1.1 Testing Tradition vs. Assessment System

Generally speaking, in the historical development of the discipline, a distinction has been made in the related literature between language testing and language assessment in terms of their definitions, foci, purposes, and scopes (Bachman and Damböck, 2018; Cheng and Fox, 2017). Initially, the primary focus of assessing language proficiency tended to be on tests as the sole measuring instruments. In the new millennium, however, the focus has shifted away from language testing to language assessment, which encompasses a broader range of multiple evaluation purposes, methods, and measures beyond standardized tests. The two systems and their defining characteristics are briefly introduced here.

Language testing refers to the tradition of measuring the language ability, competence, or proficiency of an individual, typically dominated by using paper-and-pencil tests. In other words, it is used specifically for the process of measuring individuals’ language proficiency through standardized tests (Bachman, 1990; Bachman and Damböck, 2018). It typically involves the administration of formal tests or examinations designed to measure specific language skills or overall proficiency according to predefined criteria. Conventional language testing often focuses on producing numerical or categorical scores or ratings that can be used for purposes such as certification, admission, placement, or employment. They tend to emphasize the measurement aspect and may not always encompass the broader goals and practices associated with language assessment. By providing one single sample of a student’s performance in a specific domain on the measure, these tests are designed to measure their learning achievement status at a particular point in time. They are usually administered after the instruction process ends under strict formal procedures and time limitations.

On the other hand, language assessment, in its broadest sense, is used for the entire discipline and its tools. It refers to any formal and informal processes of systematic gathering of information for the purposes of judgments about individuals using tests as well as non-tests, whether done quantitatively or qualitatively and used to make decisions or not (Brown and Abeywickrama, 2019). As defined by Farhady (2021: 123), it is “an ongoing process of collecting information about the students learning”. In brief, language assessment includes various assessment practices aimed at promoting learning, providing feedback, and supporting language development; language testing can be considered a subset or one aspect of language assessment.

Traditionally, language/translation testing has been seen as an outside force imposed upon the curriculum generally and the learner specifically. In the last century, the prevalent use of standardized paper-and-pencil tests in academic contexts was universally taken for granted with no serious objection. The approval of these tests by many shareholders was largely due to the relative efficiency with which a large number of students on multiple learning outcomes could be measured. Moreover, they are assumed to be inexpensive, easy to score, and simple to record and interpret the results. In sum, due to their optimal practicality and high reliability, conventional tests have typically been welcomed by practitioners in the field (Salmani Nodoushan, 2008). Nevertheless, due to the dramatic changes in assessment purposes, methods, and measures, these tests have been criticized for the way they measure student achievement.

1.2 Criticisms against Conventional Tests

Conventional tests are still extremely popular in educational contexts as they enjoy so-called high reliability and optimal practicality. They are cost-efficient, relatively easy to administer, and score in a little time; in addition, the results are also rather easy to report and interpret. The results are said to be ‘objective’ in the sense that personal biases and values of the scorers do not affect the scores, or at least their impact is minimized. As the main – and usually the only – measuring instruments, conventional tests enjoy uniformity in all students taking the same test, which adds to their objectivity. Used as typical gatekeepers, they are accepted as a standard norm in educational programs in general and translation programs in particular.

However, severe criticisms have been leveled against the widespread use of paper-and-pencil tests. According to Brown and Abeywickrama (2019: 16), “In the public eye, tests have acquired an aura of infallibility in our culture of mass-producing everything”. In fact, the test results tend to be accepted by many ‘uncritically’ (Bailey, 1998). Surprisingly enough, most telling criticisms commonly arise from the same characteristics contributing to their benefits. As discussed earlier, these features can be summarized as one single sample taken at the end of instruction under time limit pressure and formal procedure. It is rightly proclaimed that one single sample of performance is not enough and may not be representative of the real performance of an individual in different contexts and occasions. If the student has a bad day (like an illness at the time of the test), it sticks with them! On the other hand, there are also deep concerns about potential cheating, coaching to the test, test-taking strategies, and test-wiseness regarding students taking a test successfully without having adequate knowledge of the subject matter in question. In both cases, the sample taken is faulty, and the decisions made on the basis of test results are under serious question in terms of reliability and validity. In sum, conventional language tests cannot serve as true indicators of how students might perform in real, authentic situations.

Moreover, tests tend to provoke adverse feelings such as fear of failure, stress, anxiety, and fatigue more as they are administered under time limit pressure and formal procedure, thus impeding cognitive abilities. The main point, as explained by Brown and Abeywickrama (2019: 1), is that traditional tests “have a way of scaring students”, whereas language assessment practices “need not be degrading and threatening” (ibid: 2). Traditional testing tends to act as an “add-on” to instruction and a crushing burden to students in terms of workload and adverse consequences. With the same token, another criticism comes from the negative washback effects traditional testing has on teaching. The emphasis on preparing students for standardized tests can lead to a “teaching to the test” approach, where educators prioritize test-taking strategies and content coverage over deeper learning experiences. The authoritative power of standardized tests is said to force teachers to narrow their instruction to certain discrete points so that students are prepared for the mandated test, helping them improve their scores. Traditional tests have frequently been criticized for having both potentially negative academic and social consequences.

Conventional tests are also criticized for looking for perfection rather than student progress. Instead of stimulating collaboration and weighting progress or the act of learning itself, tests promote competition among students as individuals in isolation, overemphasizing achievements and outcomes alone. They have a contribution to the ‘academic race’ where only the most knowledgeable student matters, not others. A standardized test “penalizes students who are achieving well but are not in the top percentage of students, and thus can cause these students to lose their motivation and interest in learning” (Bachman and Damböck, 2018: 222).

Instead of evaluating a student’s work according to some standard rating scale, traditional tests assess their ability in comparison with others. As Cheng and Fox (2017: 188) explain, “For assessment to be effective and to enhance, not harm, students’ learning, students must compete with themselves to continue to improve, and teachers should use assessment events to help students to develop effective learning strategies”. This alternative is beneficial to them for more enduring and memorable learning in the classroom and real-life situations outside the academic center, making them prepared to be integrated into the 21st-century world. In short, effective assessment should engage students in showing rather than telling what they learn, in explaining how they reach the answer, in showing their depth of understanding, and in making something new from what they know. Conventional tests, to a great extent, fail to do so.

Another major dissatisfaction hinges on the argument against the traditional view that conventional tests can universally and uniformly measure all individuals and all abilities; that is to say, a “one-size-fits-all” testing mindset. The traditional testing system ignores individual differences, including students’ diverse backgrounds, learning styles, abilities, and needs. Thus, it may not effectively capture these differences. Critics argue for the use of multiple measures of assessment that encompass a variety of methods and perspectives, providing a more comprehensive and nuanced understanding of student learning and achievement. They also point out that traditional tests may perpetuate inequality and bias, as they may be culturally biased or disadvantage students from underprivileged backgrounds who lack access to resources for test preparation. The emphasis on grades in traditional testing systems can lead to a narrow focus on performance outcomes rather than fostering a love of learning or intrinsic motivation to explore and understand new concepts.

Addressing these criticisms has led to calls for alternative assessment methods, including PBA, that may offer more comprehensive and meaningful evaluations of students’ existing knowledge and actual performance. Today, the significance of language assessment is not just limited to its role as an instrument for measuring learning. Instead of just auditing instruction (sorting and selecting students or justifying a grade), the assessment system must inform language teaching and promote language learning. A single testing form is not sufficient to measure complex subject matters in units, degrading language assessment to a numerical “snapshot” of student achievement. Using several types of assessments gives a more comprehensive appreciation of student learning. What is assessed, how it is assessed and evaluated, and how the results are reported all send out a strong signal to students about what they should learn, how they should learn it, what elements of quality are most important, and how well they are expected to perform. Thus, since the turn of the century or even earlier in the late1990s, there has been a momentous movement from traditional testing toward language assessment in general and alternatives in assessment in particular.

2.1 Alternatives in Assessment

Alternative practices in language assessment mainly emerged in the early 1990s as a reaction to the apparent inadequacies and shortcomings of more traditional methods of testing and assessing an individual’s ability and as a means for educational reform. These two competing systems of language assessment can be best understood as the two extremes of the same spectrum of assessment possibilities. The main impetus behind the new paradigm is that instructors understood that “not all students and not all skills can be measured by traditional tests” (Reardon, 2017: 195) and see that traditional testing “alone cannot offer the panacea to account for students’ learning outcomes” (ibid: 196). Alternative assessment, or more accurately, ‘alternatives in assessment’ (Brown and Hudson, 1998), by definition, is the process of assessing and evaluating without tests. It includes a wide range of practices such as portfolios, journals, conferences, interviews, observations, self and peer assessment, diaries, inquiry-based learning projects, and PBA.

As a generic umbrella term, alternative assessment refers to any unconventional assessing procedure of realizing what an individual not only knows but also can do and perform. As they are closely connected to the teaching and learning process, alternative assessment procedures mainly aim to inform and improve instruction, showing students’ growth. According to Farhady (2022: 59), traditional “testing is retrospective, i.e., measuring what students have already learned”. In contrast, alternatives in assessment are regarded to be “prospective”, aiming “to guide learning”. Moreover, traditional testing is more teacher-dominated, leaving no room for students’ voices. Alternatives in assessment are more learner-centered, allowing students to collaborate with the teacher in designing and implementing the assessment process.

Accordingly, traditional testing provides minimal or no educative feedback to the students, whereas socially constructed feedback is the core of alternatives in assessment. Alternative assessment practices give the students a second chance to think over their answers and correct their possible mistakes by providing them with adequate, appropriate, individualized feedback. In sum, conventional testing typically measures only the “learners’ existing knowledge, i.e., the product of learning”, while alternative assessment principles and practices also focus on “the learners’ changing state of the knowledge, i.e., the process of learning” (Farhady, 2022: 59). As such, they support learning by placing assessment at the heart of the teaching-learning process and providing positive washback for instruction.

Admitting the growing awareness about the powerful impact of testing and assessment methods on instruction, the proponents of this paradigm promote alternative assessment as a complementary procedure (if not as a replacement) for conventional test-driven instruction. They argue that in a learner-centered classroom, alternative assessment contributes much more to the appropriate balance of power relationships between teacher and students, which results in fairness and eliminating inequalities (i.e., the issue of parity). In addition to providing students with equal chances for learning, “Fair assessment avoids student stereotyping and bias in assessment tasks and procedures” (Cheng and Fox, 2017: 11). Alternative assessment practices are said to have a positive washback effect and enjoy apparently greater consequential and face validity.

This paradigmatic shift in assessment purposes naturally led to changes in methods and measures of assessment and evaluation. These different systems of evaluating students’ competence and performance help stakeholders navigate the diverse, complex landscape of language assessments. These systems also empower them to make informed decisions in developing or selecting tests/tasks that align with their specific needs, objectives, and contexts. In an attempt to classify these various language assessment paradigms, approaches, methods, and measures, McNamara (2000) proposed a taxonomy of language tests/tasks in terms of their purpose (what they are for) and method (how they measure), which is illustrated in Table 1 as follows.

2.2 Rethinking Language Testing and Assessment

As presented in Table 1, conventional language testing, as a separate process from instruction, makes value-free, objectively quantifiable judgments about student achievement against some instructional objectives and standards. The measurement is commonly done once as a snapshot by administering a final test at the very end of a period of instruction on a particular date set in advance. However, a radical change has happened in the purposes for which assessment and evaluation are used in the new millennium.

View Full Size

Test classification in terms of purpose and method (adapted from McNamara, 2000)

Citation: Contrastive Pragmatics 6, 3 (2025) ; 10.1163/26660393-bja10139

• Download Figure
• Download figure as PowerPoint slide

Nowadays, language assessment is believed to be an integral, seamless part of the instruction, contributing to and supporting the learning process. The ultimate goal of evaluation is not ‘to measure’ but rather ‘to learn’, and assessment is viewed as a ‘rehearsal’ for the purpose of learning. In this sense, both teachers and students are actively involved in the dynamic, ongoing process of assessment, formally or informally. It is promoted that while teachers use assessment practices both to measure student learning and to improve their own teaching, students should be educated to account for their past and future learning, too.

In sum, analogizing the process of language assessment to that of chain- making, Bachman and Damböck (2018: 40) maintain that teachers should “create a series of links from students’ assessment performance to assessment records, to interpretations, to decisions, to consequences”. Salmani Nodoushan (2008: 1) clarifies the point more, “If curriculum, instruction, and assessment are integrated, the assessment itself becomes a valuable learning experience”. While in language testing, traditionally, teaching, learning, and testing were treated as separate components in a linear fashion, in more recently developed models of language assessment (e.g., Farhady, 2022), instruction and assessment are incorporated into a unified cycle, weaving assessment into the instructional process. Figure 1 depicts such a shift from traditional testing to a more expanded conceptualization of modern language assessment as an act of communication.

View Full Size

Relationship among teaching, learning, and assessment in traditional testing versus language assessment (adopted from Farhady, 2022: 60)

Citation: Contrastive Pragmatics 6, 3 (2025) ; 10.1163/26660393-bja10139

• Download Figure
• Download figure as PowerPoint slide

2.3 Revisiting Language Assessment Purposes

Against this backdrop, Cheng and Fox (2017) proposed a three-dimensional assessment model that allows optimal outcomes for assessment practices in language and translation classrooms. They introduced the ‘guiding principle of alignment’, arguing that for a language learning program to be successful, there should be maximum harmony among language curriculum, instruction, and assessment. Accordingly, language assessment is employed for three major purposes: the conventional function of language testing, generally known as Assessment of Learning (teachers as the only key assessors), and more modern coexisting functions of Assessment for Learning (teachers as the leading assessors involving students in assessment too) and Assessment as Leaning (students as active, engaged and critical assessors of their own learning). Assessment for Learning is defined as “the process of seeking and interpreting evidence for use by students and their teachers to decide where students are in their learning process, where they need to go, and how best to get there” (Cheng and Fox, 2017: 4). Assessment as Learning adds to these pieces of evidence by empowering students to “reflect on and monitor their progress to inform their future learning goals” (Ibid: 6). At the heart of Assessment as Learning lies self and peer, student-led assessment, which promotes language learner autonomy.

While Assessment of Learning simply deals with documenting learning status, such as scoring, ranking, and reporting, the other two explicitly aim to stimulate and enhance learning further to contribute to the teaching process. These three approaches, serving intertwined but distinct assessment roles and purposes, act as robust devices for measuring and boosting learning but in immensely different ways. All dimensions of assessment practices are essential in any language program, yet the trick is to get the balance right. In brief, Cheng and Fox (2017: 4) believe that student learning can best be supported through “the synergy of assessment for [/as] learning punctuated with the use of assessment of learning”.

Earl (2013) urges the critical need to adjust the weighting and emphasis placed on various types of assessments within educational systems. Her concept of shifting the balance in the assessment alignments reflects a broader movement in education towards more holistic and student-centered assessment practices that prioritize learning and growth over simply measuring outcomes. Figure 2 depicts shifting the reconfigured balance in the assessment alignments, illustrating the traditional dimension of language testing in conventional classrooms compared to that of modern assessment practices:

View Full Size

Shifting the balance in assessment alignments (adapted from Earl, 2013: 31–32)

Citation: Contrastive Pragmatics 6, 3 (2025) ; 10.1163/26660393-bja10139

• Download Figure
• Download figure as PowerPoint slide

At the current juncture, of course, Assessment of Learning still dominates evaluative practices for assessing students’ academic success within language curricula (Bachman and Damböck, 2018; Brown and Abeywickrama, 2019; Cheng and Fox, 2017; Farhady, 2021, 2022, Naderian et al., 2018), leaving little room for the other two dimensions of language assessment. Typically, every instruction is followed by a summative traditional test, primarily designed to evaluate learning retrospectively, categorize students, and communicate these assessments externally. However, more and more teachers are implementing Assessment for Learning by incorporating diagnostic processes such as formative assessment and providing feedback at various stages of their instruction. They also offer students opportunities to improve their grades and enhance their learning outcomes. Assessment as Learning remains largely absent, perhaps because of the hegemony of traditional assessment culture or institutional as well as individual resistance to change, making it difficult to shift towards this dimension of language assessment in practice.

Overall, the status quo in language assessment reflects a blend of traditional and innovative approaches, with a growing emphasis on technology, personalization, and inclusivity. Language assessment continues to evolve in response to technological advancements, pedagogical insights, and the changing needs of its stakeholders. As the field continues to evolve, further advancements in assessment methods and tools can be expected to meet the diverse needs of language learners and users worldwide. Ongoing research and innovations contribute to the refinement of assessment practices and the development of more effective tools for evaluating the language ability of the learners.

3.1 Translation Redefined

Following the critical evaluation of Meylaerts and Marais (2023), the landscape of translation has evolved significantly in response to the rapid proliferation of hyper and multimodal objects dispersed widely across vast spatial and temporal dimensions, coupled with the diminishing dominance of literary written texts once considered the primary focus of translation. Translation is no longer regarded as the mere art of conveying meaning from one language to another; translators are not treated as language mediators either. In fact, the evolution has led to a re-evaluation of traditional definitions and approaches to translation, and there is a need for more expanded conceptualizations of translation to capture a broader spectrum of activities and explore the intricate dynamics of translation in its full richness and diversity. Translation is increasingly understood as a complex and unpredictable process rather than merely the creation of a finished product. It encompasses a wide array of modalities beyond written texts, including visual, auditory, gestural, and interactive forms of communication. This recognition challenges the bias towards a certain communication mode (i. e., focusing solely on the ‘written’ form of language) that has historically constrained the study of translation. Moreover, the traditional binaries that have structured the understanding of translation, such as source-target, original-translation, and domestication-foreignization, are being questioned and transcended. These binaries have often served to restrict and oversimplify the complexity of translation practices.

In this expanded understanding, translation becomes a dynamic and multifaceted process that engages with the complexities of culture, communication, and meaning-making across diverse contexts. It involves negotiation, adaptation, and creative interpretation rather than mere replication or transfer of content from one language to another. Embracing this complexity allows for a more nuanced and inclusive approach to translation that acknowledges its role in shaping and mediating our interconnected global reality. According to Meylaerts and Marais (2023: 1), “Such expanded definitions consider translation not merely as a research object but also as a (research) practice, a process constructing, (re)assembling, and (re)connecting the social’ and as ‘an inter-semiotic all-encompassing epistemological tool and ontological concept’ which produces knowledge”.

Nowadays, both scholars and practitioners acknowledge the complex, dynamic, and multifaceted nature of the translation as “a socio-psychological phenomenon” (Almanna and House, 2024: 1). By framing translation as “intercultural communication and socio-cognitive action”, House (2014, 2015, 2024) argues that translation goes beyond the mere transfer of words from one language to another. Rather, it involves various social factors in navigating intercultural differences and engaging in cognitive processes to achieve accurate and meaningful cross-cultural communication. By treating translation “as an act of ‘re-contextualization’” (Almanna and House, 2024: 198), this perspective underscores the importance of considering both cultural and cognitive dimensions in translation theory and practice.

Within the framework of House (2024), on one hand, ‘translation as intercultural communication’ denotes a multifaceted process whereby texts are transferred across linguistic and cultural boundaries to facilitate understanding and exchange between individuals or groups originating from distinct cultural contexts. This concept encapsulates the intricate interplay between language, culture, and communication modalities, highlighting translation as a pivotal mechanism for bridging cultural disparities and fostering intercultural dialogue. Within this perspective, translators assume the role of cultural mediators tasked with navigating linguistic nuances, contextual intricacies, and sociocultural norms to convey meanings accurately and effectively across cultural divides. As such, translation emerges not merely as a linguistic endeavor but as a dynamic intercultural encounter, wherein the transfer of meaning transcends linguistic barriers to engender mutual comprehension and appreciation of diverse cultural perspectives.

This paradigm emphasizes the essential role of translation in fostering cross-cultural communication, promoting cultural exchange, and cultivating intercultural competence in an increasingly interconnected global milieu. In brief, intercultural communication highlights the importance of understanding cultural contexts when translating texts. Translators must consider cultural norms, values, and practices to ensure that the translated text is appropriate and understandable to the target audience. Translation acts as a bridge between cultures, enabling communication and understanding across linguistic boundaries. It involves more than just linguistic proficiency; it requires cultural competence and sensitivity to bridge cultural gaps effectively (House, 2015).

On the other hand, within the same framework, translation can be seen as a socio-cognitive action involving both social and cognitive processes, which include the translator’s cultural background, knowledge, beliefs, and cognitive processes involved in understanding and rendering meaning from one language to another. These factors are inherent in the act of translation (House, 2013). On a social level, translators interact with various stakeholders, such as authors, clients, and readers, to negotiate meaning and ensure effective communication. On a cognitive level, translation requires problem-solving skills, linguistic knowledge, and critical thinking to make informed decisions about how to best convey the source text’s meaning in the target language. This conceptualization posits translation as more than a mechanical transfer of linguistic units; rather, it emphasizes the active engagement of translators in complex cognitive operations influenced by sociocultural factors. Within this paradigm, translation is construed as a dynamic cognitive endeavor wherein translators draw upon their linguistic proficiency, cultural competence, and cognitive strategies to comprehend source texts and render them into target languages effectively.

Moreover, sociocultural factors such as cultural norms, historical contexts, and power dynamics play a crucial role in shaping translators’ interpretations and decision-making processes (House, 2024). By conceptualizing translation as a socio-cognitive action, this perspective accentuates the interactive nature of translation, highlighting the reciprocal relationship between cognitive processes, sociocultural contexts, and the production of translated texts. Consequently, this perspective enriches our understanding of translation as a socio-culturally embedded practice, shedding light on the intricate interplay between cognition, culture, and communication in the translation process.

In sum, understanding translation as intercultural communication and socio-cognitive action emphasizes the dynamic and interactive nature of the translation process, highlighting the role of cultural sensitivity, linguistic competence, and cognitive strategies in producing accurate and culturally appropriate translations. By the same token, following a crucial distinction first made by House (2001), Almanna and House (2024: 202) pointed out that “it is important to be maximally aware of the difference between scientifically based linguistic analysis and social judgment (relating to values, ideology, identity, gender, issues etc.) in evaluating any translation”. Deciding whether a translation is both linguistically accurate and culturally appropriate is not an easy, straightforward task within such an expanded conceptualization of translation within the framework of intercultural communication and socio-cognitive action (House, 2001, 2013, 2014, 2015, 2024), either in professional contexts in the translation industry or in the academic setting in translation studies as a discipline. In conclusion, “This perspective enriches our understanding of translation as a socio-culturally embedded practice, shedding light on the intricate interplay between cognition, culture, and communication in the translation process” (Heidari Tabrizi and House, in press).

3.2 Translation Evaluation in Academic Context: Status Quo

For the purposes of the present work, a distinction has been made between two different contexts where translations are judged in terms of their quality: the educational context (translation as an academic discipline) and the professional context (translation as industry/market). In so doing, it uses two different terms: “testing techniques” for pedagogical purposes and ‘evaluation of translations’ for translation criticism. The designation of ‘evaluation’ is left for the decisions usually made by the translation teachers in translation programs, which is referred to as academic translation evaluation (For further elaboration on the terminology, see Heidari Tabrizi, 2021b).

Language assessment and evaluation are essential and integral constituents of the teaching-learning process done for pedagogical purposes in any academic setting. Effective testing and assessment are critical in revealing the extent to which students learn what they should learn. Equally, translation assessment through quality evaluation is an indispensable part and parcel of every translation program. In the context of translator education, a common activity of crucial concern is evaluating students’ academic translations through assigning numerical scores or letter grading in order to detect the relation established between intended instructional objectives and learning outcomes.

Inevitably, in the academic career of every translation teacher/student, there are always diagnostic tests, mid-term and final examinations and less possibly other more formative assessments (Huertas-Barros et al., 2019), which are to be developed, administered, evaluated and scored. Based on such scores, translation teachers make decisions that might have serious impacts on students’ educational careers, particularly and their entire lives, generally. That is why translation assessment and evaluation are of significance in translation programs. The more severe these impacts are; the more critical translation evaluation will be. Therefore, evaluations made must be fair, justifiable, strongly dependable, and highly credible. The consequential validity and accountability of these measuring instruments used for evaluation are of utmost importance, too.

An instructor of translation, much like teachers of any other discipline, is expected to help learners develop their actual performance; as an academic member of staff, s/he is required to assess the quality of their students’ work. In fact, the central role with which translation teachers are normally associated is twofold: they serve as facilitators of the learning process, and, at the same time, they must act as evaluators of what students have achieved. In other words, as Honig (1998: 32) argues, in the academic context, judging the translation quality “should not be an end but a means”. Thus, as a rule of thumb, the process of evaluating students’ academic translations and decision-making is certainly one of the most challenging tasks a translation teacher faces. In the words of Adab (2000: 227), “Translation assessment in the university environment is a problematic issue”. The problem gets much more decisive as more and more students are attracted to translation programs, “mainly oriented towards training future professional commercial translators and interpreters and serve as highly valued entry-level qualifications for the professions” (Munday et al., 2022: 11).

There has been a massive spreading out of translator education programs throughout the world since the turn of the century to meet the excessive demand for translation. In consort with such rapid proliferation in quantity, today, a growingly sophisticated body of research and knowledge has also been devoted to improving the quality of translation curricula, including its various aspects such as curriculum design, teaching methods, instructional materials, etc. However, lesser attention has been given to the often neglected teacher evaluative practices of student translations. In fact, in the context of translator education, translation evaluation is a widespread yet challenging activity of vital significance typically carried out by teachers in translation courses. Accentuating its significant role in translator training and education, Campbell and Hale (2003: 221) explain that “better assessment means better translators and interpreters”.

Nevertheless, this common practice has received the least attention in Translation Studies during the past decades (Arango-Keeth and Koby, 2003; Bowker, 2000). At the turn of the last century, Hatim and Mason (1997: 197) assert that in comparison to other areas in Translation Studies, “little is published on the ubiquitous activity of [translation] testing and evaluation”; thus considering it an “under-researched and under-discussed” area where, in words of McAlester (2003), the field suffers from “its worst failure”. More recently, Tsagari and Van Deemter (2013: 11), while admitting that a growing number of studies can be found in the field of translation assessment and evaluation, argue that “there is still much work to be done”. Dungan (2013: 132) echoes the same idea that the area of assessment “requires more critical and systematic research”. In the same vein, Conde (2013: 108) also believes that translation evaluation within this field “is still a much-debated topic” on which many questions have remained unanswered yet. In the early 2020s, most recently published works confirm that the situation remains rather unchanged (Abdel Latif, 2020; Heidari Tabrizi, 2021a, 2022a; 2020; Yazdani et al., 2020, 2023).

3.3 Translation Evaluation: Existing Challenges

At present, the dominant trend for evaluating translation quality in most translation programs in academic settings follows the principles of Classical True Score Measurement Theory employing a traditional testing system, which relies heavily on inauthentic, summative, competence-measuring, decontextualized tests. The heavy weight given to these conventional tests and the excessive emphasis on their results, especially in translation programs, have terrible consequences. Translation teachers carelessly replace any awareness of the context complexity of the evaluative practices with the authority of their personal stances and opinions (Heidari Tabrizi, 2021a, 2022a, Moeinifard et al., 2014). Students feel that the evaluation of their translations is done on the basis of arbitrary, subjective practices. In most cases, translation students do not know by what criteria their work will be evaluated. They spend most of their energy adapting themselves to the personal criteria of their teachers and feel that it is a waste of time to gain insights into the nature of translation processes as provided by translation theories. Consequently, they lack the self-awareness as well as the self-confidence they need to carry out translation tasks when they are on their own in the real – and sometimes confusing – world of translations. One piece of evidence for this assertion can be the frequent negative feedback teachers are likely to receive from the students about the final translation tests every semester. Still, another piece of supporting evidence is the countless anecdotes one hears in professional conferences about the deficiencies of translation tests.

In brief, with the expansion of the definitions and conceptualizations, translation is a complex, multifaceted process; the main problem in evaluating students’ academic translations is the fact that inauthentic, summative, competence-measuring, decontextualized standardized paper-and-pencil tests fail to work effectively at least when using alone as the only measuring instrument. As mentioned earlier, translation, by nature, is more an act of performance than mere competence; it is an ongoing cognitive complex activity that is sensitive to contexts of situation and culture. As a task-focused, project-centered activity, translation, according to Darwish (2010: 105), “is rather a rational, objective-driven, result-focused process that yields a product meeting a set of specifications, implicit or explicit”. His definition suggests that the success of a translation depends more on the translator’s ability to perform the act of translation, to interpret skillfully, and to express the text in the target language rather than solely relying on linguistic competence. It underscores the dynamic and interpretive nature of translation, highlighting the role of the translator as an active participant who must engage with the text and bring it to life in the target language. While linguistic competence is undoubtedly essential for accurate translation, the performative aspect of the craft is emphasized, where the translator’s creativity, intuition, and understanding of the cultural context play crucial roles in producing a successful translation.

Translation can be more conveniently approached within the characteristics of a non-linear interactive task cycle of three stages: pre-translation, in-translation (translating proper), and post-translation activities. During the pre-translation activities, students may do some preliminary research to become familiar with the text and its topic, the writer and their style, as well as gathering and ordering ideas about which translation strategy should be selected throughout the translation process in general and which translation technique can be employed for particular cases. They can also contemplate over the workflow that the task goes through from initiation to completion, making resources and mechanisms required for successful task accomplishment available. Focused reading and several rereading of the text and connecting it to its context are key steps in this stage.

The in-translation stage begins with translators producing an initial draft translation of the source text into the target language. This step includes all the activities done by a translator while doing the actual translation. It involves transferring the meaning and nuances of the original text into the target language as accurately as possible and considering the cultural and linguistic differences between the two languages. In this stage, translators, as well as translation students, should lay a solid foundation for further refinement and improvement in subsequent drafts.

After drafting the translation, translators are advised to take breaks periodically to rest their minds and prevent fatigue, which can impact the quality of their translations. As Sainz (1992) explained, in this ‘Escabeche Period’, translators should empty their minds of all the deliberations and thoughts they have made hitherto and forget their translations, preparing themselves for the post-translation activities, which usually include reviewing, revising, editing, and proofreading. These can be done with a fresh look after a longer break, revisiting the translation, and approaching it with the same fresh perspective as if reading it for the first time. By checking for possible linguistic and translation mistakes and errors, (would be) translators can add to the quality of their translations (including accuracy, consistency, fluency, and clarity) before finalizing them for submission. More often than not, it is possible for them to seek feedback from peers, instructors, or experts to gain deeper insights and have suggestions in hand for refinements.

As the stages are non-linear and cyclic, students can jump forward or move backwards across the stages and sub-stages, too; the translation process involves a recurring sequence of actions or steps in completing translation tasks. The translation cycle does not strictly follow a linear progression from one stage to the next; translators can revisit earlier stages or skip ahead to later stages as they refine their workflow. In all these stages, like any professional translators in real-life situations, translation students have access to printed and online sources and databases, can consult dictionaries or more knowledgeable people in the field, can ask for ‘correct answers’, and may do the translation with the help or collaboration of others. Newmark (1988: 6) is right to say that “there is no such thing as a perfect, ideal or ‘correct’ translation”. To him, “Translation is for discussion … Nothing is purely objective or subjective- There are no cast-iron rules. Everything is more or less … there are no absolutes” (Ibid: 21). As he explains further, “Translation is enjoyable as a process, not as a state. Only a state is perfect” (Ibid: 225). If translation is open to discussion and negotiations, how is it possible for the students, who are left alone on their own and deprived of any discussion under the constraints of a test session, to produce a translation that is going to be treated as a ‘perfect’ product?

If formal translation programs and classes are to prepare students for their future profession as translators, they should stimulate workplaces and the conditions under which would-be translators should work effectively. In real-life situations, the translation projects can be completed within a reasonable amount of time in a couple of days or even weeks, not within a one or two-hour session allocated by the test developers. Hence, translators have enough time to go through different stages of the translation process several times by doing and re-doing their translation again and again. This allows them to apply lessons learned from previous attempts and potentially achieve better results. They can improve the quality of their final translation product and refine it through discussion, negotiation, and collaboration too. Again, it is not clear how standardized paper-and-pencil tests can take care of these issues in evaluating students’ academic translations. These tests, by nature, require testees to act individually, autonomously, and independently and regard teamwork, collaboration, and consultation as fraud and cheating.

All these are absent in evaluating the process of students’ translations in academic settings where teachers develop tests measuring students’ competence rather than their actual performance. To increase the reliability of the measurement, they go through ‘standardizing’ these tests. The more standardized the teachers try to make a translation test, the less authentic and contextualized it will be. On the one hand, translation is characterized in nature by the aforementioned defining features and more. On the other hand, standardized tests enjoy certain characteristics that make them partially unfitting, if not totally inappropriate, for translation evaluation in translation classes. They fail to provide the whole actual translating performance of translation students as they are expected to act in real-life situations.

In fact, research shows that the prevailing evaluative practices more often result in rather a sense of frustration in both the students (Chalak and Heidari Tabrizi, 2013; Heidari Tabrizi, 2008, 2021a, 2022a; Heidari Tabrizi et al., 2008) and instructors themselves (Yazdani et al., 2020, 2023). In fact, as Honig (1998: 29) declares, believing that translation tests are “subjective and arbitrary”, students “try to adapt to the standards of teachers, and they acquire neither self-awareness nor self-confidence”. More recently, Yazdani et al. (2020, 2023) have found that translation teachers at Iranian universities are least informed and familiar, if at all, with the current translation evaluation approaches and methods in the field of translator education. This is in agreement with Newmark’s (2003: 65) idea that “examination boards and examiners are not aware of the literature”. Similarly, Honig (1998: 29) asserts that “Obviously, many teachers and lecturers are not aware of the fact that there is such a wide variety of evaluation scenarios and applied criteria”. Heidari Tabrizi (2008, 2021a, 2022) showed how much discrepancy is found among the translation teachers as translation test designers. He argues that in developing these tests, they do not fuse theory-driven principles and standards. However, rather they follow their intuition and personal experience, and the criteria they employ in scoring students’ academic translations are highly subjective. Consequently, the quality and accountability of such tests are under serious question.

Thus, the question remains as to the possible solutions that can be offered to escape from this complex situation. In other words, how can the present situation be improved and promoted? The next section deals with a cohort attempt to occupy the existing niche in academic translation evaluation. Accordingly, in the next section, performance-based assessment is introduced as one of the potential solutions that may help fill the gaps left by traditional testing.

4.1 PBA: Definition and Theoretical Justifications

Originally, a performance-based test was explained by Bachman (1990: 304) as an applied test that “measures performance on tasks requiring the application of learning in an actual or simulated [near real-life] setting”. According to Richards and Schmidt (2010: 428), PBA aims “to measure student learning based on how well the learner can perform on a practical, real-world task”, and it is believed that PBA is “a better measure of learning than performance on traditional tests such as multiple-choice tests”. PBA is characteristically intended to assess more complex abilities with tasks rather than traditional fixed-response tests. PBA employs authentic, contextualized tasks that engage students in the application of learned knowledge and skills in performing real-life tasks successfully within a meaningful, culturally relevant situation rather than involving them in test items that demote mental functioning to discrete, isolated, decontextualized abstract knowledge. Most recently, Farhady (2022: 59) delineates that “any assessment that attempts to measure the produced language” may fall within the realm of PBA.

McNamara (1996) proposed two justifications for the emergence and development of PBA. First, it is getting more and more important to evaluate L2 learners’ language abilities and skills which are expected to join certain professional workplace settings as tour guides, health care providers (such as physicians, surgeons, nurses), or air traffic personnel (e.g., pilots, air traffic controllers, flight attendants). Second, nowadays, the evaluation process is not just limited to language usage and accuracy issues, but rather, it is more demandingly dedicated to language in action; that is to say, communicative language use and appropriateness in various real-life contexts. The ‘strong’ version of PBA deals with the former, whereas the former is taken care of within the weak ‘version.’

The strong version involves evaluating behavior-based performance as the target of assessment in its use context either through continuous observation over extended periods of time in natural situ (Direct Assessment) or on-the-job or on-site observation at agreed times (Work Sample Method). In this sense, PBA assesses “the extent to which the actual task itself has been achieved, with the language being the means for fulfilling the task requirements rather than an end in itself” (Wigglesworth and Frost, 2017: 121). The weak version happens in the class through simulating real-world task demands using tailor-made tasks where the focus is on integrated language skills and task performance serves only as a vehicle. As such, the objective of assessment is not the successful, effective performance of the task itself (e.g., Occupational English Tests, abbreviated as OPT, in Australia).

4.2 PBA: Major Principles and Premises

By employing meaningful, engaging, action-oriented tasks and projects, PBA rigorously integrates assessment with the curriculum and instruction (i.e., teaching and learning), promoting students’ learning. It is generally known as one form of alternative assessment practice that requires students to show, rather than tell, what they know or have learned by performing specific, specified tasks. As a common evaluative practice, it has been used for years in certain workplace contexts as well as specific vocational and technical educational settings (e.g., physical education, performing arts such as music, theater, ballet, etc.). In these settings, instruction is concentrated on actual performance; thus, PBA is widely used to measure an individual’s learning or achievement. In these cases, performance is described as a candidate’s active context-dependent production of an observable response. According to Davies et al. (1999: 144), in PBA, the ability of students is judged by asking them “to perform particular tasks, usually associated with job or study requirements”.

As a socially situated activity, language assessment practices unavoidably happen in authentic, real-world contexts, which “is particularly the case with task- and performance-based assessment” (Wigglesworth and Frost, 2017: 130). In PBA, tasks are developed to assess productive, performance-oriented integrated skills essential for real-life situations or simulations. Following Brown and Abeywickrama (2019), the broader definition of PBA entails task-based assessment, but not vice versa. In other words, task-based assessment must be treated as a subset of PBA and not a synonymous term for it. Accordingly, all debates on PBA inevitably involve some deliberation and negotiation of task-based assessment.

Traditional tests normally entail the process of objective measurement of a testee’s more abstract demon of knowledge by means of a reliable instrument, typically a formal, discrete-point (standardized) paper-and-pencil test where testees are required to answer some questions ‘correctly’. In contrast, PBA normally involves some human raters or scorers judging the quality of candidates’ performance on some tasks by using an agreed-upon rating scheme, rubric, or scale. In other words, the degree of achievement is rated and scored by performance observation and professional judgment. As such, PBA entails three major components: tasks, rating scale, and human raters. Thus, as McNamara (1996: 117) asserts, PBA “necessarily involves subjective judgments … and involve acts of interpretation on the part of the rater”. Traditional tests, in contrast, are claimed to be scored more objectively due to their rather fixed- response nature.

The majority of traditional tests are normally accompanied by relatively static scoring keys where the ‘correct’ answers are predetermined; thus, the end-all, be-all intention is eventually to obtain the right answer. As opposed to such a tradition, PBA practices do not necessarily involve clear-cut right or wrong answers. More often than not, there is more than one acceptable solution to performance-based tasks and projects. PBA tends to be more process-oriented procedures; what matters is not just the correct answer itself (the product) but rather the way students arrive at it (the process or solution). As an ongoing, more humanistic process, PBA displays what students learn not only at the end of the instruction but also, more importantly, throughout the process of learning or working on a problem.

In fact, PBA measures the extent to which an individual is successful or unsuccessful in performing a task or project. They may ask candidates to create their personally constructed responses to solve or overcome a problem, usually along with the process by which they approach the problem and solve it by explaining how they arrive at their solutions. Moreover, they are required to defend their choice or solution by explaining and justifying it. In brief, it can be concluded that while traditional testing is ‘done’ to a student, PBA is done by the student as an active participant. Figure 3 illustrates the similarities and dissimilarities between the two methods while highlighting the differences:

View Full Size

Traditional testing vs. PBA (adopted from McNamara 1996: 9)

Citation: Contrastive Pragmatics 6, 3 (2025) ; 10.1163/26660393-bja10139

• Download Figure
• Download figure as PowerPoint slide

4.3 PBA: Defining Characteristics and Affordances

According to McNamara (1996: 6), one of the defining characteristics of PBA “is that actual performances of relevant tasks are required from the candidates”. In other words, PBA aims to ‘show learning’; that is, to perform the application of knowledge rather than the regurgitation of facts. PBA intends to bring out optimal performance by requiring students to do something, execute a task, create a solution, develop a response, integrate knowledge across disciplines, produce a product, contribute to group work, or generate an action plan for new situations they may face in the real world. Teachers are expected to coach students to reach at least some level of excellence by assigning team-based projects that are oriented toward problems they probably encounter in real life and that they are expected to solve together. In sum, PBA is the process of involving students in challenging authentic tasks that necessitate various steps, which stimulates a wide range of active responses. By employing stimulus materials, the tasks are authentic (real-life approximations), motivating, challenging, direct, highly contextualized, humanistic, and learner-centered.

As such, PBA encourages teachers to vary their assessment methods, giving multiple opportunities without penalizing them for limited resources. That is to say, if students fail to demonstrate their ability on an assigned task at a specific time, they still have the chance to perform the task at a different time and in a different situation. Thus, this learning-oriented rather than test-centered assessment does not burden students with expectations of earning high grades on the often-practiced final tests, that is, acing the subjects. It informs both the teacher and students about the learning process, the strengths and weaknesses, stressing complex education-wise learning.

PBA aims to enhance student learning by assessing students on what they do in their daily class activities, promoting collaborative working and teamwork, and expecting students to present their work publicly. Based on the cognitive domain of Bloom’s Taxonomy of learning objectives, traditional tests are said to measure lower-level thinking abilities (like memorizing, recognizing, recalling, listing, identifying, describing, and classifying). In contrast, PBA, like other alternative assessment practices, is claimed to engage higher-level thinking abilities and more complex problem-solving learning skills (including analyzing, synthesizing, evaluating, justifying, and creating). In brief, while traditional testing tradition merely fosters rote memorization of ‘facts’, mechanical filling of the gaps, reproduction of models taught, and the like, PBA encourages the value of deep and reflective learning.

Table 2 summarizes the defining characteristics of PBA as compared to those of traditional testing. It should be stressed that “The differences do not lie in the superficial treatment of the terms but lie deep in their purpose, process, and implementation” (Farhady, 2022: 59). The table is mainly built on what Brown and Abeywickrama (2019) have concluded from the literature at hand. As they correctly notify, the content of the table must be treated with caution as it depicts “overgeneralized” notions. Moreover, the features are represented as if they are absolute, binary opposing sets of all-or-nothing nature, whereas, in reality, the border between the sets is fuzzy with no clear-cut line of distinction. Thus, at the top of Table 2, an arrow between two extremes of the spectrum (i.e., knowing and showing) has been added to help avoid such a misinterpretation and to indicate that these characteristics are of, rather more or less, continuum-based nature.

View Full Size

Traditional testing vs. PBA (based on Brown and Abeywickrama, 2019: 17)

Citation: Contrastive Pragmatics 6, 3 (2025) ; 10.1163/26660393-bja10139

• Download Figure
• Download figure as PowerPoint slide

In a nutshell, PBA is claimed to have the following merits. It compensates for the negative aspects of traditional standardized tests, such as negative washback, test biasedness, and consequential validity. Moreover, it is more valid than conventional tests in predicting students’ abilities in future real-world settings. It also documents and promotes creativity, critical thinking, and self-reflection. PBA decreases the so-called ‘Lake Wobegon effect’ where traditional testing tradition reinforces testees’ illusory superiority and overstimulation. Overall, PBA provides a robust framework for integrating diversity, equity, and inclusion into translation teaching, learning, and assessment in translation programs across the world. By leveraging authentic contexts, culturally relevant content, differentiated assessment, critical reflection, community engagement, and personalized support, translation instructors can create inclusive learning environments where all students feel valued, respected, and empowered to succeed.

4.4 PBA: Potential Challenges and Expected Setbacks

Just like other assessment approaches, PBA is not without its share of shortcomings and limitations. Like any other alternative assessment practice, it is fraught with certain challenges that persist, especially when it comes to implementing it in academic contexts. PBA typically faces challenges in terms of design and administration. According to Wigglesworth and Frost (2017: 129), PBA is “one of the most expensive approaches to assessment, and in terms of development and delivery, one of the most complex”. As for its practicality, it is quite exhausting in terms of time and energy that both teachers and students should spend on the rehearsal and for the actual performances. For teachers, it is also intensively laborious to judge and evaluate the quality of the learner’s performance and grade. Performance assessments can be resource-intensive, requiring significant time and effort from both teachers and students. This can be challenging to scale up, especially when assessing large groups of individuals. Scaling such assessments can be challenging without compromising the quality of the evaluation.

Another noteworthy drawback of PBA that must be taken into account is related to the issue of generalizability. Despite the daunting workload imposed on its stakeholders, the outcomes tend to be less generalizable to other contexts as performance tasks, by nature, are precisely contextualized and complex. Another concern related to the problem of reduced transferability of the findings is that the performance judgments may not be replicable in other settings as tasks tend to be hard to replicate, keeping the measurement reliable and consistent. Addressing these challenges requires careful planning, ongoing evaluation, and the development of robust assessment practices that prioritize fairness, validity, and reliability while also recognizing the importance of authentic skill demonstration.