator#84

While this was presented as a translator-agency issue, it actually raises more theoretical issues of whether translators can make explicit information that is implicit in gendered grammar.

Agency respondents stressed how they often worked together with translators to solve quality issues but also saw this as involving the client:

After translation, we usually send a draft first to the client to confirm if all details (names, dates, addresses) are correct. Upon confirmation from the client, we then provide the final document.

LSP#1

Half the LSPs surveyed had no specific guidelines for translators; some referred to “NAATI guidelines,” which may mean the best practice guidelines of the Australian Institute of Interpreters and Translators (AUSIT 2022)[4]. Where there were guidelines, some were brief (e.g., “[t]ranslations must be stamped and certified”) or had templates including disclaimer statements. Only one service provider had its own guidelines which were “language and document specific.” Service providers were divided on whether further guidelines would be helpful.

Human resources were also identified as a challenge for LSPs, especially in new and emerging languages, where the numbers of qualified translators are limited:

[re] difficulties meeting quality standards, I have had a few letters which used a lot of Congolese Swahili, usually accompanied by translations made by the client or a friend of theirs. This was challenging when there are only 3 NAATI accredited Swahili translators listed on the NAATI website, and none with Congolese Swahili, yet the client needed an accredited translation. I adapted their provided translation which did help me understand the Congolese Swahili letter, and sent to a Congolese colleague to check if my translation was OK.

LSP#7

LSPs were asked to comment on several aspects of quality. First, regarding translators’ English proficiency, the LSPs overwhelmingly stated this was not a problem in official document translations, with a few citing the only concerns were with new and emerging languages, where the response was to try to increase the pool of translators in those languages. When asked about errors detected in translations, the overall response of LSPs again was that this was rare, with spelling of names being the most common of the few cases observed. Other errors mentioned were those arising from handwritten documents, cut-and-paste typos (when translators deal with several documents or templates at once), omissions/oversights or “simple human error.”

Our survey of end users (public services), however, found several quality issues. One educational provider listed several concerns, including:

Mistranslation (literal translation): confusing “graduation year” with “year of study before graduation”; Not translating all the document: omitting footer; Not using a standard term: “Diploma” (translation) should be “Degree”; Misspelling words: name, date of birth.

End user #2

Another concurred and added that “the variation of key words/descriptors could potentially impact on the outcome of an assessment.”

Educational bodies in our survey often used available native speakers among their staff to check on quality or integrity issues, covered below. Where end users did have translation guidelines, issues tended to arise when these were not followed; this could result in work needing to be re-done, particularly for some translations sourced outside Australia. One regulatory authority with strict guidelines on full and certified translations commented that

Our applicants usually complain about the difficulties they encounter arranging for the documents to be translated in accordance with our guidelines, rather than the quality of the translations.

End user #1

For end users, the challenge was “to establish the translator’s eligibility requirements (certified translator), which includes their full name, address, telephone number, job title and details of their qualifications.”

5.2. Stakeholder views on authenticity and integrity

5.2.1. Concerns over authenticity

Translators, LSPs and end users were aware of authenticity issues, and likewise aware that often it was difficult to know for certain the status of documents presented. Of the translators surveyed, 35.71% reported having had concerns over the authenticity of some documents assigned to them, though the incidence of such occurrences was rare, most reporting only one or two such cases. When it came to dealing with authenticity concerns, fewer than a dozen respondents reported rejecting the work (usually from direct clients) and a similar number had accepted it but included a disclaimer regarding authenticity; two translators had reported to the particular authorities who were likely to receive such documents. The most common action among translators was to report authenticity concerns to the translation commissioners. This practice was supported by the surveyed LSPs, who stated they mostly relied on their translators to alert them to such problems, after which the relevant solutions, including all those mentioned here, were undertaken jointly.

LSPs also reported a few cases of authenticity issues, as in the following example:

Tampering of original documents, changing personal statements documents to reflect the ‘intention’ of the writer rather than what the writer wrote. Interestingly, this was mostly found in documents used for the court cases and where the individual stood to lose (wills and testament or affidavit statements).

LSP#3

Among the end-user respondents, only one reported such document authenticity issues. The others were not aware of any such issues and relied on clients and translators to submit trustworthy documents/translations. One end user commented: “We do not authenticate the validity of the documents. We require certified copies of the original documents”; another said: “we presume the translated documents are authentic.” Under a different section, one tertiary education provider also mentioned that they contact issuing institutions to verify the authenticity of qualifications.

Typical responses from those translators who had occasionally queried authenticity stressed that it was not up to the translator to determine authenticity, as often there could be ambiguous circumstances surrounding a particular document where there might be suspicions: “But even then, the translator cannot know if the issuing officer made an honest mistake, if something had changed in the client’s details or if there is a valid reason for different details.” Where problems were clear, however, or requests for certain changes were invalid, translators did act, for example, by refusing to translate an incomplete document or adding notes about unusual features in the source document. On rare occasions, concerns over authenticity have led translators/agencies to contact particular authorities. As one large translation company reported,

Depends on the level of deceit - we have on times just advised the client straight out and refunded the money and we have also alerted Immigration/border for more serious matters- Most of our translators will alert us to their suspicions about the document and why and then referred onto the director for final decision.

LSP#4

More commonly, issues related to document integrity revolved around previously covered aspects in regard to unclear copies, alterations, obscurity of stamps, seals or parts of text, with solutions often worked out in communication between agencies and translators.

5.2.2 Translation format and integrity

A central area in the translation of official documents which relates to both quality and authenticity is the completeness (full vs. extract translation) and integrity (as defined in Section 2) of the translation. The translator respondents were very aware of how critical such translations were for the holders and for the receiving institutions. The risks involved are exacerbated by the complexity of many types of official documents, including financial and property documents, medical records, and military certificates. Some complexities came from language–specific circumstances: birth or marriage records in some countries ran up to five pages; religious documents presented issues of their own, and police clearances varied greatly within the same language, in those cases (e.g., Spanish or Arabic) where the language is spoken in a large number of countries. Language-specific or culture-specific peculiarities which differ sharply from Australian or English-speaking norms included the Spanish-language use of two surnames (paternal and maternal) with occasional inconsistencies in use of surnames across different documents, while Arab culture does the opposite with documents not stating surnames at all in certain cases. Calendar systems can also differ and dates may be defined obliquely (e.g., formal locutions such as “in the current month”). Education systems will also differ, such as the ‘tertiary vocational education’ or ‘middle-level specialised school’ educational institutions in China.

Translators in the Australian context and marketplace will typically seek to provide accuracy but also economy and succinctness, and these can potentially be conflicting goals when taking the desirable ‘documentary’ approach to official translations. For this reason, annotated extract translations are a common and indeed preferred presentation for many kinds of documents and purposes – the standard civil registry documents in particular – while full translations remain an option if and where specified.

Our survey and interviews brought to the fore the issue of extract translations versus full translations and how this was viewed by translators, clients and translation agencies. Just over half of the survey participants (53.04%) expressed satisfaction with extract translations and saw them as a much more efficient way of conveying information. Extract translations are also less costly for clients, as several translators and one LSP noted.

However, there was concern over this issue in three directions. First, the translators all mentioned that while some receiving institutions make clear which format they prefer, this was not uniform, leaving translators guessing and sometimes not able to advise clients. Such a concern was also apparent in the responses provided by LSPs, who often did not receive specifications from their clients, thus needing further communication with clients or relying on the translator’s advice, as expressed in two short extracts from the LSP survey:

Find out from the client for what purpose the documents are needed, then only translate the information pertinent to that purpose. Only rarely would do full translations.

LSP#2

It would be relevant to the intended purpose of the client and the department they are dealing with but some languages that have a lot of family information can be condensed - […] and mostly a lot of banking transactions. Most translators we use can specify if they believe it would be more beneficial.

LSP#4

Translators clearly saw a need for some overall uniformity in this area, with time and effort wasted in trying to guess or affirm requirements and different translators making different decisions on formats. One interview participant summarised a broadly held view:

I would like to see a standardised [format] being used for these sorts of documents, like birth, death, marriage and license and ID documents and passports. Those are the things that you really want, details titles you’d want. You don’t want all the extra language you just want the detail. I’d love to see that.

Translator interview#7

An associated concern here was where translations were intended for overseas and where extract translations were much less common, and some respondents mentioned that this international dimension needed to be addressed by the Department of Foreign Affairs and Trade and NAATI to ensure requirements in both directions were consistently understood.

Secondly, the nature of some kinds of documents, and particularly the purpose for which they would be used, can be crucial: a document for family identity may be able to have an extract translation, but not if the document would be used in Family Court proceedings:

Now, in case we are going into the family court and we are seeking certain settlements or, I don’t know, related properties to children’s custody and so on, in this case, I cannot but translate it in full, I cannot provide an extract. Because it’s every word can weigh and it’s important for the case because now I’m dealing from a different legal perspective.

Translator interview#14

Thirdly, translators reported that many clients were uneasy with extract translations, not having ever seen such formats, and believing that not only essential information but also the essential official status of the translation may be in question in extract form – this is why there was often considerable communication between translators or agencies and clients to ensure client needs were satisfied. As one agency reported:

We always confirm with the clients that unnecessary information will be excluded and only important details will be translated.

LSP#1

Cultural differences could be crucial. As one translator succinctly put it, “Spanish speakers want paper!” and this translator will always provide a full translation, even where an extract may be acceptable by a receiving institution. This led to views of how the advantages of extract translation in many contexts might need to be thought about to make them more acceptable to sceptical clients who might regard the translator as doing a less than complete job:

Yeah, I think the templates could perhaps be updated and improved. Just so they look a bit more, I don’t want to say legitimate, but perhaps so the clients might be more likely to accept an extract translation if it looks more official and they can trust that it will be accepted by the authorities.

Translator interview#11

Translations done for the Department of Home Affairs come with the departmental letterhead, which does lend them a “more official” appearance, but extract translations done for direct clients will not have the imprimatur of any authority beyond that of the translator’s certifying stamp.

5.3. Translator strategies and responsibilities

Translators have a limited number of strategies when they feel issues within documents they are translating are difficult to convey to the target reader – namely, the officer in the relevant receiving institution. Three strategies are: a) reverting to full translations (See 5.2.2 above); b) encouraging the client to make their own clarifying representations to the receiving institution; or c) supplying translator notes (a common practice in some other areas of translation but drawing a mixed response from the respondents where official documents are concerned). For some, the inclusion of translator notes is a necessary step where a document may be misleading:

In my opinion, it only adds to the clarification and, of course, if he feels that the clarification can add to the complication, then that is his responsibility how to word it. Or if the client feels that they’re not happy with the note it is up to the client to seek further clarification.

Translator interview#6

For others, however, such notes are seen as a last resort, or even a sign of less ability in translation:

In my writing anyway, I don’t use footnotes, I just try and make my writing so clear that he doesn’t need an explanation. It’s not to say that that there’s no place for them but it’s just I feel like it’s a show of confidence in your translation if you can do it without having to then go on to justify the choice that you’ve made.

Translator interview#2

Some respondents following this line of reasoning offered the view that making notes is a sign of “weakness” of the translator, who needs to be able to convey accurately the meaning of the source text. This decision is made easier in the case of extract translations in that one of the features of many templates is a space designated for “Additional essential information” or “Notes,” as well as space for the obligatory NAATI stamp. Meanwhile, our sample of LSPs all encouraged the use of notes as a means of clarifying often cryptic translations of particular terms.

An important view of many of the respondents was that ultimately for a document having educational or employment consequences, the translator can only do so much in expressing equivalences in the translation. If the client is unhappy with the way something is translated, as it may not convey a strong enough sense of qualifications equivalence, they need to make representations themselves to the receiving institution about their qualifications, or to obtain advocacy through other means. One interview respondent saw this as not only a matter of tactics but also a question of the fundamental loyalty of the translator:

We don’t have a choice but to render it in a way that is faithful and loyal to the source text. But then I guess such observation or comments are actually better done by another advocate, not a translator because the translator is always, you know, the loyalty always lies with the assignment rather than with the client requesting such assignments.

Translator interview#10

This view was strongly supported by other respondents who set limits on the extent of the translator’s involvement on how a translation will be read or what use will be made of it to advance the cause of the particular client.

Yet, this did not mean that translators’ loyalty was only to the source text no matter how the reader will understand it; it is significant that the loyalty is seen as not to the source text per se but to the assignment, and here the purpose or skopos of the assignment, to make the nature of qualifications or achievements clear, figures strongly in the thinking of the translators. Overwhelmingly, the concern was to produce a communicative document and one which could convey most clearly the particulars of the source text to an audience which does not have an understanding of the source language, source culture and source institutions which produced the document. Thus, the loyalty of the translator is also ultimately to the reader.

This represents, of course, the balance of bilateral interests and loyalties that all translators must negotiate (Nord 1991): the need to produce a meaningful rendering for those without knowledge of the source, while remaining mindful – particularly with official documents – of potentially high-stakes contexts where clients may exert pressure to state equivalences beyond what the document would warrant (Mayoral Asensio 2014: 12-13). A key issue here is the recognition that it must be the receiving institution, not the translator, who ultimately determines qualifications equivalence.

Thus, regarding qualifications documents, it is not for the translator to decide on the status – in our case, the equivalent status in Australia – of a foreign qualification that the client brings. From the earliest days of COPQ to the current operation of the now numerous educational and professional bodies that oversee international qualification recognition, the concern has been that it is the specific educational bodies and professions that will ultimately determine this status; the translations provide the raw material for considering equivalent status but do not of themselves declare it. This was universally recognised by the translators, who expressed this often in strong terms; it was the ultimate justification for their relations with clients and relations with the receiving institutions.

This view was supported by the surveyed educational and employment institutions, who see the task of the translator as very much to provide clear information on the actual grades, awards or occupational tasks achieved or positions held by the applicant, with the high-stakes judgments of equivalence being the preserve of the receiving institutions. This is also the clear recommendation of the AUSIT guidelines for the translation of official documents, to which we now turn.

One area in which there is clearly a good deal of discrepancy among translators is the authentication of documents. In relation to source material authenticity, translators in both the survey and interviews recognised they could sometimes meet with false documents, but as a rule did not see source vetting as their primary task – limiting themselves mostly to making notes to the end user of the translation if something was clearly suspicious or out of order. Rather than worry about authenticity, the most common practice among translators is to translate what is in front of them: “For me, I don’t really care, but I wouldn’t really dig into the authenticity of this document. So, I’m the kind of person who just translates.” The AUSIT Best Practices identifies this as an important area in which translators must protect themselves:

20.3 The translator may choose to add a disclaimer underneath the certification (e.g. in smaller font) as to the authenticity of the source document. Example:
The translator, in providing this certification, gives no warrant as to the authenticity of the source document. Any unauthorised change to the translation renders this certification invalid.

AUSIT 2022: 9[5]

6. Discussion

The three surveys (of translators, language service managers and end users) suggest an overall agreement on the need for the translation of official personal documents to meet high standards of quality and integrity. There was a general agreement about risks involved in this type of translation, and the need for translations that are accurate, linguistically and stylistically appropriate, and documentary (describing the official features of the source document). However, a number of issues and uncertainties were identified, including the existence of mistranslations, incomplete translations and descriptions, the risks associated with extract translations, lack of a clear and robust quality assurance mechanism, and lack of clear guidelines from some language services and institutional end users.

The translator participants are mostly confident that they are able to meet the quality requirements (nearly 65%) and the language service managers report that errors and quality issues are rare. Some of the institutional end users, however, do report quality and integrity issues, such as mistranslations, incomplete translations or factual errors. Our quality assessment of a corpus of translations of personal documents (Taibi and Ozolins 2023) also confirms that, while most of the translations demonstrated a high quality standard, approximately 15% of the corpus was suboptimal. The shortcomings identified included incomplete translation, inaccuracies, failure to document the source text (i.e., type of source text, source language, official features, etc.) and language appropriateness.

Quality in this area, as in other areas of translation, revolves around two key factors: translator training and translation checking. In another paper (Taibi and Ozolins 2022), we reported that 64.34% of the translators surveyed had not completed any training, including professional development activities, on the translation of official documents. As for translation checking, while translators and language services are aware of its importance, practices in this regard may vary. Freelance translators will typically check their own translations and, when there are oversights, will only be made aware of them when their direct client reports back. Language services for their part would need a clear, efficient and consistent other-revision process, but may be hampered in this by a shortage of staff with relevant language skills. While not uniformly available, mutual revision as a quality assurance step may still be implemented in some circumstances, such as languages with an appreciable pool of translators, for example.

A third key factor may be added, which consists of clear guidelines from both language services and end users in relation to the requirements of personal document translations. As pointed out above, half the participating LSPs did not have guidelines for translators. Although the Australian professional body has specific guidelines for the translation of legal and personal documents, these did not emerge as a well-known or commonly used reference. This best practice guide covers 20 sections, including recommendations on issues covered above (notes, formats, education records) but also highly detailed specifications for certification, attestation and authentication, person and place names, numbers and dates, translator’s certification, and handling of specific features such as stamps and seals.

TIS Multilingual also provides useful information about its translations of official documents on its website, including a section entitled ‘What will the translation look like’ (Department of Home Affairs 2021).[6] In relation to the issue of full vs. extract translation, the same website advises:

All translations appear on an official letterhead and with an official stamp. The translations will be provided on a template and will be in extract summary format (up to 100 words).

TIS then makes exceptions for a small number of documents, including:

  • custody and medical documents, which will be translated in what it calls “word for word format” up to five pages per document

  • vaccination documents in extract summary format up to 5 pages

  • academic transcripts in extract summary format up to 5 pages per document

  • employment-related documents in extract summary format up to 600 words.

Regarding academic transcripts, TIS advises clients that if receiving institutions demand complete word-for-word translations, these will be done free up to a limit of 10 documents. This is a significant caveat, since in education and the related area of recognition of overseas qualifications by professional associations there is a considerable divergence of practice in terms of requirements regarding document verification, translation formats and translation supply.

As for end users, it is clear that guidelines are not consistently available and requirements vary from one institution to another. The Department of Foreign Affairs and Trade,[7] for example, clearly advises clients on the requirement of full translation, in addition to other details such as eligible translators and translation services, and legalisation of documents. In the higher education sector, the University of Sydney states that documents in a LOTE “must be accompanied by a complete English translation made by an accredited translator with a government body or the Registrar of the issuing institution”[8]; the university further specifies additional requirements for documents from China but makes no mention of whether source documents must be certified. The University of Adelaide,[9] on the other hand, gives no information about translation in its public website, but does give extensive advice on how to have documents certified. TAFE NSW (Technical and Further Education) demands certified copies of academic transcripts and English test results but does not specify any translation source or format, advising only that “if your documents are not in English you’re required to provide official certified English translations of your documents.”[10]

This small sample shows there is not always a publicly expressed view about the format of translations, an observation made repeatedly in our survey and interviews of translators. Where educational bodies do refer to translation, the recommendation to use NAATI certified translators is universal. Among the university international offices and regulatory agencies in our survey, full translations were preferred in some cases and demanded in others. Generally speaking, however, extract translations are often submitted and accepted.

As mentioned above, policies are not always explicitly communicated and, when they are, they are not sufficiently specific. The most significant exception to this is the Australian Health Practitioner Regulation Agency (AHPRA), which assesses foreign qualifications across the range of health professions. AHPRA has probably the most detailed guidelines on translation of any institution in Australia[11]. AHPRA stipulates that all translations must be full, not extract translations. It allows translations to be done overseas, but only by members of associations affiliated with FIT, and reserves the right to have documents translated again in Australia if the overseas translation is deemed unsatisfactory.

If the translation is performed in Australia, all work must be undertaken by NAATI certified translators. Translators working directly for an applicant must also supply additional information about themselves and a declaration of accuracy and must work from originals or certified copies. Where AHPRA receives foreign language documents with an assessment application, it requires the applicant to find an accredited translator. The subsequent exchange occurs between AHPRA and the nominated translator (once their status has been verified): AHPRA forwards the document directly to the translator, who completes and returns the translation to AHPRA.

These are possibly the tightest controls over translations of any government authority; most other areas of government (e.g., immigration, social security, local government, road traffic authorities) are happy to receive extract translations or do not state any particular requirements.

7. Conclusion: Unresolved issues and ways forward

In this paper we have reported the views of three key stakeholders in the industry of translation of official documents: translators, language service managers, and receiving institutions. The three surveys have revealed overall agreement on the high-stakes nature of these documents and the need for rigorous translation and quality assurance mechanisms. Examples of good practice identified include (for translators) pre-translation research, rigorous compliance with accuracy and style requirements, and self- and other-checking; and (for language services) providing guidelines, close collaboration with translators to ensure quality, and having checking processes in place.

However, the data also shows that translation of official documents, while providing a significant service to clients and receiving institutions, is characterised by a number of unresolved issues. The unique features of the material itself pose a significant difficulty in terms of policy initiatives, given the divergent administrative systems the documents come from and the range of styles they use. However, even here clarity over what is necessary/optional to convey in official documents may be of benefit. Unresolved issues pertaining to the policies of both government and receiving institutions seem more amenable to some regulation, though again the range of different institutional needs can be daunting.

Some of the inherent uncertainties could be resolved if institutional end users were to establish and communicate their requirements (e.g., extract or full translation); uniform guidelines across the translation sector would also be helpful. However, in this field we face a triple complexity that defies easy solutions:

  • Lack of standardisation of the official documents brought to Australia, even within one language and from one country;

  • Lack of standard criteria for acceptance by receiving institutions – while some state requirements explicitly, others have no publicly available guidelines or may indeed have no policy, accepting documents in any format;

  • Lack of uniform practice among practitioners, at least partially mirroring the lack of source document standardisation.

Australia has evolved a system for the provision and acceptance of official document translations that does its best to resolve the abovementioned difficulties and provide satisfactory outcomes for clients. Further improvement will come from greater adherence to guidelines such as those of AUSIT to bring about a uniformity of practice. Notwithstanding the still landmark authority represented by the Department of Home Affairs, the issuance of uniform guidelines across individual state and federal jurisdictions would assist receiving institutions, make the work of translators and translation companies more predictable, and also provide understandable guidelines to individual clients regarding what they can expect from a translation. However, different professions enjoy a great deal of autonomy in setting their own requirements on documents and translations, thereby frustrating any easy path to uniformity.

For the above reasons, the promotion of clear, explicit and accessible information on translation approach, formatting and style would be highly desirable. In a complex system of public, outsourced and private bodies, overlaid by a federal system of government, it is clarity of norms, rather than necessarily uniformity of practice, that emerges as a priority.