Corps de l’article

Every translation entails a dual form of reading: the partial or integral reading of the source text (ST) before the actual translation activity, and the reading of the target text (TT) while it comes into being on a paper page or electronic file. Both forms trigger that series of inferential, interpretative and creative acts that any translator performs in order to reproduce his/her idea of the text into another language, adopting foreignizing or domesticating strategies to produce a translated text that strives to recreate for the target reader the same experience as the reader of the original text. Because of the intimate, unique relationship with the text he/she is working on, the translator is widely considered the first, real reader of a text and translation has often been described as the most authentic form of reading. It is so for Calvino, who sees in the translator’s act of reading a maieutic activity which brings forth what the author of the text sometimes even ignores of himself/herself as an individual and a writer:

Tradurre è il vero modo di leggere un testo: credo sia già stato detto molte volte; posso aggiungere che per un autore il riflettere sulla traduzione di un proprio testo, il discutere col traduttore, è il vero modo di leggere se stesso, di capire bene cosa ha scritto e perché.[1]

Bufalino focuses on the peculiarities of the translator’s reading and identifies in its passionate involvement with the text something similar to a love affair:

Il traduttore è con evidenza l’unico autentico lettore di un testo. Certo più d’ogni critico, forse più dello stesso autore. Poiché d’un testo il critico è solamente il corteggiatore volante, l’autore il padre e marito, mentre il traduttore è l’amante.[2]

Finally, Vaucluse[3] underlines how the clear-sightedness of the translator’s reading may be so stark and bold, so dangerous in its authenticity, that all authors fear the translator because he/she is the only “lecteur lucide.”

Reading, however, is at the core of another stage of the translation process, that is, revision. In the Italian definition of the term “revision,” the act of reading or re-reading (“rilettura”) is actually used as a synonym for the revision activity itself, a synonym that can be interpreted in two different, yet complementary ways: it may refer to any subsequent, “further” reading of the same text, or to a completely “new” reading, with a novel view of the text. This duality of meaning corresponds by and large to two main types of revision and the reading activities they involve: self-revision (implying one or more further readings of one’s own translation) and other-revision (where a translation is read through the new eyes of someone who is not the translator). The reading activity underlying both forms of revision is always the first in a series of analytical, critical and operational steps aimed at improving the quality of the translated text. Translator and reviser, however, proceed in different ways, with different tools and sometimes with different objectives in mind. As a matter of fact, the self-revising translator reads the product of his/her translation activity for a second or third time—this being a further reading of something already known—and because of his/her intimate knowledge of the text, he/she might not be able to look at the translation with new and fresh eyes. On the contrary, the reviser reads the translation for the first time, and thanks to this “premiere view” of the translated text, the revision activity will have the fresh and impartial quality it needs to be performed at its best. We could say that while the translator is the most careful and authentic reader of a text, the reviser is, time-wise, the very first reader of the translated text, and a special reader at that, because he/she combines two separate but complementary reading attitudes and roles: the “naïve” and the “professional” reader. The attitude of the naïve reader is that of any end-reader of the translation once it is published, a person who probably does not know the original text or the original language, who sees the text not as a translation but as a product of his/her own language and reads it for informative or entertaining purposes. What makes the reviser also a “professional” reader is the critical, corrective and improving stance he/she takes on before the translated text, as well as the fact that his/her reading activity is agreed upon by contract with a commissioning agent (the publisher), who might also provide indications on specific house-style requirements and/or reader expectations to be met. Finally, the professional reader-reviser approaches the translated text with a “comparative” frame of mind: in other words, he/she is fully aware of the existence of an original text and his/her reading will constantly move back and forth between the two texts to check on consistencies and discrepancies, to identify mistakes, and to underline possible improvements or raise controversial issues.

While translator and reviser share some sort of “privileged” reading, their different relationship with the translated text makes for significant differences in the way they actually read and the aim of this work is to highlight the peculiarities of reading or “rereading” in both self-revision and other-revision, offering a brief overview of some significant Translation Studies research in this field and providing first-hand data on the professional practice of self- and other-revision in the Italian publishing market.

Reading and Translation Revision in Scholarly Research

Over the last fifteen years or so, Translation Studies research has shown an increasing interest in revision as a key stage of the translation process and the studies that have been conducted so far fall into two main categories, depending on the chosen object of investigation: the self-revising translator and his/her revision activity (self-revision) or the external reviser and his/her revision activity (other-revision). In both cases, research has been conducted using qualitative and quantitative methods and a variety of approaches, sometimes also crossing the boundaries of Translation Studies and moving towards cognitive, writing and editing studies. All these different perspectives, however, share a similar interest in the reading stage of the revision activity and what follows is a selection of those scholarly contributions both on self- and other-revision where a specific focus on reading is provided.

In one of the earliest contributions about the process of self-revision, Mossop[4] offers a clear and detailed account of this activity, dividing it into three separate stages, two of which are specifically carried out through reading. As a first step, Mossop envisages the monolingual reading of a small extract of the translation—without looking at the source text—as it can help identify general problems of language and style. As a second step of the procedure, Mossop suggests reading an extract of the translated text followed by a comparison with the source text. Thanks to this comparative activity, specific omissions and mistranslations can be identified. It also helps to form an overall idea of the quality of the translation. The self-revising translator may then choose whether to proceed with revision proper or to repeat these two “reading” steps for subsequent portions of the text. As the revision activity unfolds, the self-revising translator may have to reconsider choices previously made or adopt different strategies and solutions, while deliberately leaving as a third and last step the final decision on specific issues or translation/revision problems that are particularly hard to solve. In a more recent publication,[5] Mossop confirms the central role of reading in revision by defining revision itself as a “primarily reading exercise,” as opposed to what revision is more commonly thought of, i.e. a primarily writing or correcting task, and he also identifies the “ability to read very carefully” as the key skill of a reviser. Speaking about self-revision in particular, he suggests that every translator should perform at least one “full unilingual re-reading” of his/her translation followed—time permitting—by a comparative re-reading, as in the three-step approach illustrated above.

A specific problem encountered by any self-revising translator while rereading his/her own work—i.e. the difficulty of looking at the translation with fresh and objective eyes—is addressed and explained by Hansen:

One reason why self-revision is difficult is that people fall in love with their own formulations. The same myelin threads are used again and again. The space of time between writing and revising the translation, looking at the task with ‘fresh eyes’, plays an important role here.[6]

As a matter of fact, the translator’s involvement with the text he/she has produced, the deep knowledge of all its strengths and weaknesses is such that any subsequent reading of the translated text is near-sighted. Putting as much time as possible between the drafting and the revision stage of one’s own translation activity, a period also known as “drawer time,” is an unquestionably efficient strategy, possibly the most efficient of all, if one is to achieve the necessary detachment and objectivity to look back on one’s own work.[7] However, translators do not only lack objectivity towards their own work: they often also lack time, or maybe the time they have is not enough to put a distance between themselves and the translation. In order to compensate for this lack of time and objectivity, translators more or less consciously adopt alternative “distancing” strategies. Chesterman and Wagner[8] define distancing as a “stepping back mentally from what you are creating to get a better perspective on it,” and they advise on strategies that, when applied during self-revision, can produce a kind of “artificial forgetting . . . a clearing of the mental screen . . . in order to get a naïve and fresh view” on the translated text. When reading through and polishing one’s own work, then, a translator should first of all “learn to forget.” This is a very interesting approach to reading, as the translator is advised to become fully aware of how the experience of reading one’s own translation differs from the experience of reading for the first time something that someone else wrote. Once the differences in terms of body posture, physiology and internal reading strategies are identified, the self-revising translator should “apply the ‘first-time’ strategy when reading through [his/her] translation.” Chesterman and Wagner offer other efficient “reading strategies” for the self-revising translator: changing the medium (i.e. from screen to paper), starting to read at some point in the middle of the document, reading the text aloud to someone else or pretending to be someone else while reading one’s own translation.

Reading is obviously part and parcel of other-revision too, and the different modalities in which reading is performed can affect both the revision process and its product. This has been the topic of investigation for several works on other-revision, starting with Horguelin and Brunette’s practical handbook[9] and followed again by Mossop. In an attempt to describe an “ideal” revision process, Horguelin and Brunette identify a three-step activity consisting of a reading of the ST, followed by a comparative reading of ST and TT (what they refer to as “bilingual revision”), and finally by a correction and re-reading of the TT. On the contrary, Mossop advocates for “unilingual revision,” that is, the reviser’s reading of the target text alone, going back to the source text only when the reviser detects a problem and subsequently makes a change. About the pros and cons of this type of unilingual re-reading, Mossop says it “may well produce a translation that is not quite as close in meaning to the source as a comparative re-reading will produce. On the other hand, it will often read better because the reviser has been attending more to the flow and logic of the translation.”[10] Another very interesting contribution on the role of the procedure chosen by the reviser and the order in which he/she performs it comes from Künzli[11] who, in line with Mossop, seems to defend the “unilingual” approach when he says, “it might be an advantage to start by reading the draft translation without looking at the source text,” because in this case revisers “have the unique opportunity to avoid coming under the spell of source-language structures.”

Robert[12] conducted a particularly thorough and articulated research work on the impact revision procedures and strategies may have on the revision product (that is, its quality) and process (its duration and potential for error detection). By using a mix of data collection tools (TAPs, keystroke logging, revision product analysis and retrospective interviews), Robert compares the revision product and process of sixteen professional revisers when varying the revision procedure and its reading phase in particular. The four types of revision procedure used as variables are the following: one monolingual proofreading[13] without consulting the source text, except in doubt; one bilingual proofreading; a bilingual proofreading followed by a monolingual proofreading; and a monolingual proofreading followed by a bilingual proofreading. Following data analysis and interpretation, Robert concluded that procedure may indeed affect revision quality, revision duration and its error detection potential. Reading in other-revision was also researched by Künzli,[14] who analyzes not so much the type of reading involved (monolingual or bilingual) but the order in which ST and TT are read before proceeding with revision proper and how they may affect the quality of the revised translation. Also, Rochard[15] investigates different revision procedures to find out whether an ideal one may exist, and he particularly analyzes a combination of the translator’s reading aloud of the his/her translation with a simultaneous comparative check of ST and TT by an external reviser (thus a combination of self- and other-revision), describing it as

formidable pour apprendre à détecter à la fois les qualités de sa propre traduction et à les défendre par la conviction que l’on met dans la lecture mais aussi les imperfections qui viennent entraver cette même lecture à haute voix.

This collaborative form of reading is also described by Parra Galiano[16] as revisión de concordancia while the cross, monolingual reading by two different translators of the same translated text to identify and correct formal errors is called lectura cruzada. In a later study, Parra Galiano[17] investigates reading in revision as part of a wider research work on quality principles and parameters. According to the scholar, while the first, necessary condition to produce a quality revision is to know the “translation brief,” the second logical and practical step is

la lectura del TL como si de un texto de origen (TO) se tratase. En el supuesto de que en esa primera lectura se detected errores, a priori, un traductor [or reviser] con amplia experiencia puede deducir si merece la pena o no corregir la traducción, o sea, si la revisión del TL serà o no rentable.

Reading, here, bears an “economic” impact on the revision activity, because it helps estimate the time needed for revision as well as measure the cost-effectiveness of the revision effort. With a view to making the most of working time—especially when translation and revision work take place in a structured environment—Parra Galiano underlines that starting by evaluating the quality of a translation through a monolingual reading of the target text allows one to spot immediately “los textos de infima calidad o muy mediocres, lo cual supoine ganar tiempo.”[18]

The professional environment introduced by these last contributions will be the object of the following section, in which the different forms of reading performed by professional translators and revisers will be illustrated by highlighting their peculiarities in terms of how, where, when and why they are carried out.

Reading and Translation Revision in the Professional Context

Despite differences in quantity and quality, the research work illustrated in the previous section deals with reading as a “universal,” characterizing trait of revision practice. This section will instead focus on a specific professional field, that is, revising a translation for the Italian publishing market, but because of the variety of translated texts and practices involved, we believe that the data offered for analysis and interpretation might be interestingly triangulated and compared with observations coming from research work on revision in other working contexts, as can be found in Hernandez-Morin,[19] Rasmussen and Schjoldager,[20] and Bertaccini and Di Nisio.[21] In line with the section on scholarly research, data herewith presented refer to both self-revision and other-revision practice, and they were elicited through a questionnaire-based survey conducted in 2013 among English>Italian professional translators and revisers working in the Italian publishing market.[22] By applying empirical and qualitative research methods, the survey aimed at collecting data on three main investigation parameters, namely the agents of revision, the revision product and the revision process. The survey followed Lasswell’s communication model in structure and consisted of two separate questionnaires (translators vs revisers) articulated around five main investigation areas: what, who, how, where and when, and why in editorial/literary translation revision.

The first section of the survey addressed the translators’ view of self-revision and the data collected through the questionnaire confirmed the tendency already mentioned at the beginning of this work, that is, the use of “rereading” as a synonym for the revision activity, a term which was given as an answer by 40 % of the participants when asked to provide their own definition of revision.

A second, significant part of the survey was devoted to understanding how revision is carried out, and a good number of questions focused on the “rereading” process, in particular on the kind of medium—screen or paper—used. The answers collected from the 55 participating translators show a high level of variability, mostly related to text length and difficulty, and to the time available. However, they can be brought down to four main categories: rereading done exclusively/mostly on screen (37/55, 67.3%); rereading done mostly on paper with subsequent introduction of text changes on file (4/55, 7.3%); a first rereading on screen, followed by one or more rereadings on paper (12/55, 21.8%); a first rereading on paper, followed by one or more rereadings on screen (2/55, 3.6%). Most translators fully detailed their answers and provided extremely interesting insights into their reading practices, explaining procedures and personal preferences. In particular, they confirmed the need to adopt distancing strategies during the rereading stage, one of the most popular being a change in text visualization. This strategy is resorted to by the majority of respondents, those mainly working on screen, and it consists in changing the visual aspect of the file on screen by altering format, character size and type, spacing, page layout, and orientation. By showing a different view of the translated text, this strategy also allows for a different reading attitude, which one respondent described as “relaxed reading”: in other words, the strategy enables the translator to take on the role of the naïve reader and in fact, once the text on screen has changed its appearance, the self-revising translators is actually seeing and reading that new text for the very first time. This new attitude towards one’s own translated work may help the translator identify errors and shortcomings that had not been previously detected because, having spent so much time and effort on the translated text, he/she already knows what to expect and fails to grasp any better or even possible alternatives.

Translators were also asked whether they used a reading aloud strategy during their self-revising procedure. About a third of the total respondents (20/55; 36.4%) said they regularly resort to this type of “control” reading, while 34.6% (19/55) read aloud only occasionally, either because they do not have enough time or because they do it only for specific parts of the translated text where sound and rhythm effects are particularly important or hard to recreate. The remaining 29% (16/55) said they never read their translations aloud. To better understand the reasons behind these contrasting trends on the practice of reading aloud, it might be worth analyzing some full answers given by participants as those shown below:

I do read aloud my translation but not all of it, because I do not want to write something that sounds too much like spoken Italian in terms of rhythm and prosody, because the language we speak is not the same as the language we write. I usually read aloud, preferably to other people, only those parts of the translation where I sense there might be a problem of understanding or readability, or to loosen up stiff formulations.

I do not read aloud, but I have been using for some years now a very useful tool, especially during my second round of revision: I ask the publisher (sometimes I buy it myself) for the audio version of the book and I listen to it while I am rereading my translation. In this way I can feel if I got the rhythm right, as well as the style and the author’s intention. It is good to hear the actor or actors’ voice and it is an excellent way to detect minor mistakes or involuntary omissions.

I never read aloud, but during the last, final rereading, I “whisper” my translation to myself: it’s a good way to hear how it works in terms of readability.[23]

In another section of the questionnaire, translators were asked to give information on where and when their rereading activity takes place. A very high percentage of translators (81.8%; 45/55) said they translate and revise at the same working place and position, mainly because they work from home and do not have alternative rooms to dedicate to these separate activities. Only 10 out of 55 translators said that, although it is not always possible, they tend to move their revision activity elsewhere. This change of place is explained as the need to put more distance between oneself and the translated text (yet another distancing strategy!) and this works best when the change is radical (indoor vs. outdoor). However, minor changes (desk vs. sofa or bed) may also help, as they contribute to adopting a more relaxed posture and attitude. Again, it is interesting to note how this strategy aims at recreating a reading situation that is closer to entertainment than to professional work. In other words, it is an attempt to combine both sides of the reviser’s role as a naïve and professional reader.

Translators were also asked to indicate when the rereading takes place during the translation process, and they were provided the following possible answers: all along the translation process, mid-way, only after the draft translation is completed, in different stages, or other. Answers outline two main groups of translators: those who reread only after the draft translation is completed (32/55, 58.2%) and those who reread in different stages, both during the translation process and at the end (21/55, 38.2%). Among those translators who describe their rereading activity as distributed over the translation process, it is possible to identify two main trends: 6 translators out of 21 (28.5%) reread and revise their translation after variable portions of text (a given number of pages, or chapter by chapter), while 4 translators out of 21 (19%) reread and revise every day the pages they drafted the day before and only then do they proceed with the translation work.

As to the “why” of self-revision, translators were asked to describe the aim of their rereading activity. The answers partly overlap with those given at the beginning of the questionnaire—when translators were asked to provide their own definition of the term “revision”—but they rather focus on the actual, short-term objectives the translators have in mind when doing their last reading of the translated text before delivery, objectives which might not be necessarily the same as those envisaged at the beginning of the revision activity. A recurring element in the translators’ answers is the objective of fluency and increased readability, a refinement of the language so that the text is more easily received by the reader and any traces of “translationese” are eliminated (35.9%). Another common reading objective is the translation check and correction of both form- and content-related errors (25.6%), while the third most common objective is the overall improvement of the text as a “commercial product,” that is, with an eye to the prospective market and to the expectations of the final reader (16.6%). Although in fewer occurrences, other reading objectives were mentioned in the answers: consistency/coherence and stylistic homogeneity (6.4%), correspondence to the author’s style (6.4%), and finally, conformity to the publisher’s house-style (2.6%).

The last part of the questionnaire aimed at identifying any peculiarities the translators attributed to or expected from the reading activity of the reviser as compared to their own rereading of the translation. Translators were asked to indicate whether the reviser usually detects something more or something less than they do and how the reviser’s better/worse view of the translated text differs from theirs. Data show a quite clear situation: thanks to his/her “fresh” eye and to the lack of any emotional involvement with the translated text (see Hansen, 2009), the reviser is considered a better “detector” of stiff language formulations that affect the flow of the language (29.7%), typos/misspelt words (15.6%), calques (12.5%), idiosyncratic formulations, personal preferences and styles, regional language (12.5%), unwanted rhyming words and repetitions (12.5%), interpretation errors (9.4%), and errors of content or overall style (7.8%). Compared to the advantages mentioned above, the number of disadvantages in the reviser’s reading is considerably low—maybe thanks again to those “fresh eyes”—and they fall into two main groups: a weaker understanding of the rationale behind certain translation choices and strategies, maybe due to a knowledge of the text that is not as intimate as the translator’s (67%), and a weaker ability to grasp any network of recurrent meanings and images within a text (33%).

A similar questionnaire was submitted to revisers, who were asked to answer on their reading activity during other-revision. As expected, the term “rereading” does not appear to be used as a synonym for revision, and this is because revisers are not reading for a second time a text they already know. However, they still describe revision as a type of reading, namely a “critical” one. As a way to investigate their relationship with the source text, revisers were asked whether any full or partial reading of the ST was done before starting to revise and which advantages or disadvantages they attributed to their choice. The same question was asked with reference to the full or partial reading of the translated text. The answers indicate that 40% of respondents (10/25 revisers) read the source text before starting to revise (8% of them read it in full, 32% only in part, 20–30 pages at most). The remaining 60% do not read the source text before starting to revise and the reasons for this procedural choice are mostly lack of time and cost-effectiveness. The same distribution of percentages applies to the reading of the translated text. Although a preventive reading of both the source text and the translated text is not at all a common practice among revisers, this does not mean that, provided with enough time and money, they would not prefer to read as much as possible before starting to revise. Analyzing in detail some of the answers, one can see that revisers fully agree on the usefulness of such a reading but cannot afford it. Advantages and (a few disadvantages) of ST and TT reading before revision are illustrated in these contributions:

I would like to read the source text before starting to revise, but I can hardly do it because of the little time I have. An advantage would be the opportunity to have an overall grasp and view of the ST to identify its dominant trait or traits and therefore keep it mind as guidance when analyzing the TT and suggesting changes.

I usually start to revise by reading the ST; then I read the translation Unfortunately it takes time, but it is a good way to detect mistakes and misunderstandings that would remain otherwise hidden under the veil of a well-written translation.

I quickly read a part of the translation—at least 20 pages—before starting the line-by-line comparison with the ST. This first reading is important to have a general idea of the translation as a text in itself. It is not too time-consuming and therefore it can be justified by the revision fee.

I partially read the translation to understand the translator’s style and approach. The good thing is that already after a few pages you can anticipate what the revision work will be like; the bad thing is that sometimes this quick reading may prove misleading.

As anticipated by these authentic contributions, the reviser’s reading is mostly a comparative one and that is why revisers where asked to describe the kind of ST-TT comparison they make (word-by-word and line-by-line; sampling; only in the presence of translation problems, following specific indications or other). Nearly half of the respondents (52%, 13/25) said they always revise by comparing ST and TT line-by-line (full comparative reading), 20% say they only compare samples of both texts, while the remaining 28% have a variable approach. Lack of time is once again the main reason behind the choice of not comparing ST and TT in full. However, a particular good quality of the translation and the reliability and good reputation of the translator may also convince the reviser that a full, integral comparison of the two texts is not necessary.

Revisers were also asked to indicate the medium/media they use when reading the translation. Answers show that 64% read and revise only or mostly on screen; 16% read and revise mostly on paper and then introduce text changes on file; 12% do a first reading on screen, then a second or further reading on paper; finally, 8% read on paper first, and then do a second or further reading on screen. Looking closely at the answers, one can see that the choice of one medium or the other—and the order in which they are used—depends on a series of variables. Generally speaking, it appears that when the translated text is already very fluent and easy to read, the reviser indulges in the luxury of reading on paper. On the contrary, when a first, quick reading of the translated text indicates a hard revision work to come, the reviser seems to prefer the more comfortable and more efficient working environment the electronic file and computer screen can provide. As it was with translators, revisers mention the use of the audio file, as detailed in the answer below:

I only read and revise on screen. I open both files and I view them side by side. Alternatively, I use a reading software to read the translation aloudI choose a suitable reading speedwhile I scroll down the ST.

The “reading aloud” possibility was the object of a further question to revisers, who answered as follows: 20% said they regularly read the translated text aloud, 44% never read aloud the translation they are revising, and the remaining 36% read the translation aloud only occasionally or when the sound and rhythm effects are particularly hard to recreate.

As to the where and when of their reading activity in revision, revisers were asked to indicate their usual work place and any change of working setup during the revision activity. At the same time, they were asked to describe when the revision happens during the translation production process. The vast majority of revisers work from home (75%) while the remaining portion (25%) work both from home and from an editorial/publisher’s office.

Time-related data were obtained by asking revisers to indicate when their revision work actually happens (only after the translation is completed/simultaneously/all in one/in different segments): for 84% of the respondents, revision work starts only after the translation has been delivered to the publisher, while the remaining 16% start revising while the translation process is ongoing, and this happens when a particular text is to due to be published in a very short time (the translation is delivered in instalments and the revision proceeds at the same pace as the translation) or in the case of a “collaborative translation,” where the work of different translators has to be coordinated and revised to produce a text with a consistent, single “voice.”

Like translators, revisers too were asked to identify the main goal of their reading activity in revision, and it is interesting to see how their objectives slightly, but obviously, differ from the translators’: the prevailing objective is to give support to the translation to improve its overall quality (34%), followed by the objective to improve the readability of the text (27%), support to the translator (23%), support to the author to make sure his/her intention is clearly conveyed (9%), and improve conformity with the publisher’s requirements (7%). It is interesting to note that the negative attitude that many translators seem to have towards the reviser (often stereotyped as someone who thrives on someone else’s errors and shortcomings, someone whose work on the translated text is often triggered by envy, someone who wants to replace the translator and threaten his/her reputation), proves to be only a biased misconception: the reviser’s objective is to move towards a better quality of the translation, and not against anything or anyone. This comes out clearly in the following, concluding contribution, in which revision work is presented as an attempt to bring forth the best qualities of a translation and its author:

Revision is a support to the translator, a form of help. The reviser is a mediator, someone who knows the publisher’s needs and how to combine them with the author’s and the translator’s needs, offering suggestions, corrections, new ways of seeing his/her own work. We could say that the reviser is for the translator what Jiminy Cricket is for Pinocchio: his/her conscience. Without him, the translator might fail or succeed without even knowing it.

Concluding Remarks

The aim of this work was to shed light on the reading activity that is part of any translation revision work. Recognizing different forms of reading, as well as different attitudes and objectives depending on the agent of reading (the self-revising translator or the external reviser), this essay offered a brief overview of scholarly research in the field. Although the aim of most studies was to investigate the revision process in order to identify best practices and best procedures leading to better revision products, results seem to indicate that no absolute recipe for success exists, but this should not come as a surprise, let alone as a failure: if it is true that reading also means interpreting, we must welcome the idea that every different type, form and mode of reading may add a new and valuable perspective to any revision work. As a matter of fact, the data obtained from a survey on professional translators’ and revisers’ reading activity during translation revision, and presented here as a valuable addition to research-based contributions, seem to confirm the complementary character of different forms of reading. This variety and complementarity of approaches also applies to the revision objectives identified by translators and revisers: while such goals as fluency and readability in the target language are widely shared by both groups, translators and revisers may have different specific priorities in mind when performing the revision activity. If translators revise mostly for the benefit of the target text and its relationship with the source text, revisers tend to have a much wider range of interlocutors in mind—both present and absent—when reading the translated text: their revision work is considered an activity in support of the text itself, but also in support of the translator, the author, the publisher, and the final reader too. In other words, the reviser is the mediator par excellence and by virtue of these multiple liabilities, his/her view and work on a translated text always provide additional and valuable insights.

By exposing, both in research and in professional practice, different forms of reading in revision, this work has hopefully raised awareness on their complementarity as well as on the different yet equally necessary roles that different actors play in revision, ultimately striving towards the same goal: to provide the final readers with a text that they will read and enjoy in their very own, special way.