How to teach revision: tips from an interview study

ABSTRACT

Revision is an important – sometimes mandatory – step in the production of translations. As such, it is part of the EMT competence framework and increasingly taught in training institutions. Material is available to teachers, but it mainly consists of texts based on personal experience or accounts from the classroom. This article takes stock of a large interview study on revision to make research-informed proposals for revision courses in translation curricula. It addresses four topics: the diversity of revision, the situated nature of revision, the difficulty of the task (including the main faults of young revisers), and the social dimension of revision. Concrete proposals for enriching teaching are put forward, including revising with different time constraints, dividing the task into subtasks and asking students to perform only part of the task, asking students to justify the changes made, and teaching the principles of good communication.

KEYWORDS:

• Revision didactics
• interview study
• difficulty
• social dimension

1. Introduction

Revision, which is defined as the verification of a human translation by a second translator, is a significant part of a translation project and activity. First, revision is a mandatory step of a translation project, as defined by the international standard specifying requirements for translation services (ISO 17100; ISO, International Organization for Standardization 2015), even if revision is not provided in all settings and contexts. Second, revision is carried out by translators, and hence forms part of the occupation. The importance of revision can also be seen in its inclusion in the competence framework of the European Master’s in Translation (EMT), which requires that ‘students know how to […] revise […] their own work and that of others’ (EMT 2022, subcompetence 11). Translation curricula – at least those of institutions that are members of the EMT network – should therefore incorporate revision (see Konttinen, Salmi, and Koponen 2021 for a concrete example of how revision can be implemented in a curriculum).

Although the literature on revision didactics is quite extensive, it still has little empirical foundation. This article aims at enriching the teaching of revision using a large study conducted in Switzerland. In this article, I will argue that a training course in revision in a translation curriculum must take into account the (i) diversity, (ii) situated nature, (iii) difficulty and (iv) psycho-social dimension of revision. To do so, I will put forward fourteen proposals for teaching.

The rest of the paper is organised as follows. First, the literature on revision didactics is reviewed (Section 2) and the study upon which this article is based is presented (Section 3). Next, the results section addresses the topics mentioned above: the diversity of revision practices (Section 4.1), the situated nature of a revision task (Section 4.2), the difficulty of the linguistic changes and most common failings of young revisers (Section 4.3), and the psycho-social implications of revision (Section 4.4). Finally, fourteen proposals for teaching based on the results are described and discussed (Section 5).

2. Literature on revision didactics

The literature on revision didactics varies in line with the different meanings of the term. In the broader sense of the word, didactics refers to learning something (this is the meaning of the adjective ‘didactic’: ‘intended to instruct’, OED online¹). The best examples of works intended for people who want to learn how to revise are revision handbooks. Horguelin and Brunette (1998) focus on the revision of English-French translations, and also look at the editing of original French texts. They provide exercises, a list of tools, as well as a discussion of revision methods, principles, techniques and relationships. Mossop’s (2020) is not oriented towards a specific language pair and addresses editing (modifying of original texts), other-revision, self-revision and post-editing. With regard to revision, he tackles issues such as degrees of revision, revision parameters, procedures and relationships. Other examples of didactic works include old introductory texts in encyclopaedias (Graham 1989; Samuelsson-Brown 1996) and articles in which practitioners give advice based on their expertise (Ko 2011; B. Nord 2018; Oueillet Simard 1984). In general, these authors emphasise that revisers must show moderation (Graham 1989; Ko 2011; B. Nord 2018; Oueillet Simard 1984; Samuelsson-Brown 1996); indicate that revision is specific to each situation (Graham 1989; Ko 2011; B. Nord 2018); and call for translators and revisers to have contact and to communicate in a respectful manner (Graham 1989; Ko 2011; B. Nord 2018; Oueillet Simard 1984).

In the narrower sense, didactics relates to the teaching of something, i.e. how to teach somebody (‘the science, art, or practice of teaching’, OED online²). Every teaching activity depends on (i) the objectives to be achieved or skills to develop, (ii) the activities (i.e. the means to achieve objectives), and (iii) the assessment (i.e. verifying that the objectives have been achieved). Good examples of how the three are linked are provided by Künzli (2006b) for revision and Kruger (2008) for editing. As far as objectives are concerned, Mossop (1992) structures his chapter around the objectives of an undergraduate course on editing and revision, which notably include learning to see the text from the final reader’s point of view, focussing on larger text structures, justifying changes, or making the mental switch from a retranslating to a revising frame of mind (other lists of goals in a revision course can be found in Brunette and Gagnon 2013; Schjoldager, Wølch Rasmussen, and Thomsen 2008; Scocchera 2014). As for skills, three revision competence models exist: Künzli (2006b) groups revision competences under three labels: (i) strategic, (ii) interpersonal, (iii) professional and instrumental. Hansen (2008) developed a competence model for revision based on a previous model designed for translation, highlighting two skills that are specific to revision: the ability to show tolerance and fairness; and the ability to explain the changes that have been made. The third and most comprehensive model is based on previous translation competence models, the two above-cited revision competence models and, more generally, the literature on revision (Robert, Remael, and Ureel 2017). It features one main subcompetence (strategic subcompetence), eight subsequent subcompetences and three other components. Five of the eight subcompetences are common to translation and the three specific to revision are: knowledge about revision subcompetence, revision routine activation subcompetence, and interpersonal subcompetence. The three other components are: psycho-physiological components; translation and revision norms and brief; translator and reviser self-concept and professional ethos.

Of the three components of teaching (objectives, activities, assessment), the activities to be carried out in class make up the largest component. In the oldest texts, authors call for revision to be taught (Parra Galiano 2001) and advocate for particular attention to be paid to the parameters (logic and language correction, Brunette 1998, 2003, respectively). Several chapters or articles take a retrospective look at a particular teaching experience. This may come in the form of a testimonial (Hine 2003) or research that exploits empirical data related to teaching, such as student feedback (Schjoldager, Wølch Rasmussen, and Thomsen 2008), material produced in class (Brunette and Gagnon 2013; Scocchera 2014) or both (Hagemann 2019; Scocchera 2020). Schjoldager et al. (2008) conceived of and then taught a module on editing, précis-writing and revision, and they were the first to report on the feedback given by students. Overall, students were happy with the content and modalities of the module, but quite unsatisfied with the format of the final exam (a written exam in a computer room). Brunette and Gagnon (2013) report on a revision course in which they asked students to revise translations of Wikipedia pages, giving the would-be revisers an opportunity to practice communication with translators. In her first article on revision didactics, Scocchera (2014) focuses on the tools used in a literary revision course, whether practical, like a software solution (e.g. MS Word in track changes mode) or analytical, like Mossop’s parameters (2020). In her second article, Scocchera (2020) shows that analytical tools are beneficial but need time to be integrated, and that writing comments is useful to reduce the number of changes and to teach students to express themselves appropriately. Finally, Hagemann (2019) reports on a special teaching configuration. She taught German-to-English revision to a group of native English students as a native German speaker with English as B-language. She shows that students were happy with the course, and the need to justify the changes made in revision to the instructor proved to be useful.

Authors also offer suggestions on the syllabus, topics to cover, and tasks to undertake. Hine (2003) presents the syllabus of his 15-week course and describes the activities carried out in class, which mainly consist of discussion of in-class exercises or revision assignments, with the instructor taking on the role of facilitator. Mossop (2020) gives advice to teachers and explains how he distributes the content of his textbook over the 24-meeting course he teaches on editing, revision and post-editing. Rodríguez Rodríguez (2012) lists six activities to undertake in a revision course: studying the European EN 15038 standard (the predecessor of the ISO standard; see European Committee for Standardization 2006), analysing the professional revision market, writing a revision brief, analysing and practising communication between translator and reviser, using technology, and doing actual revision exercises. In her in-class experiment, Scocchera (2020) asked her students to revise the same text, first (a) without access to the source text, then (b) with the source text, and (c) adding comments for the translator, and finally, (d) at the end of the course. Mossop (2023) presents the topics he discusses and competences he teaches in full-day or half-day continuing education workshops, depending on the wishes of the organisers. The sixteen topics include revision definition and terminology, principles for making corrections, the need for other-revision, relationships with revisees and justification of changes, conflicts of loyalty, and consistency among revisers in large translation services.

As for evaluation, attention has mainly been drawn to evaluating the revision product. Evaluation has a long tradition in revision research, with the most recent and comprehensive work published by van Rensburg (2017). In addition, Mossop (2020, 235–240) proposes specific ways of assessing student work, e.g. focussing on problem detection, with students receiving points if the quality of the text is not reduced, irrespective of whether or not the change is fully adequate.

Overall, the resources are quite extensive, whether they are didactic texts or accounts from the classroom. However, the empirical foundation could be enhanced. The relatively few existing empirical studies tend to be based on data generated in the classroom, an exception being Scocchera (2014), which partly exploits the results of a questionnaire study conducted by the author. The aim of the present article is to strengthen the empirical basis of revision teaching and shed new light on it by looking at the results of an empirical study on revision.

3. Method: an interview study on revision and additional interviews on revision teaching

This article builds on interview data collected as part of a PhD project (Riondel 2023) and additional interview data produced specifically for this article. The original, primary study was dedicated to non-literary translation and deals with two main topics: (i) the relationships between translator and reviser, and (ii) the practices and perceptions of revision. It was conducted in Switzerland, mainly in translation teams (two freelancers who work with agencies were also interviewed). The teams fall into four categories, two public and two private: the Swiss Confederation, an intergovernmental organisation in Geneva, private companies with in-house translators and translation agencies. These contexts were chosen to represent the different sectors of the translation market and the diverse approaches to revision, including, for example, the horizontal and vertical models. In the former, translators validate the final version of the text and revision therefore has an advisory role, while in the latter, revisers correct the translation and pass it on directly to the next step (for a description and analysis of these two models in an institutional context, see Lafeber 2018, 74–76).

The data set of the main study consists of 45 in-depth interviews conducted with translator-revisers (32), heads of translation department (10), projects managers (2) and one proof-reader (mean length: 1h13min). It includes eleven departments: six translation teams of the Swiss Confederation, one intergovernmental organisation, two private companies with in-house language departments, and two translation agencies. The interviews were conducted in line with the concept of ‘responsive interviewing’ (Rubin and Rubin 2012), an approach to in-depth interviewing that emphasises the involvement of the researcher and their responsiveness to the words of the interviewee, who is called a ‘conversational partner’. As a result, not every topic had the same weight in each interview. Broadly speaking, meetings with the heads were more about how revision is organised and carried out in the department (as well as general information about the team), while interviews with translator-revisers focused on their professional background and tasks, revising method, attitudes towards revision, and communication before and after a revision task. Revision didactics was not a topic of conversation, with two exceptions: (i) seven participants were asked either about what skills are needed to revise or what makes a good or bad reviser (the question was posed at the beginning of the project, then abandoned because it was not productive); (ii) participants were asked if they had had (or would like to have had) any training in revision, either in their initial education or as continuing education.

Interviews were conducted between March 2019 and February 2021. Most took place in person at the participant’s workplace, but nine were organised in other settings (4 in cafés, 3 at the homes of participants and 2 at the university of the researcher) and three took place online via videoconference (these exceptions were mainly due to restrictions related to the COVID-19 pandemic). All interviews were conducted in French, with the exception of four in German. They were transcribed verbatim and summarised, then coded in accordance with the ‘template analysis’ approach (King 2004, 2012), which is a mix of induction and deduction and is specially tailored to compare the points of view of different groups of staff. Two full rounds of coding were carried out with the software QDA Miner Lite.³

The study received ethical clearance from the Ethical Review Board of the Faculty of Translation and Interpreting of the University of Geneva (approval No 10). The participants were given an Information and Consent Form, which they had to sign before the start of the interview. They could choose not to be recorded and their right to withdraw from the study was clearly stated.

In addition to the original study, three additional interviews were conducted in March 2024 with revision trainers in order to get expert opinions on the course proposals that arose from the original study. The three experts have an average of 21 years’ experience in revising (20, 28 and 15) and 10 years’ experience in teaching revision (10, 14 and 7). They teach French-to-German revision, English-to-French revision, and Spanish-to-English revision.

4. Results

The first subsection outlines the broad nature of revision by putting forward the diversity of practices found in the study (4.1.). The second addresses the situated nature of revision, i.e. the fact that the task is anchored in a specific situation and varies from case to case (4.2.). The third describe the difficulty of revision (4.3.), while the fourth brings the social dimension to the fore (4.4.).

4.1. Revision as a diverse activity

The terminology and the boundaries of revision are not well identified. Revision terminology is very loose, especially since different terms are used to designate the activity (see Koponen et al. 2021, 1–2; Mossop 2020, 116–118). As for the boundaries of this field, some group together revision with other related activities, i.e. the editing of TM suggestions and post-editing (Mossop 2020; Robert 2018), while others, including myself, prefer to differentiate between them (see do Carmo and Moorkens2021).

This fogginess can be set against – and is perhaps the result of – the diversity of revision practices, which is one of the main results of my study. This diversity can be identified at various levels, including the scope of revision, the person in charge of revision, the revision methods, and the relationship between translators and revisers. In terms of the scope of revision, nine out of the eleven departments involved in the study revise all texts, with or without some exceptions (e.g. short texts or texts with only limited uses). In the other two departments, only selected texts are revised, for example, translations that are particularly important or ones that are carried out by junior colleagues and freelancers. As for who is in charge of revision, beginners may revise from the onset in some departments, while in others, the activity is introduced after some time (e.g. one year) or restricted to translators with extensive experience. Revision methods also reveal the same diversity. Although bilingual, written revision is the type that is most widely used, one department revises all texts orally, using the method described by Allain (2010), and monolingual revision (as described by Mossop 2020, 168–170, 251) is used frequently in one team and occasionally in two others. The relationships between translators and revisers vary dramatically, from revising close colleagues to verifying the work of unknown freelancers. Communication between actors also differs from place to place. For example, some departments organise post-revision talks, which may involve explanations of the suggested modifications or more comprehensive feedback.

4.2. Revision as a situated task

Revising can be described as a situated activity, i.e. an activity that depends on the situation. Revising may even be seen as being doubly situated: revision varies (i) from one translational situation to another and (ii) in each specific revisional situation. In translation, functionalism has made it clear that translation choices depend on the situation (for an introduction to functionalist approaches, see C. Nord 2010): a good translation is a translation that is appropriate for its intended use and purpose. This means that translators do not always do the same thing. In order to apply functionalism, translators need information that should ideally be provided in the ‘translation brief’. This also applies to revision, thereby constituting the first way in which this activity may be considered to be situated.

In addition, revisers need specific instructions in order to perform their task properly. For 15 years, authors have described or advocated for a ‘revision brief’ (Allman 2008, 35, 36, 39–40, 47; Künzli 2007a, 50–51, 53–54; Parra Galiano 2016, 46–47; Scarpa 2010, 256–257; Yousif 2009, 12, 31). This brief may specify the scope of revision (e.g. parts of the text or parameters to be verified), aspects to focus on, procedure to follow or tools to use to present the corrections. It may also stipulate how much time should be devoted to the task or indicate whether or not corrections must be justified. The revision brief may be explicit (instructions are given) or implicit (when revision is always carried out the same way, specific instructions are superfluous). A revision brief complements the translation brief, but is not a substitute for it. As a result, a revision task can be described as being governed by both the translation and revision briefs.

Translation and revision briefs are structural elements that shape a revision task. Besides the structural framework, the task may be influenced by personal decisions, as shown by my study. During the interviews, participants stated that they do not always behave the same way. Data analysis highlighted six factors that may influence the revision task. Of course, not all factors apply simultaneously. Furthermore, the study – which is qualitative – does not allow factors to be ranked or frequencies to be identified.

First, revisers may take into account how much time is available. The reality is that if you do not have much time, you have to concentrate on the essentials, i.e. solve the most important problems. Conversely, if you have more time, you can go into more detail. This factor is not specific to revision or even to translation.

Second, revisers may not behave the same way with all texts. Concretely, they may pay more attention and make more changes to high-stakes texts and, conversely, make fewer changes to texts with a more minor impact. Once again, this is not a revision issue – but it is a translation issue. Not all translations have the same quality requirements, depending on their nature or use.

Third, revisers may take into account what they know about the translator when revising. This is a major trend observed among the group of interviewees: only four revisers said they try to free themselves from their prior knowledge or do not try to find out who translated the text. Adapting to the translator means paying particular attention to the person’s flaws and perhaps paying less attention to the areas in which they are more skilled.

Fourth, revisers may adjust their changes to the quality of the draft translation in two different ways. First, when dealing with a poor-quality translation, revisers may make more modifications and rework the translation because they have abandoned the principle of moderation. Conversely, revisers may intervene more with good translations by changing features that they would not modify if the draft translation were average or poor, because they have more time to go into detail or the risk of offending the translator is lower.

Fifth, revision may be governed by a department’s revision aims, or the goals that the revisers have set for themselves. In one department of the study, revision is clearly intended to improve the skills of junior translators. Revisers insert stylistic improvements so that translators can see how they can improve the text, which is an approach akin to didactic revision, as defined by Horguelin (see for instance Horguelin and Brunette 1998, 4, 50). In other departments, interns or young colleagues can be revised more extensively so that they can fully benefit from revision.

Sixth, three interrelated factors can influence the revision process: the nature of the changes (corrections vs. suggestions), the person who is responsible for the final product and the work process step that follows revision. In most of the departments involved in the study, the revised text goes back to the translator for validation and revision is advisory; consequently, changes are suggested by the reviser, and the translator is the one who is primarily responsible for the text. In these teams, however, revision may also be corrective, particularly when the translation has been outsourced or a colleague is out of the office when the translation has to be delivered. Revisers do not behave the same way when the revised translation goes directly to the client as opposed to when the text goes back to the translator. For example, when delivering the text directly to the client, they cannot make suggestions.

4.3. Revision as a difficult task, especially for beginners

This subsection focuses on the textual dimension, which is at the core of revision, since revision consists of verifying a translation, i.e. finding problems and changing the text in order to solve them (see Mossop’s oft-quoted definition 2020, 115).

Both empirical studies on the product of revision as well as my own study show that revision is difficult. The research that has been carried out on the quality of revision is sound and substantial enough to provide a scientifically valid picture. Two major trends emerge: reviewers tend to (i) overlook problems (Daems and Macken 2021; Künzli 2006a, 2007b; Nitzke and Gros 2021; Robert 2012; Schaeffer et al. 2019; van Rensburg 2017) and (ii) make unnecessary changes (Daems and Macken 2021; Künzli 2006a, 2007b; Nitzke and Gros 2021; Robert 2012). In addition, errors may be introduced during revision, but this trend is less significant than the other two (Künzli 2006a, 2007b; Robert 2012; van Rensburg 2017). Since revisers tend to overlook problems, make unnecessary changes and, to a lesser extent, introduce errors in the translation, the added value of revision can be limited. In Künzli’s experiment (2007b, 120–121; Künzli 2006a, 93–94), four of the ten revisers checking a legal translation and five of the ten revisers checking an advertising translation lowered the quality of the text.

My indirect, interview-based study confirms the overall picture that emerges from direct textual observation. The analysis of the interviews led to a long list of potential difficulties, including the difficulty of restricting the volume of changes, deciding whether to intervene in the translation or not, classifying or explaining the problems spotted and staying focused while revising (see Riondel, A. Manuscript submitted for publication-a). The difficulty of revision was also referred to in general, whether directly or indirectly. Of the six revisers at the international organisation (who have from 11 to 32 years of experience in translation), three insisted very much on the difficulty of the task and two said that revising was difficult in the beginning, while the final person instead finds revision quite easy. Two participants from other teams who had been revising for at least six years said that there was nothing easy about revision.

Revising seems especially difficult for young revisers, i.e. professional translators who have revised for less than three years. According to my study, young revisers may have trouble finding the right balance: they make either too many or too few changes, the latter resulting from feelings of inhibition. Making too many changes is the most common problem that emerged. Twelve of the 22 revisers who have at least four years of experience and are not heads of department said that the number of changes they made in revision has diminished over time (the topic was not addressed in every interview). In addition, two participants working in the international organisation – one translator and one reviser – said that young revisers tend to make more changes than revisers with more experience. This finding corresponds to what Mossop (2020, 201) writes about beginners who tend to modify too much, rather than not enough. While my study supports Mossop’s statement, it points to another, secondary model that involves young revisers who do not dare to make changes to texts. This tends to happen in situations where people have to verify the work of older or more experienced colleagues. In this model, changes increase with time. Three participants – with 1.5, 6 and 7 years of experience of revision – said they make more changes now than when they started revising.

4.4. Revision as a social activity

Considering only the textual component of revision results in too narrow a view, because revision is also a social activity for two reasons in particular. First, revision is collaborative, because revisers take on the work of translators and the revised translation is the result of a joint effort by two people. The second reason is that communication (at least in some situations) and psychosocial aspects play an important role. Examples of psychosocial phenomena are the security that translators feel when they are revised, or the negative experience of receiving a corrected text (for a detailed discussion, see Riondel, A. Manuscript submitted for publication-b). Considering revision to be a social act means looking at the actors involved, namely translators (or revisees) and revisers.

Being revised is not something that should be taken for granted; it is difficult, unpleasant and not very appreciated by translators (Horguelin and Brunette 1998, 65; Samuelsson-Brown 1996, 116; Scocchera 2015, 179). My interview study shows that young translators feel strong emotions when receiving their revised texts, and that even experienced translators can have difficulties with the act of being revised. More precisely, the study indicates that translators generally learn to tolerate revision during the first years of professional practice, as if they need to develop a ‘competence of being revised’, which consists of being able to:

• accept criticism,

• be open to discussion,

• assess the changes made in the translation, and

• comment on one’s work in a way that is useful to revisers when translating.

Commenting on one’s own work means learning what information, doubts and explanations to provide to the reviser and in what manner. In addition to promoting communication with the reviser, it can also foster the translator’s reflexivity on their work, which can be seen as positive.

On the revisers’ end, the social dimension encompasses two aspects: justifying the changes that are made and, more broadly, communicating in an appropriate manner. In revision, it is crucial to make changes that are justified, as put forward by handbooks (Horguelin and Brunette 1998; Mossop 2020). The study conducted shows that justifications can also (i) reduce the bitterness induced by revision and (ii) help translators understand the changes (by simply looking at the changes, justifications may not be apparent to translators). Justifying changes is therefore useful for revision from a textual point of view (reducing unnecessary changes), from a translator development point of view (translators improve more when they understand the reasons behind the decisions that have been made), and from a social point of view, as respectful explanations reduce feelings of bitterness.

In addition to being able to justify the changes, future revisers should learn how to communicate. As put forward by the study, revision involves a great deal of contact, at least in the context of translation departments. According to participants, most of the communication takes place after revision, in the form of explanations as to why changes have been made or more general feedback for the translator (when revisers assume a mentoring role). Throughout all this, revisers should communicate in an appropriate way. Even if everyone must be respectful, it is particularly true for revisers, as revision is a form of evaluation. Revising means assessing the translation choices and fixing any problems that are identified. In other words, the revisee is being criticised, which means that the reviser must be particularly careful.

5. Impact on teaching

Based on the findings of the previous section, I initially formulated fifteen proposals for teaching, which were submitted to three revision trainers (see Section 3). This step did not lead to removing any proposals, as none were evaluated negatively by the three experts. It did, however, result in the merging of two proposals, the deletion of parts of proposals, and some reformulation. In other words, the interviews clarified the proposals, in addition to making it possible to determine which were more or less important. The resulting fourteen proposals can be broken down into three groups: content to be taught, revision exercises and activities related to the social aspects of revision (Table 1).

Table 1. Revision teaching proposals that aim to take into account the diversity, situatedness, difficulty and social dimension of revision.

Content to be taught	Revision exercises	Activities related to the social aspects of revision
1. Making students aware of the reality of professional revision, including the diversity	4. Revising with different time constraints, including focussing on major errors	11. Teaching the principles of good communication
2. Bringing up the difficulty of revising	5. Revising translations of different quality	12. Doing peer revision
3. Describing the two models of young revisers (changing too much or not enough)	6. Revising people with different profiles	13. Organising post-revision talks
	7. Testing monolingual revision	14. Asking students to provide comments to the reviser
	8. Doing advisory and corrective revision
	9. Dividing the task into subtasks and asking students to perform only part of the task
	10. Asking students to justify the changes made

The first column lists content to be taught in class. Proposal 1 is based on the diversity described in sub-section 4.1, which requires that students be prepared to work in a vast array of configurations. To do so, it is important to describe the different models and modalities used in the professional world. The three teachers introduce students to the professional reality of revision and consider it important, although one said it should be kept short. Proposals 2 and 3 arise from the difficulty of revising described in sub-section 4.3. On that basis, I suggest that teachers tell students that revising is difficult, notably in order to reassure them, and present the two main failures of young revisers: changing too much or too little. Both proposals were praised by the experts.

Column 2 groups together revision exercises, which have been drawn up based on the results of the study. Proposals 4 and 5 pertain to the situated nature of revision. As put forward in sub-section 4.2, revision varies from situation to situation, and different factors can influence the task. Two kinds of exercises can be implemented in the classroom in order to make this clear to students. First, revising with different time constraints: high, medium, or none. As put forward by Teacher 3, time pressure can help students focus on major errors and thus ignore less important items. Next, using draft translations of different quality to show that the number of changes varies depending on the situation. Both proposals were validated by the teachers, who use them in their courses. For example, Teacher 2 uses low-quality translations at the beginning of the term, because they prove to be easier for students, and better translations later on.

Proposals 6–8 derive from the diversity of revision described in sub-section 4.1. Overall, they were deemed less important by the experts, but nonetheless useful. As for revising texts translated by people with different profiles, asking students to revise a member of the teaching staff (e.g. the teacher themself) is a type of exercise used by one teacher and considered interesting by another, notably to minimise the number of changes by students who tend to change too much. One of the experts, who teaches revision into English, also uses translations from native speakers and non-native speakers. Monolingual revision was considered interesting to test, rather than to train extensively. Finally, proposal 8 consists of preparing for situations where all modifications are corrections, as the revised text is delivered directly to the client (corrective revision) and situations where the revised text goes back to the translator for validation (advisory revision). Two of the teachers implement this proposal by alternating classic revision assignments (corrective revision) and peer revision exercises (advisory revision).

Proposal 9 directly originates from the difficulty of revision outlined in sub-section 4.3. In the classroom, it is possible to reduce the challenge of revising by offering exercises that focus on parts of the revision task, which can be divided up as follows: detecting problems, describing problems, modifying the text so that the problem is solved without adding new ones, and justifying the change. According to the interviews with the teachers, the two most productive exercises are: (i) only detecting and describing problems, (ii) solving problems that are already spotted. Proposal 10 – justifying the change – is linked to the results reported in sub-sections 4.3. and 4.4. It should be considered a general rule of revision and revision teaching, because it distinguishes between necessary and superfluous changes, and helps students limit modifications to essential ones. Proposals 9 and 10 were deemed both very useful and important by the experts.

The last column features proposals that relate to the social dimension of revision and are linked to the results reported in sub-section 4.4. Proposal 11 involves teaching students the principles of respectful communication: speaking for oneself, describing facts, discussing the performance or product (not the person or their skill), avoiding being too prescriptive, and being open to dialogue. There was consensus among the experts regarding this issue. Proposals 12 and 13 often go together since peer revision generally gives rise to a talk where potential changes are discussed. Two of the teachers have their students do this in class. The third teacher, however, finds peer revision too sensitive. Instead, she organises a little role-play: she gives students revised translations and asks them to play different roles, like an awful reviser or a translator who refuses all changes. Finally, teaching students to write comments that are useful for the reviser was approved by the experts, even if they noted that this is more a translation issue (they generally do it in translation courses rather than in revision courses).

There is no one-to-one correspondence between the above-described proposals and classroom activities. Classroom activities often deal with several of the above proposals. For example, Teacher 1 organises elaborate peer revision exercises. She first asks a student to choose a text to translate, which is then translated by one of his or her colleagues (at home). In class, the person who has chosen the text revises his or her colleague’s translation. After revision, the translator receives the revisions and takes a few minutes to review them. Finally, the two discuss the text and work out the final version. This one activity covers proposals 8, 10, 11, 12 and 13.

The teacher with the most experience in revision training stressed during the interview that, while every proposal is interesting, classroom time is limited, which may result in prioritising the proposals. According to the interviews, the most important proposals are:

• making students aware of the reality of professional revision (No. 1)

• revising with different time constraints (No. 4)

• dividing the task into subtasks and asking students to perform only part of the task (No. 9)

• asking students to justify the changes made (No. 10)

• teaching the principles of good communication (No. 11)

• organising post-revision talks (No. 13)

To this list, I would add describing the two models of young revisers (No. 3), which requires very little time while being potentially very useful to students.

6. Discussion and conclusion

In this article, I have used an interview study to describe four aspects of revision that are important from the perspective of revision teaching and independent of language pairs: the diversity, situatedness, difficulty and social dimension of revision. I have then come up with proposals for teaching, which were submitted to three experienced revision trainers during additional interviews.

The fourteen proposals for teaching can be put into perspective with the most comprehensive model of revision competence (Robert, Remael, and Ureel 2017). They can be said to contribute to the main feature of the model, i.e. the ‘strategic subcompetence’. In addition, the first proposal (making students aware of the reality of professional revision) directly relates to the ‘knowledge about revision subcompetence’ in Robert et al’.s model. Next, the third group of proposals (Nos. 11–14) and the requirement of justifying the changes made (proposal No. 9) are closely linked to the ‘interpersonal subcompetence’ and ‘psycho-physiological components’. Manifestly, the proposals made in this article do not cover all the subcompetences required for revision. In particular, as regards Robert et al’.s model, they do not relate to the ‘revision routine activation subcompetence’ or the ‘tools and research subcompetence’ (which is partly specific to revision and partly common with the translation competence). This is logical, since the proposals made in this article do not cover all the content of a revision course; they are only suggestions for teaching based on the findings of an interview study conducted on revision.

The fourteen proposals for teaching place great emphasis on the social dimension of revision, since the original interview study showed that this issue is paramount in revision. It should therefore be prominent in revision teaching as well, especially since students do not always seem to be aware of the interpersonal implications of revision (Scocchera 2020, 36). Addressing the social dimension notably means that a revision course must include both revising and being revised. The original interview study highlighted that being revised cannot be taken for granted, which is why I have suggested that the ‘competence of being revised’ be defined as accepting criticism, being open to discussion and being able to weigh the changes that have been made. In my opinion, a revision course should involve learning how to be revised, a dimension that has received little attention in the literature, although it is mentioned in an early text by Mossop (1992).

The first limitation of the article is that it is not attached to any particular language pair, even though revision is primarily about linguistic choices. Other limitations relate to the methodology. The original interview study, though far-reaching, is qualitative, which means that participants are not intended to be representative of a larger group; moreover, the data produced in interviews are indirect. In addition, the validation step of the teaching proposals is of an exploratory nature, since only three experts were interviewed. The most pressing avenue of research would therefore be to implement the proposals and assess them. This testing may take various forms, like asking students for feedback at the end of the course to identify what works and what does not, or surveying students a few years after the course to determine whether it prepared them well for professional life.

Acknowledgments

I would like to express my sincere gratitude to all the participants, who gave me some of their precious time, and especially to the heads of departments who very generously supported my study. The support of the three revision trainers was also greatly appreciated.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This work was partly supported by the Swiss National Science Foundation under Grant [P1GEP1_195089].

Notes

1. https://www.oed.com/dictionary/didactic_n?tab=meaning_and_use#6731126. Accessed on 22/09/2023

2. https://www.oed.com/dictionary/didactics_n?tab=meaning_and_use#6731254. Accessed on 22/09/2023

3. https://provalisresearch.com/products/qualitative-data-analysis-software/freeware/. Accessed on 12/05/2023