Ebook A Literary Translation in The Making A Process Oriented Perspective 1St Edition Claudine Borg Online PDF All Chapter

A Literary Translation in the Making A
Process Oriented Perspective 1st

Edition Claudine Borg
Visit to download the full and correct content document:
https://ebookmeta.com/product/a-literary-translation-in-the-making-a-process-oriented
-perspective-1st-edition-claudine-borg/
More products digital (pdf, epub, mobi) instant
download maybe you interests ...
Translation of Contemporary Taiwan Literature in a

Cross Cultural Context A Translation Studies
Perspective 1st Edition Szu-Wen Kung
https://ebookmeta.com/product/translation-of-contemporary-taiwan-
literature-in-a-cross-cultural-context-a-translation-studies-
perspective-1st-edition-szu-wen-kung/
The Object Oriented Thought Process 5th Edition Matt

Weisfeld
https://ebookmeta.com/product/the-object-oriented-thought-
process-5th-edition-matt-weisfeld/
Understanding Korean Film A Cross Cultural Perspective

Routledge Studies in East Asian Translation 1st
Edition Jieun Kiaer
https://ebookmeta.com/product/understanding-korean-film-a-cross-
cultural-perspective-routledge-studies-in-east-asian-
translation-1st-edition-jieun-kiaer/
Advances in Corpus Applications in Literary and

Translation Studies Routledge Advances in Translation
and Interpreting Studies 1st Edition Riccardo Moratto
(Editor)
https://ebookmeta.com/product/advances-in-corpus-applications-in-
literary-and-translation-studies-routledge-advances-in-
translation-and-interpreting-studies-1st-edition-riccardo-
Common Data Sense for Professionals: A Process-Oriented
Approach for Data-Science Projects 1st Edition Rajesh
Jugulum
https://ebookmeta.com/product/common-data-sense-for-
professionals-a-process-oriented-approach-for-data-science-
projects-1st-edition-rajesh-jugulum/
German Philosophy in English Translation Postwar

Translation History and the Making of the Contemporary
Anglophone Humanities Routledge Advances in Translation
and Interpreting Studies 1st Edition Spencer Hawkins
https://ebookmeta.com/product/german-philosophy-in-english-
translation-postwar-translation-history-and-the-making-of-the-
contemporary-anglophone-humanities-routledge-advances-in-
translation-and-interpreting-studies-1st-edition-sp/
The Routledge Handbook of Latin American Literary

Translation 1st Edition Delfina Cabrera
https://ebookmeta.com/product/the-routledge-handbook-of-latin-
american-literary-translation-1st-edition-delfina-cabrera/
Retracing the History of Literary Translation in

Poland: People, Politics, Poetics 1st Edition Magda
Heydel (Editor)
https://ebookmeta.com/product/retracing-the-history-of-literary-
translation-in-poland-people-politics-poetics-1st-edition-magda-
heydel-editor/
Making a Mark Image and Process in Neolithic Britain

and Ireland 1st Edition Andrew Meirion Jones Marta Díaz
Guardamino
https://ebookmeta.com/product/making-a-mark-image-and-process-in-
neolithic-britain-and-ireland-1st-edition-andrew-meirion-jones-
marta-diaz-guardamino/
Routledge Studies in Literary Translation
A LITERARY TRANSLATION
IN THE MAKING
A PROCESS-ORIENTED PERSPECTIVE
Claudine Borg
i
A Literary Translation in the Making
This book presents a holistic picture of the practice of an experienced lit-

erary translator working in situ, highlighting the value of in-depth process
studies for the discipline and offering a model for future similar studies.
Bringing together Cognitive Translation Studies (CTS) and literary
translation, Borg interrogates existing assumptions in CTS and sheds light
on the value of a combined look at both cognitive and social processes in
literary translation. The volume extends the scope of existing CTS studies
with its comprehensive examination of the work of one translator and
exploration of the wide range of materials from draft to finished transla-
tion. This unique model allows for a greater understanding of the actions,
decisions, motivations and work practices of individual translators as
well as of their interactions with other participants in the practice of a
literary translation.
Making the case for in- depth process research in illuminating the
dynamics of translation production and working practices, this innova-
tive book will be of interest to students and scholars in translation and
interpreting studies, especially those interested in literary translation and
cognitive approaches.
Claudine Borg received her PhD in Translation Studies (2017) from Aston
University, UK. She is a Lecturer in Translation Studies at the University
of Malta and a literary translator. Her research interests lie primarily in
cognitive translation studies, translator studies and literary translation.
ii
Routledge Studies in Literary Translation

Series Editors: Jacob Blakesley and Duncan Large
Routledge Studies in Literary Translation highlights pioneering research

in literary translation, exploring emerging developments, new voices,
and key issues of relevance in core literary genres. The series questions
the definition of literary translation as a sub-discipline in its own right
with its own particular methodological and theoretical considerations
as well as the extent to which its study extends to genres beyond the
traditional categories of fiction, poetry, and drama. The series extends
its scope beyond Anglophone literary traditions to feature research on
translated literary works across a range of languages as well as the inter-
face between literary translation and such topics as multilingual litera-
ture, literary canons, publishing markets, classics, and digital humanities.
With its dedicated focus on literary translation, this series will appeal to
students and scholars interested in the interface of translation studies and
literary studies, as well as those in related disciplines such as comparative
literature, literary criticism, sociology, and media studies.
Translating Transgender Identity

(Re)Writing Undecidable Texts and Bodies
Emily Rose
The Afterlife of Dante’s Vita Nova in the Anglophone World

Interdisciplinary Perspectives on Translation and Reception History
Edited by Federica Coluzzi and Jacob Blakesley
A Literary Translation in the Making

A Process-Oriented Perspective
Claudine Borg
For more information about the series, please visit: https://www.routledge.

com/Routledge-Studies-in-Literary-Translation/book-series/RRSLT
iii
A Literary Translation in
the Making
A Process-Oriented Perspective
Claudine Borg
iv
First published 2023

by Routledge
605 Third Avenue, New York, NY 10158
and by Routledge
4 Park Square, Milton Park, Abingdon, Oxon, OX14 4RN
Routledge is an imprint of the Taylor & Francis Group, an informa business
© 2023 Claudine Borg
The right of Claudine Borg to be identified as author of this work has been asserted in
accordance with sections 77 and 78 of the Copyright, Designs and Patents Act 1988.
All rights reserved. No part of this book may be reprinted or reproduced or utilised
in any form or by any electronic, mechanical, or other means, now known or
hereafter invented, including photocopying and recording, or in any information
storage or retrieval system, without permission in writing from the publishers.
Trademark notice: Product or corporate names may be trademarks or registered trademarks,
and are used only for identification and explanation without intent to infringe.
Library of Congress Cataloging-in-Publication Data
A catalog record for this title has been requested
ISBN: 978-0-367-71438-3 (hbk)
ISBN: 978-0-367-71443-7 (pbk)
ISBN: 978-1-003-15090-9 (ebk)
DOI: 10.4324/9781003150909
Typeset in Sabon
by Newgen Publishing UK
v
Contents
List of Figures vii

List of Tables viii
List of Abbreviations ix
Acknowledgements x
Introduction 1
1 Theoretical Framework: Venturing beyond classic

process-oriented research 5
1.1 Cognitive Translation Studies: An overview 5
1.2 The three phases of the translation process 8
1.3 Decisions in the translation process 9
1.4 Alternative translation solutions, postponed decisions
and choices 15
1.5 Revision in CTS 19
1.6 Individual variation in translation processes 29
1.7 Literary translation in CTS 31
1.8 Studies of draft translations 32
1.9 Ethnographic approaches in CTS 34
1.10 Case studies and generalisation in TS and CTS 35
2 Context 38
2.1 The wider context: The “typical” literary translator 38
2.2 The specific context 40
3 Methods and Data 46

3.1 Research design 46
3.2 Data analysis methodology 57
vi
vi Contents
4 A Literary Translation in the Making 69
4.1 Part I. Analysis of the initial interview: How does the
translator envisage his own literary translation process? 69
4.2 Part II. Textual analysis of Draft 1 74
4.3 Part III. Rich description of the translator’s approach
to the translation 81
4.4 Part IV. Written alternative translation solutions 121
4.5 Summing up 131
5 From a Handwritten First Draft to the Published

Version 134
5.1 The translation process 135
5.2 The post-translation process 143
5.3 A process characterised by time gaps 147
5.4 Summing up 149
6 Decision-Making and Underlying Motivations 151

6.1 Alternative translation solutions 151
6.2 Self-revisions 157
6.3 The translator’s self-concept 169
6.4 Matching intentions with actions 172
6.5 Summing up 172
7 Conclusion 174
7.1 Evaluation and implications of the findings 174
7.2 Limitations 180
7.3 Contributions 181
7.4 (Future-)Looking at the individual translator 184
Appendix 1: Evolution of the Translation –

Corpus Sample 186
Appendix 2: Sample of a TPP (Extract) 191
Appendix 3: Analysis of Reasons behind the
Choice of Written ATSs in D2 and
Self-Revisions in D7 195
Appendix 4: Codes used in NVivo 200
Appendix 5: Phases 3 and 4 Sessions 202
References 205
Index 226
vi
Figures
1.1 Decisions and choices in the translation process 10

1.2 My rearrangement of Wilss’s six-stage decision-making
process 12
1.3 The extended translation process 20
4.1 The translator self-revising D1 82
4.2 Extract from TPP08 showing the self-revisions
undertaken in Phase 3 and their chronology 84
4.3 Extract from TPP19 illustrating the self-revisions
undertaken in Phase 4 and their chronology 103
4.4 Example of an ATS suggested by the proofreader
(D5 –hard copy, p3) 106
4.5 The number of other-revisions in D5 compared with
revisions in D6 112
4.6 Distribution of ATSs in D1 123
6.1 The four steps of the decision-making process 156
6.2 Accounting for written ATSs in the decision-making
process 156
6.3 Accounting for the generation of further ATSs in the
decision-making process 156
6.4 Number and type of revisions in the drafts, excluding D2 158
vi
Tables
3.1 Methods used to gather data on the different phases of the

translation process and the data yielded 48
3.2 Summary of the insights provided by each method and
their main benefits and drawbacks 51
3.3 Examples of ATSs in D1 60
4.1 Examples of creative solutions in D1 77
4.2 Examples of OSRs 79
4.3 OSRs in D1 sorted according to linguistic categories 80
4.4 Examples of creative solutions in D1 and their trajectory
in the translation process 92
4.5 Self-revisions in D3 104
4.6 Distribution of ATSs in the first, second and third
parts of D1 124
4.7 Going beyond linguistic categories: Number and
percentage of ATSs in each category 126
5.1 Comparison of the phases of the extended translation
process in TS literature and in this study 135
6.1 The effect of self-revision on the TT 161
xi
Abbreviations
ATS alternative translation solution

CTS Cognitive Translation Studies
D draft
D1, D2 D1 stands for Draft 1, D2 for Draft 2 etc.
ISSI initial semi-structured interview
OSR online self-revision
RI retrospective interview
RS1 retrospective session re: proofreading
RS2 retrospective session re: D7
SL source language
ST source text
TAP think-aloud protocol
TL target language
TPP translation process protocol
TPP01/001 the number after TPP refers to the translation process
protocol number, while the number (or letter or symbol)
after the slash refers to the note reference number (or
letter or symbol) in the TPP
TS Translation Studies
TT target text
x
Acknowledgements
A book is rarely a one-person venture and this one is certainly not so.
Several people have helped me in several ways and without their support
I would not be writing these Acknowledgements.
My heartfelt thanks to Séverine Hubscher-Davidson and Christina
Schäffner for believing in the project and in me from the very start.
I would like to thank members of the TREC research network, par-
ticularly Ricardo Muñoz Martín, Maureen Ehrensberger-Dow, Birgitta
Englund Dimitrova, Hanna Risku –and especially Arnt Jakobsen –for
inspiring this work and for their valuable insights. My thanks also to
Cecilia Alvstad and Olga Castro for their guidance and encouragement.
A special mention goes to Toni Aquilina for agreeing to be part of this
project and for giving so generously of his time. This project would not
have been possible without his collaboration and the data he made avail-
able. I am grateful to Joanne Micallef (Faraxa Publishing) for allowing
me to reproduce parts of the Maltese translation Is-Sur Ibrahim u l-Fjuri
fil-Koran and to Mario Cassar for his expert advice.
Parts of this book build on material that appeared in three articles:
“Decision-making and alternative translation solutions in the literary
translation process: A case study” (2017) in Across Languages and
Cultures, Volume 18: Issue 2 published by Akadémiai Kiadó Zrt https://
akjournals.com/view/journals/084/084-overview.xml; “The effect of self-
revision on the target text: Do self-revisions deliteralise the final transla-
tion? A case study” (2018) in Transletters, Volume 1: Issue 1 published
by the University of Córdoba https://www.uco.es/ucopress/ojs/index.php/
tl/index; and “Written alternative translation solutions in the transla-
tion process. A case study” (2019) in Translation, Cognition & Behavior
Volume 2: Issue 1 published by John Benjamins Publishing Company
https://benjamins.com/catalog/tcb. Thanks to the publishers for giving
me permission to include them here in their entirety or parts thereof.
Thanks go to Louisa Semlyen at Routledge for proposing the book
idea during the EST Congress in Stellenbosch and Elysse Preposi for her
patience and support, to the four anonymous reviewers of the book pro-
posal and, of course, to the series editors Jacob Blakesley and Duncan
Large for supporting this monograph and providing helpful comments
xi
newgenprepdf
Acknowledgements xi
and suggestions for improvement. Thank you, too, for the enjoyable and
enriching experience.
Most importantly, I thank my family, particularly my parents, and
my husband Trevor for their infinite patience, continuous support and
unconditional love. This book is dedicated to you.
xi
1
Introduction
The glamour of the finished translation hides the depths of a text’s

Becoming. For makers we are, but to make, one must remake. But it
is not only that our work is hidden; we hide it. What is lost behind the
scenes, beneath the linear order of the published text, are the out-of-voice
rehearsals where inchoate drafts turn into more drafts, then into sense,
then sense into style, and style into the feeling and ideas evoked in a
reader; we lose the textual threads that could lead us into the multiverse
of possibilities courted and abandoned. Confronted with the devasta-
tion of Miguel de Cervantes’s (or rather, the narrator of Don Quixote’s)
infamous image –that translations are like the reverse side of a Flemish
tapestry –translator Carol Maier once replied (personal communication,
2008) that the reverse is the interesting side anyway. It shows the work.
[Kelly Washbourne 2019:2; emphasis in original]
Until a few years ago, the literary translation process was mystified and the
literary translator out of sight. Literary translations were almost treated
like magic tricks: they happened but you did not see them happening.
But at least in magic, the magician is in full sight, acknowledged, held in
awe and applauded for his or her skill. Literary translation was largely
perceived as a second-rate activity (Bassnett 2013:15) and little or no
value was attributed to translation drafts –not least by translators them-
selves; they were meant to be discarded, not exhibited and studied.
Having access only to the final polished product, readers of translations,
unlike in tapestry, cannot turn around the book to see “the reverse side”.
And as Kelly Washbourne aptly put it, “it is not only that our work is
hidden; we hide it,” thereby hiding the intricate and painstaking process
that literary translation is.
Admittedly, in recent years the visibility of literary translators has
increased significantly, and translators are little by little becoming less shy
about their raw messy processes and more forthcoming in sharing them:
Daniel Hahn’s (2021) superb Translation Diary is a case in point. Over
the years, numerous researchers, too, have provided valuable insights into
the work of translators; however, in-depth studies on particular aspects
of the translation process and on how translations come into existence
DOI: 10.4324/9781003150909-1
2
2 Introduction
are still relatively scarce (Risku et al. 2013:169). We need to better
understand how literary translations are produced, Susan Bassnett has
argued just recently (2019:6). Most of the knowledge we have about the
literary translation process comes either from translators reflecting and
writing about their own practice such as Hahn’s Diary and in the volumes
edited by Susan Bassnett and Peter Bush (2006), Esther Allen and Susan
Bernofsky (2013) and Kelly Washbourne and Ben Van Wyke (2019), or
by researchers working within the sociological or genetic approaches in
Translation Studies (TS).
Traditionally, literary translation has mostly been studied from
a product perspective: the great majority of the analyses involve final
translations, source text–target text comparisons and sometimes draft
versions. But studying translations from a product point of view will not
lay bare the factors that made them the way they are; these are only
revealed through analyses of actual processes and the influences bearing
on them (Risku 2014:334). Cees Koster (2014:153) has argued that lit-
erary translation has lost prominence in TS, that it is no longer central
to the discipline and that the future of research into literary translation
depends on its position within different paradigms. Moreover, Cognitive
Translation Studies (CTS) –the branch in TS that brings the process
into view –has largely been concerned with non-literary translation,
neglecting literary translation. This is another reason why “the depths of
a text’s Becoming” remain hidden. Literary translation has not yet bene-
fitted fully from the wealth of insights that CTS could bring. As Waltraud
Kolb writes:
If we want to gain insight into those processes and learn more about
what actually goes on in the translator’s mind and what goes into
the making of a literary translation we need tools that permit us to
observe those processes closely while they occur.
[Kolb 2011:260]
This book addresses the coming into being of a literary translation from
a situated process perspective, bringing together CTS and literary trans-
lation. Taking the Maltese translation of Éric-Emmanuel Schmitt’s philo-
sophical novella Monsieur Ibrahim et les Fleurs du Coran as a case study,
it examines in- depth its making, revealing the behind- the-
scenes, its
reverse side. Its aims are threefold: i) to analyse in detail how a translator
goes about the task of translating a literary text, ii) to investigate the tra-
jectory of a literary translation from first draft through to publication and
iii) to demonstrate the value of in-depth process studies for both CTS and
literary translation, thus encouraging this line of research. The translator,
Toni Aquilina, is closely studied while he revises his draft translation until
it is deemed finalised and goes to print. This book places the translator
in the limelight, centring upon his work and the process leading to it. In
Washbourne’s words: “It shows the work.”
3
Introduction 3
Situated within a CTS framework, the book combines methods from
this research paradigm (think-aloud, draft versions, and video recordings)
with ethnographic techniques (observations and interviews) to provide
thorough insights into the practice of an experienced literary trans-
lator working in situ. It provides a detailed account of the translator’s
actions and decisions, alternative solutions considered and discarded,
underlying motivations, his work practices, his approach to the task and
his interactions with other participants as the translation evolves from
its first handwritten draft to the published version available on book-
shelves. In line with recent work in CTS, this volume takes into account
both cognitive and social processes. It is practice driven, focusing on the
situated and embedded nature of translation. Veteran translation scholar
Ian Mason (2014:38) highlights the importance of studying translatorial
decisions and the contextual factors impinging on them in real transla-
tion practice. By investigating a translator self-revising an entire literary
translation from start to finish, this study aims to shed some further light
on the decisions and choices arrived at by the translator in the process
of translation and which ultimately shape the final work. In so doing, it
aspires to increase our understanding of translatorial decisions and of the
various ways in which these can shape the target text.
While drawing on pioneering work in the field, the book ventures
beyond classic process-oriented research in several ways by i) focusing
on literary translation, ii) zooming in on one translator working on a
book-length piece of work, iii) taking into account both the translation
process and the post-translation process, including the publication phase
and iv) using a much bigger quantity of material, including all the draft
versions and 90 hours of fieldwork. The book therefore extends and
enhances CTS, relating its findings with previous studies’ findings and
questioning several key assumptions in the field such as the division of the
translation process into three phases and the deliteralisation hypothesis.
It also offers an alternative way of studying literary translation. The great
majority of studies involving literary translation investigate translations
and translators retrospectively, after the completion of the task and not
while the translator is at work. This study applies an innovative approach
in an attempt to provide fresh and innovative insights into the literary
translation process.
Moreover, the volume builds on calls for research focusing on indi-
vidual situated translators and their creative processes (Halverson
2014:117). It is also in line with both Andrew Chesterman’s (2009:13)
vision for Translator Studies, which would study a translator’s agency
from sociological, cultural and cognitive perspectives, and Kaindl et al.’s
recent publication Literary Translator Studies (2021), as it examines
how the literary translator makes decisions and moulds and influences
the translation within his particular context. It underlines the potential
and value of in-depth process studies for the discipline and illustrates the
sheer amount of insight one such study brings to TS. By describing in
4
4 Introduction
detail the study’s methodology, the book also provides a method for and
an example of how similar studies could be carried out.
Research questions and structure of the book

This study asks and attempts to answer the following questions:
i) How does a literary translation come into being?

ii) What happens in the translation process after the first draft of a
translation is produced?
iii) How does the translator in question approach the target text?
iv) How does the translator under study arrive at his decisions?
v) Which decisions and choices does the translator make during the
post-drafting phase? What are the underlying motivations?
vi) How does the revision process shape the translation?
The structure of the book is as follows. Chapter 1 provides a theoretical

framework for the study and discusses relevant issues and previous key
studies in CTS. In view of the importance of contextualisation for case
studies and in particular for the interpretation of the results, Chapter 2
provides a context for the study. For replicability’s sake, the methodology
is explained in detail in Chapter 3, which provides a thorough account
of the research design, the data collection methods, the data yielded and
the data analysis methodology. Chapter 4 analyses the data and presents
the findings. Chapters 5 and 6 discuss and expand on some of the salient
findings presented in Chapter 4. Drawing mostly on CTS but also on
literary translation literature, these two chapters triangulate the results,
weigh the evidence and create links with previous TS findings. Chapter 5
focuses on the phases the translation went through as it evolved from first
draft to published product and on the translator’s approach to the task,
whereas Chapter 6 centres on translatorial decision-making, particularly
on alternative translation solutions and self-revisions, giving attention to
the reasons that motivate decision-making. The Conclusion evaluates the
implications of the findings for CTS and literary translation, as well as
the contributions and limitations of the research. It also highlights the
benefits of in-depth process studies for TS and recommends avenues for
further research.
5
1 Theoretical Framework
Venturing beyond classic
process-oriented research
This chapter outlines the concepts, definitions and models underpinning

the theoretical framework of this study. It surveys previous research on
relevant topics, discusses key studies for the current project and identi-
fies gaps in the literature. First, a brief overview of Cognitive Translation
Studies (CTS) is presented in order to locate the present study in this
branch of Translation Studies (TS).
1.1 Cognitive Translation Studies: An overview

Cognitive Translation Studies is one of the three areas of research within
the descriptive branch of Holmes’ (1988:72– 73) map of Translation
Studies that was graphically presented by Toury (1995:10). It should
be highlighted that this research area is also referred to by other labels
such as process-oriented research, translation psychology (Holmes 1988;
Jääskeläinen 2012) and translation process research (TPR). Until recently,
the term TPR was widely used, but it has now become associated with
a line of research based on technology –using mainly the Translog soft-
ware –and drawing on the information process paradigm. Following
Alves and Jakobsen (2021b:546), this study applies the term CTS and
understands it in its broad sense, i.e. to refer to process-oriented transla-
tion studies. Drawing heavily on psychology and cognitive science, CTS
endeavours to understand the mental processes at play when a trans-
lator works on a translation. Since cognitive processes are influenced by
internal and external factors, affective and ergonomic processes also form
part of CTS’ investigations. What happens in the translator’s mind during
the translation process is known as translation acts, while translation
events incorporate the sociological processes surrounding the act or its
situational context (Toury 1995; Chesterman 2013).
Muñoz (2016b:155–156) has however challenged this dual act/events
perspective, arguing that acts cannot be separated from events since they
impinge on and influence each other. This led him to propose the term cog-
nitive translatology (Muñoz 2010), which sought to recontextualise TPR
and broaden its scope (Muñoz 2016a). While classic TPR focuses on acts,
draws on information processing theory and views cognition as restricted
DOI: 10.4324/9781003150909-2
6
to the brain, cognitive translatology advocates translation as a situated
action that takes into consideration the embodied, enacted, embedded,
extended and affective (4EA) aspects of cognition. CTS embraces these
two paradigms of process research (see Alves & Jakobsen 2021a). As will
be detailed later, the present study falls within the cognitive translatology
paradigm of CTS.
Highly empirical, CTS investigates translators’ actual actions as
opposed to assumptions about what they do while translating. Research
in this area started in the mid-1980s and interest in CTS has grown over
the years. Gaining access to the mind is no easy task and, since mental
processes cannot be observed directly, process researchers attempt to take
a glimpse into the workings of the translator’s black box by employing a
number of research methods and tools. The first pioneering studies (e.g.
Gerloff 1987; Krings 1987; Séguinot 1989; Lörscher 1991) elicited data
through think-aloud, a verbal report procedure imported into TS from
cognitive psychology. Verbal reports are widely used in process research
to tap the subjects’ cognitive processes through their own verbalisations
(e.g. Krings 2001; Englund Dimitrova 2005; Göpferich 2009; Dorer
2020). They mainly comprise concurrent and retrospective verbalisa-
tion as well as dialogue think-aloud, and their theoretical and methodo-
logical foundation is largely based on Ericsson and Simon’s seminal work
(1984/1993, 1987). Think-aloud was very popular in the eighties and
nineties, but nowadays its use has decreased drastically mainly because
of concerns related to its validity and the advent of other methods (Sun
et al. 2021:133). Nevertheless, it is still perhaps the best method to elicit
information about translators’ minds (Jakobsen & Alves 2021:4).
In the 1990s, researchers started exploiting technology to scrutinise
the translation process. Jakobsen and Schou (1999) developed Translog,
a key-logging software that records all the actions a translator makes
on the keyboard and the mouse, such as additions, deletions, revisions,
pauses and mouse clicks. These actions are time stamped and saved in a
log file and the researcher can replay the process on screen. Keystroke
logging programs have been used to study various aspects of the transla-
tion process such as segmentation (e.g. Jakobsen 2003; Englund Dimitrova
2005) and pauses (e.g. Immonen 2006) and they are often combined with
an eye-tracker to trace a translator’s eye movements and fixations on a
screen. An eye-tracker provides data on, for instance, reading, compre-
hension and monitoring processes in translation (Jakobsen 2014:75) and
complements the data supplied by keystroke logging. Moreover, screen
recording software registers data about online research activities carried
out by translators and about other software they use; it is often used in
conjunction with other methods to monitor the participants’ actions on
the computer (e.g. PACTE 2009, 2011).
Whereas the three technological methods discussed above pro-
vide quantitative data, methods borrowed from ethnography such as
interviews, observations and translation diaries yield qualitative data.
7
Theoretical Framework 7
Here, the participants’ actions and verbal reports are often captured on
video or audio. Other data collection procedures employed in process-
oriented research include questionnaires and psychometric instruments
(e.g. Hubscher-Davidson 2018b). Researchers are also increasingly
experimenting with methods from the neurosciences such as electroen-
cephalography (EEG) and functional magnetic resonance imaging (fMRI)
(e.g. García & Muñoz 2021). These innovative methods enable process
researchers to measure, for example, brain activity and cognitive effort.
Hence, the methods used in CTS are varied and are often combined
to mitigate the limitations of individual methods and to provide com-
plementary sets of data and results which are then triangulated (Alves
2003:vii). The methods chosen depend on the aims and objectives of the
study. There are therefore different ways of doing process research, such
as i) the psychological approach, which looks into psychological phe-
nomena of translation such as emotion, personality and self-confidence;
ii) the technological approach, often undertaken in laboratory condi
tions, which applies sophisticated technology and focuses mostly on
translatorial micro-behaviour; and iii) the ethnographic approach, where
research is carried out in naturalistic environments. Yet one approach
does not necessarily eliminate others; in fact, researchers are increasingly
combining different approaches (see, e.g., Ehrensberger-Dow 2014).
As happened with the research methods, over the years the range of
topics investigated in CTS, too, has become more and more varied. They
encompass, for instance, cognitive effort (e.g. O’Brien 2006; Dragsted
2012; Sun et al. 2021); creativity (e.g. Kuβmaul 2000; Bayer-Hohenwarter
2010); decision-making, problem-solving and strategies (e.g. Krings 1986,
2001; Jääskeläinen 2009; Prassl 2010; Ferreira et al. 2018); emotions
(e.g. Lehr 2014; Rojo & Ramos 2016; Hubscher-Davidson 2018b); ergo-
nomics (e.g. Ehrensberger-Dow & O’Brien 2015); expertise (e.g. Englund
Dimitrova 2005); intuition (e.g. Hubscher-Davidson 2013); metaphors
(e.g. Tirkkonen-Condit 2002; Schäffner & Shuttlewood 2013; Massey &
Ehrensberger-Dow 2017); post-editing (e.g. Vieira 2017); revision (see
Section 1.5); time-pressure (e.g. Alves & Liparini Campos 2009); transla-
tion competence and its acquisition (e.g. PACTE 2003, 2019; Göpferich
2013); and translator performance (e.g. Jakobsen 2005).
CTS research started in laboratory settings and in classrooms, and it
focused on how translators process texts or aspects thereof, involving
mainly the examination of decontextualised translational microlevel
behaviour (e.g. Hansen 1999, 2002). Although a great deal of valu-
able research is still happening in laboratory conditions (e.g. Carl et al.
2016), process researchers have ventured out of the laboratory in recent
years and into translators’ workplaces, a move that has reinstated the
human, social and cultural dimensions of cognition and opened many
research avenues (Muñoz 2014a:67). In fact, various process researchers
are now investigating translation as a situated activity. They examine
how translation materialises at the workplace, in other words, examining
8
real working practices and real texts, but they are also looking beyond
this at how translators interact with their environment: with each other,
with other colleagues, with their tools etc. (e.g. Kuznik & Verd 2010;
O’Brien 2012; Ehrensberger-Dow & Massey 2014; Risku 2014; Risku
et al. 2017; Ehrensberger-Dow & Englund Dimitrova 2018). As Muñoz
(2015:11) puts it, innovative research “trends are not a change of
course, but rather a widening and deepening of the approaches from the
eighties”. As mentioned above, he gathers these new trends under cogni-
tive translatology. A lot has been done and great inroads have been made
in process-oriented translation research, nevertheless “much still remains
insufficiently explored” (Jakobsen 2014:78).
The present process study combines methods from cognitive psych-
ology (think-aloud and retrospection) with ethnographic methods (obser-
vation and interviews) to investigate how a literary translation comes into
being. In line with studies of translation as a situated activity, it examines
the translator at his place of work when preparing a translation for pub-
lication. Although the current research draws on the work of numerous
process scholars (e.g. Jakobsen 2002; Englund Dimitrova 2005; Göpferich
2009) and shares similarities with, for instance, Risku’s (2014) ethno-
graphic study, Ehrensberger-Dow’s (2014) workplace study and Kolb’s
(2011, 2013, 2017) work on decision-making in literary translation, it
does not neatly align with one particular type of process approach. The
present research expands the boundaries of classic translation process
research by studying in-depth the process of one translator working on
a book-length literary text until it is published (see Sections 1.6, 7.1 and
7.3). It touches upon the three levels of the translation process identified
by Muñoz (2010) –level one encompasses mental states and operations;
level two, sub- tasks and observable operations for example reading,
consulting resources and self-revising; and level three, the situated nature
of the translation process –since it studies i) translatorial decision-making
pertaining to level one; ii) sub-tasks such as self-revision relating to level
two; and iii) the evolution of a literary translation from first draft to
publication. Therefore, my work intersects with various CTS studies and
Muñoz’s (2010) three levels of the translation process.
Next, the different phases of the translation process are defined because
they feature prominently in this book, and understanding the makeup of
the translation process is important for this study.
1.2 The three phases of the translation process

Generally, researchers (e.g. Mossop 2000; Jakobsen 2002; Englund
Dimitrova 2005; Carl et al. 2010; Huang 2016; Sun et al. 2021) agree that
the translation process is divided into three distinct phases, although they
employ different terminology for these phases. For example, Jakobsen
(2002:191) uses orientation, drafting and revision; Englund Dimitrova
9
(2005:86) uses pre- writing, writing and post- writing; and Mossop
(2000:40)1 uses pre-drafting, drafting and post-drafting.
Despite applying different terminology, these three scholars define
the three phases very similarly. In this study, I adopt Mossop’s termin-
ology because my data indicate that the first full target text (TT) version
produced by the translator in the second phase is a draft version, not a
finished product, and some work still needs to be done in the third phase.
Besides, the translator refers to the first version of the TT as the first draft.
Here, the three phases are understood as follows: in the first phase, the
translator familiarises himself with the source text (ST); in the middle
phase, a full version of the translation is produced; and the third phase
begins after the first draft is completed.
Phase 1 serves as a familiarisation phase where translators read and
interpret the ST before composing the TT. Comprehension is a main fea-
ture of this phase (Jakobsen 2002:192) as is planning (Englund Dimitrova
2005:24). Yet previous studies (e.g. Jakobsen 2002; Carl et al. 2011; Sun
et al. 2021) have shown that the length and activities performed during
this phase differ considerably among translators and this variation seems
unrelated to translation experience (Englund Dimitrova 2005:22). Some
translators jot down words/phrases and/or carry out research, some read
the whole ST and others browse it quickly, whereas certain translators
start translating immediately, skipping this phase altogether.
In Phase 2, the translation is drafted alongside other actions: translators
read the ST (again) and engage with it, resort to external resources such
as dictionaries and the Internet, and evaluate the emerging text as well as
self-revise it. Most studies have found the second phase to be the longest
(e.g. Jakobsen 2002; Englund Dimitrova 2005; Dragsted & Carl 2013;
Shih 2013), but individual differences and exceptions were reported here
as well in terms of duration, approach and activities performed.
Since this study centres on post-drafting, the third phase is discussed in
more detail in Section 1.5.3 after important related concepts are tackled.
1.3 Decisions in the translation process

Decision experts Judith Orasanu and Terry Connolly (1995:6) stated that
“in everyday situations, decisions are embedded in larger tasks that the
decision maker is trying to accomplish”. In translation, the larger task,
i.e. the production of the TT, involves extensive and continuous decision-
making activities (Wilss 1998; PACTE 2011) interlaced with the trans-
lation process. Decisions form such an integral part of the production
process that various scholars (e.g. Levý 1967/2000; Hatim & Munday
2004; Munday 2008a) define the translation process as a decision-
making process. The translator, in the course of translation, constantly
makes decisions and this series of decisions ultimately forms the TT. This
study adopts this perspective and views translations as the result of the
01
translator’s decision-making process. It is therefore important to define
“decision” and other related terms.
1.3.1 Definitions
In a general context, Baron (2008:6) defines “decision” as “a choice of
action”, which people make with the aim of realising goals by basing
them on their beliefs on how best to attain them. If we apply this to trans-
lation, the translator selects a course of action, for instance which transla-
tion strategy to opt for and which to discard. Translational decisions are
taken to accomplish the translator’s goal –that is, to produce a specific
type of TT –and are guided by the translator’s beliefs and a myriad of
other factors such as the ST, the target culture and the target readers.
The term “choice” is a common denominator in many definitions of
decisions both in and outside of TS (e.g. Eilon 1969; Krings 1986). Wilss
(1994:132), too, defines decisions in terms of choice: a translator needs
to make a decision when confronted with a choice.2 Similarly, Holmes
(1988:4) argues that the translator is continuously challenged by choices,
and translation mainly involves making choices (1988:60). Hatim and
Munday (2004:52) point out that the translator makes choices at each
phase of the translation process. Choice, defined as deciding among
alternatives (Jennings & Wattam 1998:26), is thus a key feature in
decision-making and consequently in the translation process. Figure 1.1
depicts my illustration of these concepts.
Psychologists (e.g. Jennings & Wattam 1998) and TS scholars alike (e.g.
Hatim & Munday 2004; PACTE 2008; Jääskeläinen 2012) acknowledge
that decision-making is highly complex. Wilss (1994:133) maintained
that translational decision- making is even more complex because of
the presence of the ST, which is an added factor in the translatorial
• involves extensive and continuous decision-

Translation making activities
process
• involves making choices

Decisions
• deciding between alternatives

Choices
Figure 1.1 Decisions and choices in the translation process.

1
decision-making process when compared to monolingual writing, and de
Groot (1997:30) argued that for a long time psychologists side-lined this
topic because they considered it too complex to tackle. It is, definitely, an
intricate phenomenon.
Translational decisions are often grouped into two broad categories:
macrolevel and microlevel (e.g. Holmes 1988; Jääskeläinen 1996;
Tirkkonen-Condit 2005). The former are decisions at the global level
of the text and encompass strategies such as the decision to domesticate
or foreignise names (Venuti 1995, 1998), and the latter pertain to the
lower level of words or group of words (Holmes 1988:54–55) restricted,
say to, a phrase, sentence or paragraph. Wilss (1994:134) postulates that
strategies are needed for decisions at the macrolevel to avoid inconsist-
encies, and that microlevel decisions are time-consuming, especially in
literary texts. To this, one may add that macrolevel decisions could be
equally time-consuming with wide-ranging consequences: for example,
if halfway through the translation process a translator decides to change
a macrostrategy, this may necessitate considerable reworking at the
microlevel. In fact, Holmes (1988:55) argues that macrolevel decisions
govern and constrain microlevel decisions. In a similar vein, Tirkkonen-
Condit (2005:406) suggests that experts, unlike novices, make certain
global decisions towards the beginning of the process, and these decisions
then guide local decisions such as the choice between competing trans-
lation solutions (see Section 1.4). The idea that prior decisions impact
subsequent decisions seems to be significant in decision-making and is
also seen in Levý’s work.
When, back in 1967, Levý (1967/2000:148) envisaged the translation
process as a decision-making process, he compared it to a game of chess,
where one move, or decision in the case of translation, determines the
next. However, Levý’s theory seems to imply that translation decisions
are linear, one following the other, with previous decisions influencing
later decisions, but not the other way around. Whereas in a game a player
cannot amend previous moves, a translator can and often does. In fact,
there is ample empirical evidence (e.g. Englund Dimitrova 2005; Malkiel
2009) that translators revise their decisions during different phases of
the process. Thus, although it is true that former decisions determine
and influence later decisions, in translation this could also work in the
opposite direction, since, at times, prior decisions are amended in view
of later decisions and the emerging text, as will be shown by the pre-
sent study.
1.3.2 The structure of decisions

According to Jennings and Wattam (1998:19), decisions consist of three
stages:3 identification of the problem, development of one or various
solutions, and selection of a solution. This is somewhat similar to how
Krings (2001:466) conceives the basic makeup of translatorial decisions:
21
the translator evaluates a number of alternatives and then chooses between
them. Therefore, Krings omits the problem identification stage, as does
Pym (2003:489), for whom the translation process consists of producing
solutions and choosing among them. This brings to light two aspects: i) the
fundamental role of the production of various alternative translation
solutions (ATSs) and the ensuing choice in the translation process and ii) the
striking similarity between the decision-making process and the translation
process, further highlighting translation as a decision-making process.
Whereas Krings streamlines the decision- making process into two
stages, Wilss (1996:188), drawing on Corbin (1980), proposes a process
composed of six consecutive stages: i) identification of the problem, ii) its
description, iii) gathering of information, iv) thinking of a way forward,
v) choosing a solution and vi) evaluation of the outcome. Wilss designates
the sixth and last stage “post-choice behaviour (evaluation of translation
results)”, which implies that, for him, evaluation takes place after a solu-
tion is selected. Evaluation, however, also takes place earlier on in the
decision process, for instance when the translator evaluates the various
solutions generated. In fact, Wilss (1996:188) indicates that the bound-
aries between the stages may be blurred or may overlap. Similarly, Krings
(2001:464) points out that it is difficult to distinguish between evaluation
and decision-making processes, and he questions whether this distinction
is valid. Figure 1.2 presents my reworking of Wilss’s model.
In the present work, evaluation is considered an integral part of the
decision-making process that may take place at any stage of the decision
process and, by extension, at any phase of the translation process. It is
seen as a fluid and flexible process. Its prominence increases in the post-
drafting phase, when the translator reviews the TT, evaluates the draft
solutions and decides whether to validate draft solutions and move on
or else revise. The translational decision-making process is here viewed
as consisting of four key stages: identification, development, selection
and evaluation. This is supported by Shih’s (2015:86) study on decision-
making in post- drafting, which found that translators generally skip
the description stage. Since the present study considers translation as a
Identification of the problem
Description of the problem

Evaluation
Gathering of information
Thinking of a way forward
Choosing a solution
Figure 1.2 My rearrangement of Wilss’s six-stage decision-making process.

31
decision-making process, identification occurs with each segment to be
translated but it is evident when a problem is detected. The development
stage involves the generation of one or various ATSs. If the translator
comes up with more than one solution, then a choice needs to be made.
I argue that the sequence of these four stages is not static. Moreover,
some stages may be omitted according to the level of consciousness. For
example, Prassl (2010:61) suggests that, in routine decisions,4 only one
option is produced, thus the selection stage is skipped.
In the post-drafting phase, the starting point of the decision-making
process is most often the evaluation of the solutions present in the draft.
If the solution is evaluated as correct, then the translator moves on
to the next segment; if not, a decision to alter the TT is made and the
decision-making process is restarted: the problem is identified, one or
more solutions are developed, then a choice is made; possibly that choice
is evaluated again.
Uncertainty is an important element in decision- making and
translatorial decisions are imbued with uncertainty (Tirkkonen-Condit
2000:141), defined by Angelone (2010:18) as “a cognitive state of
indecision”. Managing uncertainty is a chief element in translators’
decision-making (Fraser 2000:115). Tirkkonen-Condit (2000:140)
established that experienced translators tolerate uncertainty and tackle
it in various ways, for instance by generating an abundance of solutions,
trying them out or postponing decisions. Various researchers (e.g. Fraser
1996; Angelone 2010; Angelone & Shreve 2011), in fact, found that
experienced translators tend to have a higher tolerance for uncertainty
than novices and they manage it better (Angelone 2010; Angelone &
Shreve 2011), and Hubscher-Davidson (2018a) suggests that tolerating
uncertainty can help in resolving complex decisions. Hence, being able to
deal with uncertainty is associated with proficiency in translation and is
seen as beneficial.
Decisions are based on information (Jennings & Wattam 1998:10),
and TS research has shown that translators derive information from two
distinct sources: internal and external resources (e.g. Alves & Liparini
Campos 2009; PACTE 2009). Internal resources consist of the translator’s
personal knowledge and employ automatic and non-automatic cogni-
tive resources (PACTE 2009:215), whereas external resources entail
the recourse to documentation sources, such as dictionaries and the
Internet. While translating, these two resources are often combined to
different degrees for solutions to be achieved. This leads us to another
important and frequent distinction, i.e. that of conscious and uncon-
scious decision-making.
1.3.3 Conscious and unconscious decisions

TS scholars frequently distinguish between conscious and unconscious
decisions and choices (e.g. Boase-Beier 2006; Munday 2008a; Prassl
41
2010; Hubscher-Davidson 2013). In psychology, such a distinction is
proposed by dual-process models (e.g. Evans 2007; Weber & Johnson
2009) that suggest decision-making is either deliberate or intuitive, or a
mixture of the two. Deliberate processes apply rules and computations
consciously, in a controlled way, while intuitive ones “operate (at least
partially) automatically and without conscious control” (Glöckner &
Witteman 2010:4). In TS, Lörscher (1991:203– 204) argues that the
translation process is made up of problematic and non-problematic parts.
Non-problematic parts are either unconscious or conscious, in the latter
case ensuing from quick decision-making. As Englund Dimitrova suc-
cinctly explains:
In the translation process of any individual, there are segments which

are translated apparently automatically, without any problems, and
other segments where translation is slow, full of many variants and
deliberations, which necessitates a problem-solving approach and the
application of strategies.
[Englund Dimitrova 2005:26]
Translators thus make use of both deliberate and intuitive decision-

making. Accordingly, not all translational decisions are rational,
structured and measured (Wilss 1994:132; Jääskeläinen 2009:378–379),
and unconscious, intuitive processes play a significant role in the transla-
tion process (Hubscher-Davidson 2013:214).
However, research (see Hubscher-Davidson 2013) has shown that auto-
matic, intuitive processing is not as irrational or haphazard as it was once
thought to be. Intuition is not static but develops through learning and
experience (Glöckner & Witteman 2010:5–6); it is cumulative with older
decisions informing new ones. A novel problem can become programmed
over time if it is encountered repetitively and a procedure to handle the
problem is established with time (Jennings & Wattam 1998:3–4). This
is in line with empirical findings in TS research which have shown that,
with experience, translators’ behaviour becomes more automatised and
intuitive (e.g. Jääskeläinen & Tirkkonen-Condit 1991; Göpferich 2010;
Prassl 2010). Therefore, practice and experience impact translatorial
decisions in obvious and less obvious ways. This has implications for the
present study, which looks at the translation process of an experienced
literary translator, as some decision-making processes could have become
internalised through years of practice and experience. On the other hand,
Wilss (1994:132) argues that literary translation is a non-routine activity
(compared to, for example, translating contracts) during which the trans-
lator is constantly engaged in active decision- making. Consequently,
the nature of the text involved in this study might demand more con-
scious processing and is less likely to rely on routinised behaviours. Being
aware of the different types of processes underlying decision- making
is important for this study as it enables a deeper understanding of the
51
process and allows for a fuller explanation of specific instances of the
translator’s decisions. However, its main aim is not to categorise decisions
as deliberate or automatic but to look in-depth at the decision-making
process of one individual translator working on a literary text, the factors
influencing his choices and their impact on the final translation.
Translation research has established that many factors are at play
during decision- making, such as the influence of the ST (e.g. Wilss
1994:133), the brief (e.g. Fraser 1996:89) and target-culture norms (e.g.
Toury 1995:61), as well as the specific sociocultural and ideological
setting within which the translator works (e.g. Munday 2008a:175;
Statford 2020:120). However, it is a fact that no two translations are
the same, even if all the above-mentioned factors are constant. This is
because the individual element is significant in translation (see Section
1.6). “There is obviously room for individual decision-making”, Munday
argues (2008a:48), and Holmes (1988:54) maintains that the translator
chooses on the basis of his personal knowledge, experience, tastes and
preferences. Wilss (1994:139) highlights that translators convey speci-
ficity to decision-making: translatorial decisions are influenced by the
translator’s distinct characteristics and, in Meschonnic’s view, (1999:89)
most translators have pre-conceived ideas that bear on the choices they
make. Recent research has also showed that the translator’s emotions
and well- being impact choices too (e.g. Lehr 2014; Rojo & Ramos
2014; Rojo 2017; Hubscher-Davidson 2018b). The translator, though
often camouflaged and hidden (Venuti 1995), shapes the TT, and it can
be safely assumed that s/he is one of the main reasons, if not the main
one, why every translation is different. As this book will demonstrate,
studying in-depth the decision-making process of one translator throws
interesting light on translatorial decisions and the different factors at play
during decision-making. In translation, like in other human actions, the
roles played by the individual and the social environment change with
every new situation (Paloposki 2009:190). Hence, this study also furthers
our understanding of why a translation –or rather the literary translation
involved in this study –is the way it is. Further glimpses into translatorial
decisions will also be gained by investigating ATSs, which, as we have
seen, are part of the decision-making process.
1.4 Alternative translation solutions, postponed

decisions and choices
It has been established that producing ATSs and choosing between them is
an intrinsic part of the decision-making process as well as the translation
process at large. Amongst other authors (e.g. Lörscher 1992; Pym 2003),
Levý (1967/2000:148) explains that the translator first generates possible
solutions and then chooses between these solutions. Empirical evidence
leads Englund Dimitrova and Tiselius (2009:121) to conclude that con-
templating different target language options seems to be an important
61
part of the translation process. The data in Krings (2001:423) revealed
that, for each hundred words in the final translation under study, an add-
itional sixty-two words of variants were produced and excluded in the
process. This underlines the significant role of ATSs in the translation
process.
Accordingly, ATSs,5 here defined as the consideration of different TT
solutions for the same ST segment, feature frequently in CTS literature.
Tirkkonen-Condit found that the production of tentative solutions is a
shared behaviour by professional translators (2000:141). Pym (1992:281)
associated the capability to produce various translation solutions with
translator competence. Englund Dimitrova (2005:26) linked transla-
tion variants with problematic segments in the translation process. For
Göpferich (2009:33), the amount of ATSs produced reflects process cre-
ativity, while for Jakobsen (2019:68–69) they signal possible interpret-
ations of the ST evidencing the indefiniteness of meaning. ATSs have
clearly generated substantial debate and have been variously interpreted.
Krings views ATSs as characteristic of the translation process
(2001:417). Observing that translators tend to consider various possible
solutions during the translation process, he classified translators into two
categories: those who consider few variants and those who consider many.
Translators also differ in the way they process variants: some translators
produce variants mentally, writing down only the chosen solution; others
write down a solution and then revise it with another (Krings 2001:530).
Krings thus considers self- revisions as ATSs. My work distinguishes
between self-revisions, verbal ATSs and written ATSs:
• Self-
revisions consist of one solution replacing another and are
discussed in Section 1.5.2;
• ATSs refer to both verbal and written ATSs;
• Verbal ATSs denote possible solutions generated mentally and
externalised verbally that may or may not materialise in self-revisions;
• Written ATSs are generated mentally and externalised in writing: the
translator writes down various solutions and postpones the choice to
a later phase. Written ATSs are several possible solutions simultan-
eously present in the draft.
Discussions dealing with the presence of several ATSs in the written

draft are quite rare in process literature and, although several process
researchers discuss the postponement of decisions, they only do so rather
briefly. For example, Wilss (1996:188) questions why certain decisions
are delayed, but he does not account for this phenomenon in his model.
Tirkkonen-Condit (2000:141) identified postponed decisions and tenta-
tive solutions in her subjects’ think-aloud protocols (TAPs) and linked
them with the ability to deal with uncertainty. In Englund Dimitrova’s
study, a number of participants postponed decisions to the post-drafting
phase, and the first draft of one of her professional subjects contained
71
various instances of postponed decisions and provisional solutions
(2005:108–109). Like Tirkkonen-Condit, she interpreted this as toler-
ating uncertainty and thinking over a decision. Interestingly, she also sees
the deferral of certain decisions to the post-drafting phase as a global
strategy employed by the translator in order to take ultimate decisions
once a full TT is available (Englund Dimitrova 2005:109). In a similar
vein, Göpferich (2010:17–19) proposed that translators defer decisions
either for strategic reasons, for instance to get a better grip of the TT and
thus be in a position to make more informed decisions, or else because
of an absence of strategies. Her 2010 study reveals that students post-
pone decisions because they lack strategic behaviour.6 Kolb (2013:218)
also reports that most of her experienced literary translators postpone
decisions in the first draft.
Deferred decisions seem to be a common phenomenon in the trans-
lation process of many translators. Although many authors report this
occurrence and provide interesting interpretations, most of them only
deal fleetingly with this intriguing aspect. In fact, Englund Dimitrova
suggests further research on postponed decisions (2005:148). The pre-
sent study focuses on this phenomenon in order to shed further light on
it –for example, why an experienced translator postpones decisions and
the type and nature of postponed decisions.
In the data reported in the literature, deferred decisions feature in
different forms, such as subjects typing a series of repeated symbols
(e.g. Kolb 2013:218), writing a garbled word (e.g. Englund Dimitrova
2005:109), leaving a blank space (e.g. Prassl 2010:70) and writing various
TT options for the same ST segment (e.g. Krings 2001; Munday 2012).
Moreover, researchers generally report that ATSs are generated verbally
(e.g. Tirkkonen-Condit 2000; Dragsted 2012), with the translator writing
down only one solution in the TT. Krings (2001:417) observed that many
alternative solutions are generated mentally, and if written down they
feature as revisions, with one solution replacing the other rather than
being simultaneously present in the draft as is the case in Aquilina’s D1.7
Accordingly, this intriguing phenomenon merits further investigation. Is
this a strategic behaviour on the part of the translator? How does he
deal with these written ATSs in the post-drafting phase? Taking a closer
look at the different ATSs is also interesting. What are these? Are they
produced for particular aspects of the TT? Can we categorise them in any
way? The current study explores these aspects.
1.4.1 Choosing between ATSs

Once a number of possible solutions are generated, then a choice is
necessary. As already discussed, choice is one of the stages of the deci-
sion process and it occurs when more than one solution is produced.
Yet choosing among the various translation solutions generated could be
problematic: Englund Dimitrova and Tiselius (2009:121) found that 28%
81
of the problems experienced by their subjects were related to the choice
between variants. Krings (2001:464) affirmed that choosing between
ATSs is challenging in situations where the variants are evaluated as being
all equally good or all equally bad. In other words, selection is difficult
when the choice is not clear cut. Pym (2015:76) suggested that, in order
to avoid risk, translators select simpler and more explicit alternatives.
Wilss (1994:141) not only advocated the need for understanding how
translators make decisions and choices but also emphasised the import-
ance of studying pre-choice behaviour (1994:140, 1998:59) –that is, the
factors constraining or motivating choices during the translation process.
More recently, Bangalore et al. (2016:212) reiterated that details of how
translators choose among translation alternatives and the factors guiding
their choice are still mostly unknown. How does the translator involved
in this study choose among the various solutions he produces? What can
we learn about choice behaviour? What are the factors influencing his
choices?
1.4.2 Written ATSs: Empirical evidence from the literature

One of the very few studies in the literature that deal with numerous
written ATSs present in the same draft version is Munday’s case study
(2012:118–121)8 concerning George Davis’ draft English translation of
Vargas Llosa’s short essay “El Paraíso de los libros” (1991). Munday
(2012:118) points out that this data source is particularly significant as
translation alternatives feature very clearly in the draft. Unsure about
which variant to choose, Davis includes 53 sets of ATSs in the draft
and forwards his queries to the ST author. This corroborates the claim
that alternative TT options reflect uncertainty in the translation process
(Tirkkonen-Condit 2000:141; Englund Dimitrova 2005:109). In this par-
ticular case, the translator seeks guidance from the ST author in order to
resolve his doubts.
Drawing on appraisal theory, Munday analyses and categorises the
53 sets of written ATSs contained in the draft TT. He finds that 20 of the
53 doubts concern the choice between synonyms. In 18 other queries,
the translator checks that he has understood and rendered the ST cor-
rectly. The data available to Munday is certainly very interesting, yet it
has a missing link: we do not know what happened to the variants in the
process between Davis’ draft and the final published translation –all we
have are the final choices. As Munday (2012:118) himself concedes, we
do not know how these doubts were solved, although the final solutions
are available in the published text. But who made the final choice? For
which reasons?
Munday’s novel and insightful study offers a very good basis for
the current investigation into ATSs. The first draft produced by my
study’s participant also exhibits a good number of variants. However,
this study’s data are more comprehensive as they include all the draft
91
versions leading to the final TT, some of which are accompanied by
think-aloud and observation data. Interestingly, Munday (2012:160)
concludes his monograph by suggesting that it would be “interesting to
pursue this research with other experimental methods, such as think-
aloud protocols”. Verbal reports could provide us with insights into the
reasons behind translator’s choices as well as access to verbal ATSs. The
present study takes Munday’s investigation a step further by exploring
how the translator approaches ATSs in the post-drafting phase. In-depth
analysis of such a core and abundant element of the translation process
could help us improve our understanding of translator behaviour. The
processes governing the generation of solutions during the translation
process are not yet fully understood, but they frequently lead to self-
revisions (Jakobsen 2019:67). Self-revision is, in fact, another key aspect
of translator behaviour explored in this study and we will now turn our
attention to it.
1.5 Revision in CTS

Revision is a significant feature of the translation process (Mossop 1982;
Hansen 2013; Jakobsen 2019); it is integral to translation (Malkiel
2009:150) and plays a fundamental role in the shaping of the final
translation (Englund Dimitrova 2005:143) and its quality (Lehka-Paul
2020:107). This is also highlighted by empirical evidence which shows
that revision plays a central role in translation processes (e.g. Englund
Dimitrova 2005) and that translators attribute an important role to revi-
sion (e.g. Shih 2006:296). Despite its significance, and although interest in
this area has grown considerably (see, e.g., Koponen et al. 2021), empir-
ical research on translation revision is still relatively limited (Jakobson
2019:71) and this is even more so in relation to the literary translation
process (Koponen et al. 2021:9).
1.5.1 Definitions
In TS literature, the lack of terminological agreement on revision is
evident and many scholars highlight this problem (e.g. Mossop 2011;
Rasmussen & Schjoldager 2011; Koponen et al. 2021). The term “revi-
sion” has been used both for when translators check their own work
as well as when they check the work of others, which are two distinct
activities. On the other hand, the same activity is sometimes called dif-
ferently, for example the terms “revision” (e.g. Hansen 2008:259),
“other-revision” (e.g. Mossop 2007:6) and “translation revision” (e.g.
Robert 2008:8) have all been employed interchangeably to refer to the
checking of other translators’ work. Consequently, the reader often has
to stop and wonder about which kind of activity a term is referring
to.9 Clear definitions of the terms adopted in one’s work are therefore
necessary.
02
To distinguish between revising one’s own work and revising other
translators’ work, Mossop coins two terms: “self- revision” (2001/
2010:167) for the former and “other-revision” (2007:6) for the latter.
Thus, in self-revision, the translator checks his or her own work; the
translator and the reviser are the same person. Other-revision involves
revising the work of other translators. This is an important distinction for
the current study, which is mostly concerned with self-revision.
Hansen views self- revision and other- revision as two different
processes (2008:263) and, in fact, studies of revisions can be divided
into two broad categories: those looking at other-revision and those
looking at self- revision. Scholars generally agree that self- revision is
part and parcel of the translation process and empirical evidence shows
that most translators undertake some form of self-revision (e.g. Asadi &
Séguinot 2005; Englund Dimitrova 2005; Dragsted & Carl 2013).
However, scholars take different standpoints on other- revision. For
instance, Mossop sees both other- revision and self-revision as an
integral part of the translation process (1982:6), whereas for Robert
(2008: 4–5), other-revision is a post process, occurring after the transla-
tion process. In agreement with Robert, other-revision is here seen as
part of the post-translation process, since it is not undertaken by the
same translator who translated the text but by another individual. In
other-revision, the translation changes hands; it becomes the remit of
the reviser who may or may not consult the translator on the changes
deemed necessary. For these reasons, in my study, the translation pro-
cess is considered over when the role of third parties begins. Taken
together, the translation process and the post-translation process would
then constitute the extended translation process (Figure 1.3). As this
research is mainly concerned with self-revision, other-revision is only
discussed briefly in Section 1.5.4.
Extended Translation Process

Translator and third parties
Translation process Post-translation

process
Translator Third parties and translator
Post- Other- Publication

Pre-drafting Drafting
drafting revision phase
Figure 1.3 The extended translation process.

12
1.5.2 Self-revision
Yet again, authors use a variety of terms to denote the concept of self-
revision: for example “proof reading (own work)” (Samuelsson-Brown
1993:109), “self-corrections” (Malkiel 2009:150), some simply call it
“revision” (e.g. Breedveld 2002; Englund Dimitrova 2005), while Shih
(2006:296) applies “revision” and “self-revision” interchangeably.
1.5.2.1 How is self-revision defined in the literature?

Mossop (2007:109) defines self-revision as the operation during which
translators “identify features of the draft translation that fall short of
what is acceptable and make appropriate corrections and improvements”.
Lehka-Paul (2020:117) sees self-revision as an activity involving decision-
making and quality-assurance that could be impacted by the translator’s
personality. For Antunović and Pavlović (2011:214), self-revision involves
all of the changes that translators make to a TT, excluding typographical
mistakes. A similar definition is offered by Malkiel (2009:150), who speci-
fies that self-revisions comprise the additions, deletions and changes made
to a translation. Malkiel (2009:159) also eliminates changes made to
spelling because she sees typos as simply revealing the translator’s typing
skills and hardly offering any useful insights into the translation process.
The elimination of typos from studies of self-revision is questionable,
as their correction is certainly part of the self-revision process. In fact,
other scholars (e.g. Dragsted & Carl 2013:146) incorporate them in their
investigations of self-revisions, and Muñoz (2009:183) sees much value in
typographical errors, claiming they could indicate attention lapses.
Several studies have shown that self-revision is not restrained to the
post-drafting phase but also blends in with the drafting process (e.g.
Jakobsen 2002; Asadi & Séguinot 2005; Englund Dimitrova 2005; Carl
et al. 2010; Lehka-Paul 2020). Thus, self-revision is often spread over more
than one phase, occurring both during the production of the draft and
afterwards. It is not a separate phase of the translation process but inter-
woven with it. Englund Dimitrova views self-revision (which she simply
calls revision) as one of the three cognitive processes of translation –i)
planning, ii) text generation, and iii) revision (2005:10) –that takes place
over the drafting phase and the post-drafting phase (2005:106). Jakobsen
(2002), too, observed this phenomenon and creates two terms to distin-
guish between two types of self-revisions: “online revisions” performed
while the first draft is still in production and “end-revisions” undertaken
after the completion of the first full draft. The current study distinguishes
between the two and covers both kinds of self-revisions. Self-revisions
carried out in D1 are labelled online self-revisions (OSRs), while those
undertaken after D1 are here called self-revisions.10
Self-revising one’s work is not a straightforward activity; it is maybe
“the most difficult task in the process of translation” (Samuelsson-Brown
2
1993:109). Hansen (2008:263) explains that self-revision is particularly
difficult because translators get attached to their constructions, which are
often repetitive. Despite its significance and complexity, self-revision has
not received much attention in TS research (Mossop 2001/2010; Lehka-
Paul 2018), particularly in literary translation and workplace research.
By focusing on self-revision in literary translation, this book contributes
to narrowing this gap in the translation revision literature.
1.5.2.2 Empirical studies of self-revision

Empirical studies of revisions examine how translation revision takes
place and reveal the mental processes linked with this activity. Several
scholars have explored different aspects of self-revisions. Some studies
(e.g. Dragsted & Carl 2013) look at the distribution of self-revisions over
the translation process, distinguishing between online and end-revisions.
Another such study is Carl et al. (2010:6), where the authors introduce a
further distinction by differentiating between short-distance revisions and
long-distance revisions. They define the latter as self-revisions involving
an item in the text which is more than two words away from the last
keystroke and suggest that long- distance revisions reflect translation
problems, while short-distance corrections are probably related to typos.
Innovatively, Alves and Vale (2011) apply corpus-based techniques to
investigate the distribution of self- revisions. Besides online and end-
revisions, they identify a third type of self-revision consisting of items
revised during the drafting phase and revised again during the post-
drafting phase. A qualitative analysis of the three kinds of self-revisions
observed by Alves and Vale (2011) could be revealing in terms of what is
actually revised during the different phases of the process.
Other studies investigate student self-revision processes (e.g. Malkiel
2009; Antunović and Pavlović 2011; Huang 2018) or compare self-
revision processes of novice and experienced translators. For instance,
Jakobsen (2002) found that professionals carry out less end-revisions
than students. Englund Dimitrova (2005:145) and Denver (2009:142)
also found that experienced translators make fewer self-revisions than
students. Alves and Liparini Campos (2009) studied the type of support
that professionals use during the translation process and showed that,
during self-revision, professional translators rely mostly on their personal
knowledge. They observed widespread online revision and found that
professional translators solve the majority of problems during the drafting
phase; therefore, few problems are left pending in the post- drafting
phase. In 2005, Englund Dimitrova made a similar finding: the majority
of self-revisions made by her experienced translators were done in the
drafting phase (2005:117–120). This is also corroborated by Alves and
Vale (2011:114) and by Lehka-Paul (2020:182), whose study focused
on the role of the translator’s personality in the process of self-revision.
In contrast, Carl et al. (2010:7) concluded that experienced translators
32
delayed self-revisions to the post-drafting phase. Lehka-Paul (2020:182–
183, 214), moreover, found that practising translators performed more
self-revisions in the post-drafting phase than translation students, and
that translators with a thinking type personality dedicate more time to
post-drafting and make more changes both in terms of quantity and of
meaning during this phase. Also, three out of the five literary translators
participating in Kolb’s (2017:36) study spent more time post-drafting
than drafting their work. Findings on this matter are therefore inconclu-
sive, pointing to the need for further studies.
I will now focus on Englund Dimitrova’s (2005) seminal study, which
makes a valuable contribution to the study of self-revision and offers
useful insights, particularly on how to count and categorise such revisions.
She combines think-aloud with keystroke logging and triangulates the
data yielded, which is one of the many strengths of the study. Englund
Dimitrova looks at the three phases of the translation process, comparing
the process and product of nine individuals with different levels of trans-
lation experience. She analyses self-revision through the examination of
interim solutions and covers both online and end-revisions, thus pro-
viding rich empirical evidence on self-revision.
In addition to the findings discussed earlier, some salient findings in
Englund Dimitrova’s study include the observation that all participants,
irrespective of their translation experience, made some online revisions,
leading the author to conclude that online revision is partly an auto-
matic process (2005:143). It was found that, although professionals do
check their translation in the post-drafting phase, they carry out few
self-revisions during this phase, leading the author to conclude that
professionals are able to tackle problems and be satisfied with them
earlier in the translation process (2005:147). Another relevant finding
was that, during self-revision, the translators focused mostly on the TT
(2005:125–126). It will be interesting to examine whether the current
study corroborates these findings.
Of particular interest is the counting and classification system that
Englund Dimitrova (2005:113) devised for the analysis of self-revisions.
Her counting system is similar to the one adopted by Malkiel (2009),
while Antunović and Pavlović (2011) draw on both authors to count
self-revisions, and their classification system is quite similar to Englund
Dimitrova’s. Englund Dimitrova’s classification scheme has thus proved
useful and practical for the study of self-revision. She organises self-
revisions in six categories: syntactic, lexical, morphological, content,
orthographic and other, and provides quite detailed information about
the different categories backed up with examples (2005:113–116). This
makes her methodology transparent and replicable. Her data revealed
that, overall, the most common self- revisions in both the drafting
and post-drafting phases are of the lexical type followed by syntactic
corrections. However, she observed that syntactic revisions are higher
in the drafting phase and the highest number of self-revisions made by
42
three out of her four professional subjects while drafting pertained to
syntax (2005:116); this points to the possibility that translators concen-
trate more on the construction of the sentence during the drafting phase
(2005:145). For these reasons, the classification system for self-revisions
developed for the present study is based on Englund Dimitrova’s and,
again, it will be interesting to find out whether this study corroborates the
findings mentioned above.
1.5.2.3 The effect of self-revision on the translation process:

Do self-revisions deliteralise the final translation?
Interestingly, Englund Dimitrova (2005:121) observed that profes-
sional translators first translate short segments literally, which are then
revised to a less literal version, with her analysis revealing that profes-
sional translators’ online self-revisions make the text less literal, in other
words less closer to the ST. Back in 1995, Toury studied the self-revisions
made by a translator to a literary translation and made an analogous
finding: the first solutions were more literal and were revised into less
literal alternatives (1995:204). A similar finding is also made by Munday
(2013:132). Likewise, Tirkkonen-Condit et al. (2008:4–5) report that
20.5% of the self-revisions carried out by their participants pertain to the
removal of literal translations and that this phenomenon occurs for all
linguistic categories –that is, at the lexical, morpho-syntactic, syntactic
and textual levels. Therefore, previous research has shown that one of the
effects of self-revision is a decrease in word-for-word translations. Many
translators seem to start off by translating segments literally, and after-
wards they revise their texts to remove some of their literalness, thereby
moving away from the ST as the translation process progresses. However,
Pavlović and Antunović (2013) tested the literal translation hypothesis
and their findings challenged it. Their study involved twelve professional
translators and interpreters (6+6) who translated a short non-literary text
under time constraints using Translog. They examined the self-revisions
made by these two groups of professionals to see whether they literalise
or deliteralise the text. Interestingly, they found that 39.47% of all self-
revisions deliteralised the text, 26.75% literalised the TT, and 33.77%
were neutral self-revisions that neither literalised nor deliteralised the
translation. They argue that their findings do not provide conclusive evi-
dence for the literal translation hypothesis: although deliteralising self-
revisions have the highest percentage (39.47%), they question whether
this is predominant enough to “unequivocally” claim that the self-revision
process moves from more literal to less literal renderings (2013:243).
Translation theorist Andrew Chesterman, in an article on the literal
hypothesis, puts forward a deliteralisation hypothesis that could be tested
by comparing different draft versions of the same translation. Defining the
literal hypothesis as “during the translation process, translators tend to
proceed from more literal versions to less literal ones” (2011:26), his own
52
deliteralisation hypothesis claims that “initial (or earlier) draft version
A is formally closer to the source than the later version B” (2011:26).
Chesterman (2011:26) asserts that this is the result of the ST influence on
the translator’s cognitive processes. Furthermore, this is also assumed to
be the result of cognitive overload during the drafting phase. The literal
translation hypothesis is a major hypothesis in TS; it has been discussed
by numerous scholars, although different labels were sometimes applied
to denote a similar concept (see, e.g., Ivir 1981; Toury 1995; Englund
Dimitrova 2005; Tirkkonen-Condit 2005; Schaeffer & Carl 2014). For
reasons of space, the discussion will not be replicated here. However, in
view of Chesterman’s hypothesis and the empirical findings mentioned
above, it is interesting to look at this aspect in the present study, as it
analyses self-revision in different draft versions of the same translation,
and one of its objectives is to examine the effects of revision on the TT.
Moreover, Englund Dimitrova (2005:148) encourages other researchers
to examine this phenomenon in studies involving different language pairs
and text genres. The present study can offer a contribution in this regard
since it deals with a literary translation from French into Maltese; it will
therefore investigate whether Aquilina’s self-revisions deliteralise the final
translation or not.
1.5.3 The post-drafting phase

We will now discuss the third and final phase of the translation process,
the post-drafting phase, since the present study focuses on this phase.
It was shown that self-revision is not restricted to the third phase but
is spread over both the drafting and the post-drafting phases. The pri-
mary activity of the post-drafting phase is monitoring the existing trans-
lation (Jakobsen 2002:193): translators read the TT, identify problems
and make any changes deemed necessary. Sometimes, some non-checking
work, for example researching a term, is performed here too (Mossop
2001/2010:168), and some decisions could be postponed until this phase.
Some translators choose to subdivide this phase into various subphases
(Englund Dimitrova 2005:86; Huang 2016:121, 191), which begins when
the translator finishes the first draft of the translation and ends when
s/he considers the translation complete (Jakobsen 2002:193; Englund
Dimitrova 2005:86).
From the literature, it is evident that the post-drafting phase is an
important phase in the translation process. Previous research has shown
that translators tend to spend an important part of their time on this phase.
Jakobsen found that professionals dedicated 24% of the total transla-
tion task to this phase (2002:194). Englund Dimitrova reported that all
her participants, irrespective of their translation experience, performed
a post-drafting phase (2005:106) on which they spent a large propor-
tion of their time (2005:136). This was corroborated by Antunović and
Pavlović, whose subjects spent, on average, 32% of the total translation
62
time on the post- drafting phase (2011:225). Empirical evidence thus
highlights the significance of the post-drafting phase. However, to date it
has received only scant attention in process research (Carl et al. 2010:1;
Shih 2013:32), as most studies investigate the three phases of the transla-
tion process and generally focus on the second phase and on non-literary
translation.
Hence, empirical research tends to focus on online self-revision and
rather neglects self-revision during post-drafting. Studies of self-revision
either do not distinguish between the two (e.g. Asadi & Séguinot 2005;
Malkiel 2009) or else give particular attention to online self-revision
probably because it occurs during what is considered as the main trans-
lation phase: the drafting phase. Two exceptions are Shih 2013’s and
Huang 2016’s investigations. Huang (2016) studied students’ self-
revision processes during the post-drafting phase with the aim of iden-
tifying the revision working styles of student translators, while Shih
(2013), employing TAPs to examine end-revision processes and patterns,
corroborates Englund Dimitrova’s finding that self-revisions concerning
lexical items are the most prevalent. More interestingly, she reports that
later draft versions are processed in longer chunks, with no backtracking
and few references to the ST. She also found that taking a break after the
drafting phase is beneficial not only because translators are reinvigorated
but also because it helps them to detach themselves from the ST and be
more critical when self-revising the TT (2013:42). These are interesting
insights into the post-drafting phase, even more so as the first draft
involved in the present study remained in the translator’s drawer for
quite some time. According to Shih, this implies that the translator will
approach the draft with fresh eyes.
Although Shih’s (2013), Huang’s (2016) and the current study all
explore the post-drafting phase, their research design is entirely different,
as both Shih and Huang employ a “classic”11 design, with twelve and
eighteen participants respectively working on a short general text.
Examining in detail how one translator deals with the post- drafting
phase of a literary translation intended for publication is therefore intri-
guing and will help provide a deeper understanding of the translation
process. The next section focuses on the post-translation process and
other-revision.
1.5.4 The post-translation process and other-revision

In this study, the “post-translation process” refers to the part of the pro-
cess from when the translation leaves the translator’s hands and reaches
a third party, generally the commissioner of the translation or a reviser,
until it is published (see Figure 1.3 in Section 1.5.1). It encompasses all the
activities that take place after the translator submits his or her translation
to a third party and includes other-revision and the publication phase.
In literary translation, the post-translation process is generally called the
72
publication process, which comprises editing/copyediting, the terms gen-
erally used in the industry for other-revision (see, e.g., Mossop 1982;
Freely 2013; Solum 2018). Since this study is being undertaken within
a CTS framework, a distinction is being made between other-revision
and the publication process, and both are grouped under the label “post-
translation process”. The term “other-revision” is being applied in line
with other CTS studies.
Most process studies carried out so far terminate at the post-drafting
phase once participants consider they have finished their task (e.g.
Jakobsen 2002; Englund Dimitrova 2005; Kolb 2013, 2017; Shih 2013),
while other studies examine solely other- revision (e.g. Künzli 2006,
2007; Robert 2014). Hence, the two processes have mostly been studied
separately. In the real world, once translators complete the transla-
tion, another process, here called the post-translation process, generally
commences: the translation is revised by a third party (Künzli 2006:193)
and, if it is intended for publication, the publication phase then ensues.
Ehrensberger-Dow (2014:362) affirms that “[i]‌f a translation job could
be followed from a translator’s to the reviser’s workplace as it is sent to be
revised, a realistic picture of professional translation and revision might
emerge”. To my knowledge, few process studies have attempted to cap-
ture both the translation process and the post-translation process like the
current project does, which describes the evolution of a literary transla-
tion from first draft until its publication. Hence, by examining the whole
trajectory of the translation until the publication phase, the present study
extends the scope of CTS. Some work along similar lines has been done in
the sociological approach to TS12 (e.g. Buzelin 2006, 2007a; Bogic 2010;
Luo 2020), but these studies applied a different theoretical framework
and mainly focused on publishers or the publication process and rarely
on translators or their decision-making processes. Interestingly, Buzelin
asserted that “the process of ‘making’ a literary translation has not, to my
knowledge, been the subject of any in-depth field study” (2007a:141); as
far as I know, the present study is the first to attempt such an endeavour
from a situated process perspective.
As other-revision is not the main focus of the current study, this aspect
will only be dealt with briefly. Scholars identify two main types of other-
revisions: the first involves revising the translation without referring to
the ST and the second involves comparing the ST with the TT. Mossop
(2007:6) calls the former unilingual and the latter comparative, while
Brunette et al. (2005:29) label them monolingual and bilingual respect-
ively. CTS scholars do not seem to distinguish between proofreading and
other-revision but seem to consider proofreading as unilingual other-
revision. A definition of proofreading in revision is, however, provided
by Mossop (2001/2010:200; emphasis in original): “a synonym of uni-
lingual re-reading, especially when this is limited to corrections (i.e. no
improvements are made)”. This is interesting as it allows for a distinc-
tion between proofreading and unilingual other-revision: proofreading
82
involves the correction of mistakes such as grammar and orthography,
while unilingual other-revision also involves making other improvements
to the text.
All the CTS studies centring on other- revision encountered so far
focus on non-literary translation (e.g. Künzli 2006, 2007; Robert 2012),
thus other-revision in literary translation has been largely overlooked
(Koponen et al. 2021:10) –particularly from a CTS perspective –even
though literary translations undergo other-revision too. Being able to
identify which decisions were made by the translator and which were
made by someone else, for example by a reviser, could be insightful, as
although the TT is often revised by a third party/third parties and at
times extensively so, the translation and the decisions contained therein
are attributed to the translator, especially in literary translation where the
translator’s name features in the publication. This is, in fact, what Bogic’s
2010 study of correspondence between the translator of Simone de
Beauvoir’s Le deuxième sexe and the publishing house reveals: while the
translator was slandered for the English translation, some of the decisions
he was blamed for were actually imposed on him by the publishing house.
As Toury argues, by studying only the ST and the final TT:
there is no way of knowing how many different persons were actu-

ally involved in the establishment of a translation, playing how many
different roles. Whatever the number, the common practice has been
to collapse all of them into one persona and have that conjoined
entity regarded as “the translator”.
[Toury 1995:183]
Exploring the relationship, especially the power relations, between trans-

lator and reviser and/or publisher is also worthwhile. Mossop (2001/
2010:174) lists various scenarios and asserts that, if the translator and
reviser are colleagues at the same rank, the reviser provides the trans-
lator with their recommended revisions; they might discuss them, but the
translator decides whether to accept them or not since they are respon-
sible for the final translation. He also describes other scenarios, such as
when the reviser either holds a senior position to the translator or else the
translator has been commissioned for the job, and in these two scenarios,
the reviser has the final say, as they are accountable for the translation’s
quality. To this, one must add that, apart from ranks and seniority, other
elements, for instance the translator’s prestige, self-confidence and rela-
tion with the publisher, may also come to bear. So who is finally respon-
sible for the translation depends on various factors. However, in the
literary world, the translator’s name appears in the publication but usu-
ally not the reviser’s; thus, in this case, it could be argued that the trans-
lator is ultimately responsible for the translation. Literary translation is
potentially different, as technical translations are usually anonymous.
Nevertheless, literary translations are “subject to manipulation,
92
fine-tuning and revision by third parties” (Buzelin 2007a:142), and we
find many reports in the literature where literary translators have been
overruled by revisers/publishers or had their work revised without being
consulted (e.g. Woods 2006; Munday 2008a; Venuti 2013). Novelist and
literary translator Maureen Freely (2013:121–122) recounts how, after
having translated Orhan Pamuk’s Snow into English, her publishers were
stunned by her request to be consulted during the publication process;
she insisted and got her own way, which was fortunate because various
changes had been made to the text by third parties, including a rewriting
of the last paragraph, and, in the process, various mistakes had been
introduced into the translation that would bear her name.
The Freely example highlights various points. First, that some
publishers do not treat authors and translators equally: it seems that
they consult authors but not always translators during the publication
phase. Second, it underlines that certain decisions and mistakes are not
made by translators but are still ascribed to them. Third, it demonstrates
the importance of being assertive, of insisting on being consulted by
publishers. This will not only increase the visibility of translators among
publishers but will also avoid unwarranted changes and mistakes. On a
more positive note, Marin-Lacarta and Vargas-Urpi’s (2019) study on
revision processes in a non-profit digital publisher has shown that literary
translators appreciate being involved in the revision of their translations
and engage quite willingly with revisers to improve their translations.
In their study, because translators had the final say, the revision process
turned out to be a positive experience devoid of conflict and tension.
Thus, the inclusion of translators in the revision and publication phases
seems to be fruitful and important.
1.6 Individual variation in translation processes

A recurring observation in process- oriented studies is the great vari-
ation in individual working habits. The literature is replete with evi-
dence of individual differences in the processes adopted by translators
(e.g. Jensen 1999; Tirkkonen-Condit 2000; Krings 2001; Jakobsen 2002;
Alves & Vale 2011; Hansen 2013), and these variations have been noted
for self-revision processes too (e.g. Asadi & Séguinot 2005; Englund
Dimitrova 2005; Antunović & Pavlović 2011; Lehka-Paul 2020), with
Shih (2006:310) concluding that professional translators “develop their
own idiosyncratic habits of revision”. Thus, no two translators translate
and revise in exactly the same way and, as a result, no two translations
are identical. This individual variation in the handling of the task is often
related to the notion of process profiles, discussed in Section 5.1.3.
The socio-cultural environment within which translators work cer-
tainly influences and constrains their choices (Toury 1995), but two
translators working in the same socio-cultural setting would nonetheless
produce two different products (see Munday 2008a). Hence, in addition
03
to outside influences, there are also internal influences operating on the
translator, such as personal preferences, idiosyncratic differences, skills,
knowledge and experience. These all influence translatorial decisions.
According to Krings (2001:464), most of the time a decision can take
different paths as there are often various possible solutions. Therefore,
different individuals may make different decisions in the same situation
(Wilss 1998:57). Moreover, everyone has their own style of making
decisions (Jennings & Wattam 1998:xvii). Consequently, individual
differences in translation are important on several counts, as translators
may differ not only in the outcome of their decisions, reflected in TT
differences, but also in the way they reach decisions, make choices and
arrive at their final TT. Furthermore, according to Wilss (1994:132), indi-
vidual differences are even more significant in non-routine translation,
particularly in literary translation, and therefore we must take them into
account. Yet despite clear evidence that translators vary a lot in the way
they produce translations (Carl et al. 2010:2), very few (if any) process
studies have investigated in-depth how particular translators approach
specific tasks, for instance with in- depth investigations of translators
handling a long translation.
In fact, almost all process studies conducted so far have involved short
texts, sometimes very short indeed. For instance, in Jakobsen (2002:194),
the texts chosen were between 367 and 1001 characters long, approxi-
mately equivalent to 70 and 170 words, and in Lehka-Paul (2020:154–
155), the longest ST comprised 269 words. Consequently, the duration
of the experiments was very short too, in some cases amounting to only
a few minutes (e.g. Carl et al. 2010; Dragsted & Carl 2013). Two of the
longest texts encountered in CTS literature are the STs used in Kolb’s
work (2011, 2013, 2017) and in Englund Dimitrova and Tiselius’s (2009,
2014) studies, comprising 637 and 1,093 words respectively. Therefore,
we have a multitude of insightful and mainly quantitative studies based
on short texts and short experiments, but a lack of qualitative studies
looking at translator behaviour during a long translation task. Breedveld
maintains that if small chunks of texts reveal important findings, “then it
seems promising to investigate translation processes of somewhat larger
units or even integral texts” (2002:93).
For all these reasons, qualitative and detailed process studies of how
individual translators tackle specific translation tasks are now needed.
The present study brings an original contribution to TS by investigating
in-depth the process adopted by an experienced translator to self-revise
a full literary text. This entails a different research design from the main-
stream method adopted so far: a sole research participant and a long, full
text, instead of short texts and a number of research participants. This
approach could help us gain new insights into translator behaviour and,
if such studies proliferate in the future, a wealth of knowledge on trans-
lator behaviour could be accumulated and comparisons between similar
studies could then be drawn. My argument is backed by Risku et al.’s
13
call for thorough investigations on particular aspects of the production
process that
would illustrate specific parts of the “coming-into-being” of a trans-

lation and critically engage with it and the process that led to its
existence. Taken together, these case studies will contribute towards a
better understanding of the detailed history of creation of translations.
[Risku et al. 2013:169]
Moreover, Hansen (2010, 2013), basing herself on her long experience

in empirical process research and on observations she made in her trans-
lation classes, holds that every single translator possesses an individual
competence pattern that impacts both the way translators approach the
text and their products. In view of this, Hansen (2010) calls for inte-
grative descriptions of translation processes that combine quantitative
and qualitative data and take into account translators’ life stories (e.g.
their professional career, personal experiences, values and thoughts). She
(2010:204; emphasis in original) argues that such an approach “fits the
complex character of ‘translation’ as a process and a product and has a
focus equally on human beings (and their profiles), translation processes
and texts in situations”. The present study responds to Hansen’s call,
which was later supported by O’Brien (2011:11), who saw value in inte-
grative description, as it could help us understand and explain individual
variation in translation processes.
It is interesting to note, however, that House (2013:50) disagrees with
Hansen and O’Brien on this point because, according to her, a more
personalised approach to TS would prevent the discipline from being
acknowledged as a science and taken seriously; in her view, TS scholars
should aim to generalise. It should, however, be pointed out that even
scientists are highlighting the value of idiographic research and appealing
for in-depth studies focusing on individuals in their respective disciplines
(e.g. Kravitz et al. 2008). One approach does not exclude the other, but
both approaches should be seen as complementary, with one enriching
the other. Variation is so prominent in translation processes that it cannot
be ignored: we now need to start paying attention to it to better under-
stand its roots and causes.
1.7 Literary translation in CTS

As previously mentioned, literary texts have received little attention
in CTS (Kolb 2011:260; Jääskeläinen 2012:194). Only a handful of
scholars have adopted a process-oriented approach to the study of lit-
erary translations. For example, Audet and Dancette (2005) used think-
aloud and draft versions of extracts from a novella to investigate literary
perception, while Jones (2011) examined expert poetry translation by
means of think-aloud and interviews. Kolb (2011, 2013, 2017) employed
23
verbal reports, keystroke logging and the draft and final translations of a
short story to explore situated cognitive processes in professional literary
translation from English into German. Kolb (2011) traced the translator’s
voice in the TT, and in her 2013 article she delved into how literary
translators construct meaning when the ST is ambiguous and vague. Kolb
(2011:261) maintains that certain aspects such as style, ambiguity and
the translator’s self-concept are salient in literary translation but not so
central to non-literary translation and, since process research has mainly
involved non-literary texts, such aspects have been neglected. Her 2017
study, focusing on macrolevel workplace practices, showed that literary
translation processes are very fragmented. All five home-based freelance
translators spread their work on the 637-word ST over several sessions
and days, dotted with frequent breaks or interruptions. These pauses
affected the translator’s voice, i.e. their textual choices, as did input from
spouses. Kolb is one of the very few researchers looking at literary trans-
lation as a situated activity from a CTS perspective. The lion’s share of
previous CTS studies have dealt with non-literary translation; literary
translation has hardly benefitted at all from the wealth of insights that
can be gained from CTS studies.
Literary translators are increasingly writing essays (e.g. Bassnett &
Bush 2006; Anderman 2007; Allen & Bernofsky 2013) in which they
reflect on their own translation practices and processes. For example,
Peter Bush (2006) discusses the different phases of his translation pro-
cess, Bernofsky (2013) comments on her self-revision process and Jull
Costa (2007) sheds light on translatorial problems in literary transla-
tion. In his fascinating Translation Diary (2021), Hahn recently laid bare
his own translation process of a short novel from Spanish into English.
Albeit mostly non-empirical, these writings provide interesting insights
into literary translators’ processes, approaches and practices. Given the
limited CTS research on literary translation, such contributions pro-
vide additional information against which this research’s findings will be
compared. These writings on the literary translation process will therefore
be woven into the discussion of this study’s results (Chapters 5 and 6).
1.8 Studies of draft translations

As self-revision is an intrinsic part of the translation process, studies of
self-revision necessarily entail probing process data. Comparison of an
ST and a final TT leaves us in the dark about revision processes, since
revision data are lost in the process and leave no traces in the final trans-
lation. Researchers have employed various methods to investigate phe-
nomena related to revision. The most prevalent method reported in the
literature is keystroke logging (e.g. Malkiel 2009; Antunović & Pavlović
2011; Lehka-Paul 2020). Keylogging programs are instrumental in
studying revision empirically as they capture certain process data such
as modifications and corrections, but they do not capture the full scope
Another random document with
no related content on Scribd:
in this way; so that in reality the cost of one’s entertainment falls on
the whole village.
The majority of the natives of these oases are miserably poor, and
it goes much against the grain for a European to have to live upon
them in this way. But to refuse their hospitality would be considered
as a slight, if not as an actual insult, and so would any attempt to
offer them any payment in return.
The meals, as a rule, were quite well cooked, and usually better
than I got in camp. It was the tea and cigarettes that were such a
trial. The one luxury the inhabitants of the oasis allow themselves is
tea; even the poorest of them consume enormous quantities. The
quality of the tea in the better class houses is irreproachable. The
best of it is said to come from Persia, and I was told that as much as
£1 a rotl (the Egyptian pound) is paid for it. In addition to red tea, a
green tea, and also a brown and a black are used. The last I only
tasted once; it seemed to be of an inferior quality. The richer natives
will often offer two or even three different kinds in succession.
After drinking, it is quite the correct thing to sit silent for some time
licking and smacking one’s lips, “tasting the tea” as it is called, as a
compliment to the quality supplied by one’s host. The natives have
another way of showing their appreciation of the fare set before
them, which, however, it would be better not to describe.
The greatest ordeal I had to face was not the tea but the
cigarettes. My host would extract from somewhere in the voluminous
folds of his clothing a large shiny papier mâché tobacco-box, inlaid
with mother-of-pearl, from which he would produce some tobacco
and cigarette papers and proceed to roll me a cigarette, which he
then licked down.
Eventually I found a means to avoid them. If the cigarette was
offered me before the tea, I placed it above my ear—the correct
position to carry it in the oases—and explained that I would smoke it
later, so as to avoid spoiling the tea. If it was handed to me after the
tea drinking, I was able to postpone lighting it for a time by saying
that I would not smoke it just then, as I was still “tasting” the tea;
then, while still licking and smacking my lips, with the cigarette still
unsmoked above my ear, I found that it was time to take my
departure. Once safely outside my host’s house in the desert, the
cigarette would fall down from my ear and be promptly scrambled for
by my men.
In Smint, however, no cigarettes were forthcoming. The reason
was not far to seek. Close to the village the Senussi had built a
zawia, and a large number of the inhabitants of the village had
already been converted to the tenets of the sect, or, as the natives
put it, they “followed the sheykh.” The members of this sect are
forbidden to smoke.
SENUSSI ZAWIA AT SMINT.
In company with the ’omda we went to call on the sheykh of the

zawia. After speaking to us for a minute or two, he rather sulkily
invited us to enter and treated us to the usual tea.
The zawia was an entirely unpretentious looking mud-built
building, and might have been only the house of a well-to-do villager.
The head of it—Sheykh Senussi by name—was quite a young man
in the early twenties, and had probably been given the position
owing to the fact that he had married a daughter of Sheykh
Mohammed el Mawhub, the chief Senussi sheykh in Dakhla, who
himself had a zawia at Qasr Dakhl, the largest town in the oasis,
situated in its north-western corner.
He was said to be an Arab from Tripoli way, a statement that was
borne out by his clothing, which consisted of the ordinary white hram
of a Tripolitan Arab of the poorer class. He was very silent during the
whole of our visit, and when he did condescend to speak it was
generally to sneer or laugh at some remark that we made. The
interview was consequently cut as short as possible.
Re-soling a Camel’s Foot.
The sharp rocks of the desert sometimes flay the entire skin from the sole of a camel’s
foot, the Arabs replace this with a piece of leather sewn on to the camel’s foot. (p. 35).
After Qway had succeeded in extracting some barley for his

camel off the ’omda we started again, for Mut, which lay about six
miles away to the west.
In many parts the scenery of these oases is extremely pretty. Our
road to Mut lay through cultivated fields, alternating with areas of
salt-encrusted land, and sprinkled with palm plantations and low
earthy hills. Away to the north at the foot of the cliff that bounds the
oasis lay the palm groves of the village of Hindau. The fields, with
their ripening grain and green crops of bersim (clover), the yellow
ochreous hills, the clumps of graceful date palms with their dark
green foliage, set against a background of cream-coloured sand
dunes and purple cliffs, made a lovely picture in the light of the
setting sun.
As we neared Mut, however, the country became less productive.
Large areas of land thickly encrusted with salt and barren stretches
of desert replaced the fertile fields and palm groves in the
neighbourhood of Masara and Smint. Owing probably to the sinking
of new wells at a lower level in the village of Rashida, the water
supply of Mut has for many years been falling off, and now, although
the place is the capital town, the district in which it lies is one of the
poorest in the whole oasis.
We reached Mut in the dusk soon after sunset. Built on a low hill,
and seen in the failing light, the place gave rather the impression of
an old medieval fortified town. We skirted round its southern side,
past a number of walled enclosures used to pen the cattle in at night,
and, passing through a gap in the south-western corner of the wall
that surrounds the town, arrived at a large rambling mud-built
building, mainly used as a store, in which I had received leave to
stay. It was a gloomy-looking place, and had evidently been built with
a view to defence. Entering through a gate in the wall, secured by a
bar, and turning to the right past some low outbuildings, we found
ourselves in a narrow court, surrounded on three sides by high two-
storied buildings—the upper part having apparently been used at
some time as a harem by one of its former inmates.
Doors opened from either end of a gallery that joined the two
wings. One led into the centre of three rooms on the western side
that looked over the desert, and the other into some small chambers
which, as one had a fire-place in it for cooking, I allotted to Dahab
and Khalil, retaining the three western rooms for my own use.
OLD HOUSES IN MUT.
These proved to be high, spacious and airy, and commanded a

fine view over the desert. The windows were large and fitted with a
sort of trellis. This not only made the rooms more private, but
considerably reduced the glare of the desert. So beyond the fact that
the floors in many places seemed unsafe, and that the place was
said to swarm with scorpions, I had little fault to find with my
lodgings.
I walked out in the dusk as soon as we had settled into our
quarters in the old store, to see what I could of the town. Many of the
streets were roofed over, as in Kharga Oasis, but the tunnels were
not nearly so long and very considerably higher, so that, except for
the unevenness of the roadway, we had no difficulty in getting about.
We were, however, compelled to carry a lantern in order to find our
way.
There was not much to be seen; but the monotonous thudding of
the women pounding rice, the continuous rumbling sound of the
small stone hand mills by which they were grinding grain, the smell
of wood smoke, the soft singing of the women and an occasional bar
of ruddy light, crossing the roadway from some partly open doorway,
showed that most of the inhabitants were in their houses preparing
their evening meal.
Rice enters largely into the bill of fare of the natives of the oases,
and is pounded by the women with a large stone held in both hands,
which is brought down with all their strength into a small basin-
shaped hollow scooped out of the rocky sandstone floor upon which
the town is built.
The following morning I received a state visit from the mamur
(magistrate), Ibrahim Zaky by name, the doctor, Gorgi Michael, a
Copt from Syria, and the zabit, or police officer. The mamur and
doctor spoke English fairly well.
Like most of the native officials who are to be found in the oases,
the mamur was rather under a cloud, and had been sent to Dakhla
as a punishment for some misdeeds of his in his last appointment.
These oases posts are cordially disliked by the natives, as in these
remote districts they are entirely cut off from the gay life of the towns
of the Nile Valley. The appointments, however, have certain
advantages. Being so far removed from the towns of the Nile Valley
may be dull, but it frees them from the constant supervision of the
English inspectors, a state of things of which an Egyptian is usually
not slow to take advantage, by extorting bakhshish from the
wretched fellahin of their district—often to a most outrageous extent.
One of the English inspectors had very kindly written to the
mamur to inform him that I was coming into his district, and to tell
him to help me in any way he could. The mamur’s term of office in
Dakhla being nearly at an end, he was extremely anxious to get my
good word with the inspector in order that he might be appointed to a
better district. He was accordingly most oppressive and unremitting
in his attentions—until the government removed him to another and
still worse district.
He was by no means enthusiastic about his life in the oasis, and,
from his account of the natives, he evidently looked upon them as
being little removed from beasts. He explained that he had left his
wife behind in Egypt, but as he found that he did not get on well
without one, he had married a young girl from Mut. He complained
bitterly of the expense she had put him to, for as he expressed it in
his rather defective English, it had “cost him £25 to make her clean!”
After the Egyptian officials had departed, a succession of ’omdas
from all over the oasis dropped in to pay their respects and to ask
me to come round to their villages.
After the ’omdas came various minor fry. First the camel postman,
a burly, black-bearded Arab, called ’Ali Kashuta, looked in, drank a
gallon or two of tea, took a handful of cigarettes out of the box that
was handed to him, told me several times that he was my servant,
and obviously didn’t mean it; and then asking if I had any letters for
post, departed, leaving a breezy independent atmosphere behind
him, which was a pleasant contrast to the fawning attitude of the
other natives.
Then came the clerk to the Qadi, Sheykh Senussi, who was also
a member of the Senussi sect. He was a very learned person and a
poet in his leisure moments. He drank tea, but didn’t smoke, and
was all smiles and compliments.
Next came the postmaster. He had been to school in the Nile
Valley and spoke English quite well. He explained—what I was
beginning to realise—that I was causing much mystification to the
good people of the oasis; they could not make me out at all. The
postmaster, however, who had been educated in Egypt, knew all
about it. He had read about a man called “Keristoffer Kolombos,”
who had found America, and he thought that I must be in the same
line of business. I told him that he was quite right. He beamed all
over, and immediately departed to break the good news to an
expectant oasis that the great problem had been solved. Before
going he wished that Allah would preserve me on my journey, and
hoped that I should find another America in the Libyan Desert.
In the afternoon I went round to tea with the mamur in the merkaz,
or official residence.
One of his guests was a tall intelligent looking man, who was
introduced to me as the ’omda of Rashida, the mamur adding in
English that he was one of the most hospitable men in the oasis; but
very fond of whisky.
The latter statement unfortunately proved to be true. According to
the mamur, he was a most depraved and habitual drunkard. This,
however, was an exaggeration.
Between him and this ’omda there was very little love lost. Shortly
before my arrival they had quarrelled furiously. I never heard the
cause of the dispute—it was probably a case of cherchez la femme,
for Dakhla is one of those unfortunate places where, as Byron so
nearly expressed it, “man’s love is of his wife a thing apart, ’tis
woman’s whole persistence.” These small-minded natives will
squabble over the most trivial matters and keep the quarrel going for
years. Often a tiff of the most puerile kind will become a family
matter and end in a regular hereditary feud. In the Nile Valley this
often leads to bloodshed. In the oases, however, the quarrel usually
takes the form of the two sides to abusing and telling lies about each
other behind their backs, wrangling whenever they chance to meet,
and endeavouring at every possible opportunity to subject their
opponent to an ayb (insult, slight, snub) often of a most elaborate
description.
Shortly before my arrival the ’omda, getting sick of the squabble,
or finding that the mamur was making things too unpleasant for him,
had held out the olive branch by sending him a basket of early
mulberries—a fruit much appreciated in the oasis. The mamur had
made this an opportunity to humiliate his opponent. He had thrown
the fruit out of his window into the square in front of the mosque,
where all the inhabitants had seen it. It was generally considered
that he had scored heavily by doing so, and that this was one of the
best aybs that had been seen for years. The whole oasis had been
talking about it.
The partisans of the ’omda were consequently much
discomforted; but endeavoured to cover up their defeat by explaining
that it hadn’t really been a good ayb—the mamur had not thrown the
whole of the mulberries away, as he had stated, but had taken out all
the best ones and had only thrown away the rotten ones out of his
window; so as an ayb it didn’t count at all.
The ill-feeling between these two at length rose to such a pitch
that some of the leading men in the oasis decided to try and effect a
reconciliation between them, and a ceremony known as “making the
peace” took place.
The two opponents were invited to meet together in the presence
of some of their friends, who had argued with them, and at length the
quarrel had been patched up. They had then fallen on each other’s
necks and embraced and had agreed to feed together. They had
partaken of a huge feast in which whisky apparently played a
prominent part, and had both got drunk and started quarrelling
furiously again, in their cups. The next morning, when they were both
probably feeling rather cheap, the peace-makers had got to work
again and explained to them that they had not played the game, and
again a reconciliation had been effected; but there was still a good
deal of latent ill-feeling between them which vented itself mostly in
backbiting, under a show of friendship.
CHAPTER IV
B Y Qway’s advice I started feeding my camels on bersim,

preparatory to our journey into the dunes. There are two kinds of
bersim grown in the oasis: bersim beladi[1] and bersim hajazi.[2]
Bersim hajazi, however, should not be fed to camels in its green
state, as it very frequently causes them to get hoven.
The bersim was bought off the natives by the kantar, of a hundred
Egyptian pounds. At first there was some difficulty in getting it
weighed. Abd er Rahman, however, proved equal to the emergency.
He discovered a rock, which was supposed to weigh a kantar, and
which was the standard weight for the whole oasis. He then rigged
up a pair of scales, consisting of two baskets fixed to either end of a
beam, suspended from a second beam.
In the evening of the first day I spent in Mut I climbed to the top of
a low hill close to the town to look at the dune field that I hoped to
cross. A more depressing sight it would be impossible to imagine.
Not only were the sand hills in the neighbourhood of the town much
higher than those we had encountered on leaving Kharga Oasis, but
they extended as far as it was possible to see to the horizon, and
obviously became considerably larger in the far distance, where they
were evidently of great height.
I returned to my rooms with the gloomiest forebodings, wishing I
had never been such a fool as to tackle the belad esh Shaytan, or
“Satan’s country,” as the natives call this part of the desert, and
wondering whether, when I attempted to cross those dunes, I should
not end, after a few hours’ journey, in having to return completely
beaten with my tail tightly tucked between my legs, to the Nile Valley.
I lay awake for most of the night in consequence.
But daylight as usual made things look more cheerful. Anyway I
could have a shot at it, and as my camels did not seem to be in very
good order I decided to give them a rest and to feed them up into the
best possible condition, before subjecting them to what appeared to
be an almost impossible task. In the meantime I thought I might as
well see something of the oasis, and at the same time collect what
information I could about the desert.
So a few days after my arrival at Mut I set off with the mamur, the
policeman and the doctor to stay for a night with the ’omda of
Rashida, leaving the caravan behind me.
For the first two hours after leaving Mut, till we reached the village
of Qalamun, our road lay over a barren country largely covered with
loose sand, which proved to be rather heavy going.
Qalamun is rather a picturesque village, and seems to have been
built with an eye to defence. A great deal of land in the
neighbourhood is covered with drift sand, which in places seems to
be encroaching on to the cultivation, though not to be doing any
serious damage. An unusually large proportion of land in the
neighbourhood is planted with date palms, and, as the water supply
seems to be fairly abundant, the place has a prosperous well-to-do
air. In some cases the wells appear to be failing, as a few shadufs for
raising the water were to be seen. These and a few Dom palms gave
the neighbourhood a rather distinctive appearance. Of course we
visited the ’omda. The sheykhs of this village—the Shurbujis by
name—claim to have governed the oasis ever since the time of the
Sultan Selim, “The Grim.”
On leaving Qalaman we made straight for Rashida, most of our
road lying through cultivated fields, planted mainly with cereals.
Before reaching the village, we passed a large dead tree—a sunt, or
acacia, apparently—which is known as the “tree of Sheykh Adam,”
and is supposed to possess a soul. The wood is reported to be
uninflammable.
Shortly before reaching Rashida, we were met by the ’omda and
some of his family, who had ridden out to meet us, all splendidly
mounted on Syrian horses, gorgeously caparisoned with richly
embroidered saddles and saddle cloths. These joined on to our party
and rode back with us to Rashida.
Kharashef. Sand Grooved Ridge.
The wind driven sand grooves away the
rock, sometimes leaving large ridges
standing above its surface. (p. 308).
In Old Mut.
This shows the fortified character of the houses formerly built in the oases of the
Libyan desert as a defence against raids. (p. 41).
The village is one of the prettiest and most fertile in the oasis. It is
built on a low ridge lying at the south-east corner of a very extensive
grove of palms, in whose shade were planted great numbers of fruit
trees: figs, mulberries, apricots, oranges, tangerines—known in
Egypt under the curious name of Yussef effendi, i.e. Mr. Joseph—
bananas, almonds, pomegranates, limes, lemons, olives and sweet
lemons, the last bearing a large, tasteless, but very juicy fruit,
something like a citron in appearance.
THE TREE WITH A SOUL, RASHIDA.
The village lies close to the cliff. The interior of the village was of
the normal type, and, beyond presenting an unusually prosperous
appearance and having the walls of some of its houses painted on
the outside in geometrical patterns, usually in red and white, did not
differ from the other villages in the oasis.
The ’omda’s house was delightfully situated, with palm trees
growing almost up to the walls. He took us up into his guest
chamber, a long narrow room neatly whitewashed and furnished
almost entirely in the European manner, with deck-chairs, sofas
round the walls, a large gilt hanging lamp, bent wood chairs and
three-legged tables. The windows were draped with European
curtains and the floor covered with Eastern rugs and carpets. A large
mirror in a gilt frame and an oleograph portrait of the Khedive
completed the list of furniture.
On entering the room one’s eye was at once caught by the words
“Ahlan wa Sahlen”—welcome—painted on the opposite wall. And
welcome that hospitable ’omda certainly made us. The windows had
been kept closely shuttered all the morning to keep out the heat and
the flies; but these were opened on our arrival. Then the ’omda
entered and proceeded to spray the room and its inmates with scent.
Shortly afterwards the inevitable tea and cigarettes made their
appearance.
After compliments, enquiries as to the health of all parties present
and the usual polite preliminaries had been got through—a process
that took some minutes—the conversation turned upon horses. Only
a few of the richer natives of the oases are able to afford them, and
the remainder, when they do not walk, ride on donkeys. Powerful
quarters, round cannon bones and a small head, with an especially
small muzzle and widely distended nostrils, seemed to be the points
they valued most.
After luncheon, when the heat of the day was past, we were taken
by the ’omda to see some of the sights of the village. First we were
led to a big mud ruin known as the ’Der abu Madi. He told us he had
dug up a number of mummies about a mile to the north of the village,
which he said had been buried in earthenware coffins. Fragments of
one of these coffins that he produced showed that they must have
been about three inches thick and had evidently been baked in a
kiln. Many of the mummies had been wrapped round with a cloth of
some sort, with their arms lying straight along their sides, and had
then been wound tightly round with a rope. The remains of one of
them was shown us. It was, however, entirely knocked to pieces, as
the ’omda and his family had stuck it upright on the ground and then
amused themselves by turning it into an “Aunt Sally.” One or two
coins and the skull of a gazelle had been dug up from one of the
graves. The coins unfortunately were so worn and decayed that they
could not be recognised. There seems to be plenty of work for an
archæologist in Dakhla—and still more for an inspector of antiquities.
We were next taken off to see the great sight of Rashida—the Bir
Magnun, or “foolish well.” When this well was being sunk about forty
years ago the labourers stopped working for the day, not knowing
that they had almost reached the water-bearing stratum, with the
result that the water forced its way through the small distance from
the bottom of the bore hole to the top of the water reservoir, and
gushed up with such violence that it forced the tubing, above the
bore hole, partly out of the ground and flooded the whole country
round.
On first arriving in the oases, I made enquiries on all sides from
the natives for information as to what wells, roads or oases were to
be found in the unknown parts of the desert, beyond the Senussi
frontier. For a long time I could extract no information from any of
them, and it was not till I got to Rashida, and happened to ask the
’omda whether he knew anything about the oasis of Zerzura, that I
got any information at all. There is no stopping a native of Dakhla
when he gets on that subject, and one begins dimly to realise how
very little the East has changed since the days when the “Arabian
Nights” were written.
Many of the wealthier natives of the oases, and also, I believe, of
the Nile Valley, spend an appreciable portion of their time in hunting
for buried treasure. The pursuit is an absorbing one, to which even
Europeans at times fall victims. Curious as it may seem at first sight,
the native efforts are not infrequently attended with some success.
The reason is not far to seek. In former days, when the country
was ruled by a lot of corrupt Turkish officials, a native, who was
known to be possessed of any wealth, at once became the object of
their extortionate attentions. He consequently took every precaution
to hide his riches from these rapacious officials. The plan which he
very often adopted was to bury his valuables in the ground. Not
infrequently he must have died without imparting to his relations the
whereabouts of his cache. The treasure buried in this way in Egypt
would probably amount to an enormous sum in the aggregate, if it
could only be located.
Then, too, the sites of old Roman settlements are to be found all
over Egypt. The careless way in which the Romans seem to have
scattered their petty cash about the streets of their towns is simply
amazing. You can hardly dig for an hour in any old Roman site
without coming across an old copper coin or two.
Let a native find a few coins in this way, and he will spend weeks,
when no one is looking, in prowling around the neighbourhood in the
hopes of finding more. Should he be lucky enough to find an
earthenware pot containing a handful or two of old coins hidden in
the past from a Turkish pasha, it is pretty certain that he will become
a confirmed fortune-hunter for the remainder of his life. There is no
doubt that quite considerable sums—several pounds’ worth at a time
—are occasionally found in this way. The natives are extraordinarily
secretive about this kind of thing, and have been so long under a
corrupt Government that they can hold their own counsel far better
than any white man—for even now in out-of-the-way districts such as
the oases, where the English inspectors cannot properly supervise
the native officials, the extortionate ruler is at times most
unpleasantly en evidence.
In their hunts for buried riches the natives are frequently guided
by old “books of treasure.” Every self-respecting native, who is
wealthy enough to procure one, possesses at least one copy.
Before leaving Kharga I was fortunate in meeting E. A. Johnson
Pasha, so well known as the translator of the whole of Omar
Khayyám’s “Rubaiyat” into English verse—Fitzgerald, of course, only
translated a portion of it. He was the proud possessor of the only
complete copy known to exist of a book of this description, dating
from the fifteenth century.
One of the problems of the Libyan Desert, beyond the western
frontier of Egypt, is that of the oasis known as Zerzura, or “The oasis
of the Blacks.” It was, I believe, first heard of by Rohlfs, who, in his
attempt to go westward from Dakhla Oasis to Kufara, found the sand
dunes impassable for his big caravan, and so had to turn up to the
north and make for the oasis of Siwa instead. During this journey he
encountered three blacks, who said that they were escaped slaves
from the oasis of Zerzura, a place that they described as being some
distance to the west of his route.
On mentioning this place to Johnson Pasha, he told me of this old
book, and said that it contained a description of the road to this
oasis, and of what might be found there by anyone who was
fortunate enough to reach it. His book also described the road to the
mines of King Cambyses.
He very kindly gave me a translation of the portions of this queer
old volume that related to these two places. There were two
descriptions of the road to Zerzura in a section of the book headed
“In the Oases” They ran as follows:—
“Go to the Der el Banat (the girls’ convent), near it you will find a
hollow place, three mastabas (platforms), a round hill and three red
stones. Burn incense here.” Then follow two lines of cipher writing
and cabalistic signs, which presumably give instructions for following
the road, and the description ends.
The second reference was much more to the point. It was as
follows: “Account of a city and the road to it, which lies east of the
Qala’a es Suri, where you will find palms and vines and flowing
wells. Follow the valley till you meet another valley opening to the
west between two hills. In it you will find a road. Follow it. It will lead
you to the City of Zerzura. You will find its gate closed. It is a white
city, like a dove. By the gate you will find a bird sculptured. Stretch
up your hand to its beak and take from it a key. Open the gate with it
and enter the city. You will find much wealth and the king and queen
in their palace sleeping the sleep of enchantment. Do not go near
them. Take the treasure and that is all.”
The book also contained two separate directions for finding the
mines of King Cambyses. One of them instructed the reader thus:
“Go to the Der el ’Ain, west of Esna, where there is a medical spring,
and go north from the Der and the well five farasangs, which make a
barid and a quarter, to where there is a red hill with a beacon mark
on the top of it. You are to go up and look towards the east. You will
see a pillar divided into two halves. Dig there.” Then the aggravating
book—just when it comes to giving the final definite directions for
finding the mines, breaks off into line upon line of cabalistic signs, as
it did in the case of Zerzura.
The second instruction for finding the mines, however, is much
more explicit, and goes into minute details of the road to be followed,
so much so that it would appear to be impossible for anyone to miss
it.
It runs as follows: “By the town of Esna, north of Edfu. If you go
there seek the mines of King Kambisoos (Cambyses). Ask for the
Holy Der, which is called ‘———’ but to give away directions for
finding such wealth of treasure would be foolish. King Cambyses
was a son of Cyrus the Great—the Conqueror of Babylon—and ruled
over the Medes and Persians when the Persian Empire was at about
its height. He was a real big King, and the much-vaunted mines of
King Solomon—a mere petty Sultan by comparison—probably bear
about the same relation to those of King Cambyses as a threepenny-
bit bears to the present National Debt. The mere description of them
in Johnson Pasha’s book of treasure makes one’s mouth water.”
First the directions lead you—in the clearest possible way—to a
valley called the Wady el Muluk (the valley of the kings). Here you
find the crucibles and all the apparatus and tools necessary for
smelting, merely waiting to be used. You go a little farther on and
you come to the “high class mine”—and very high class it is. You
have only to dig half a cubit deep into it and you come at once on to
a mineral “like yellow earth in stony ground.” First you find it in lumps
the size of beans, which “is sent by Allah,” and you are directed to
take “His good fortune.” Then if you dig deeper, you will find it in
lumps the size of melons. This you are explicitly told “is gold of
Egypt. There is none better”—a statement it would be rash to
contradict.
Having dealt with these particular mines, the old Arab astrologer
directs his son, to whom the book is addressed, to go on to where
two great rocks stand up, with a hollow before them, stating that in
the hollow will be found “a black earth with green veins like silver
rust,” and directs him to take it. It is “sent by Allah.” Unfortunately he
omits to mention the nature of this mysterious mineral.
He then directs his son to “go with the blessing of Allah” to
another place, where he states “You will find, oh! my son, before you
a high hill in which they used to get the peridots.” Next he tells him
how to go on to the “Emerald pits which are three in number,” and
after that, directs him to the “Copper mine which is in a cave closed
by a door,” adding that the copper ore is “green earth very like green
ginger and having veins in it like blood.”
With the dazzling prospect of acquiring such untold wealth as that
to be obtained in the very “high class mines,” described in books of

Ebook A Literary Translation in The Making A Process Oriented Perspective 1St Edition Claudine Borg Online PDF All Chapter

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Ebook A Literary Translation in The Making A Process Oriented Perspective 1St Edition Claudine Borg Online PDF All Chapter

Uploaded by

Copyright:

Available Formats

A Literary Translation in the Making A

Process Oriented Perspective 1st

Translation of Contemporary Taiwan Literature in a

The Object Oriented Thought Process 5th Edition Matt

Understanding Korean Film A Cross Cultural Perspective

Advances in Corpus Applications in Literary and

German Philosophy in English Translation Postwar

The Routledge Handbook of Latin American Literary

Retracing the History of Literary Translation in

Making a Mark Image and Process in Neolithic Britain

A Literary Translation in the Making

This book presents a holistic picture of the practice of an experienced lit-

Routledge Studies in Literary Translation

Routledge Studies in Literary Translation highlights pioneering research

Translating Transgender Identity

The Afterlife of Dante’s Vita Nova in the Anglophone World

A Literary Translation in the Making

For more information about the series, please visit: https://​www.routle​dge.

First published 2023

List of Figures vii

1 Theoretical Framework: Venturing beyond classic

3 Methods and Data 46

5 From a Handwritten First Draft to the Published

6 Decision-​Making and Underlying Motivations 151

Appendix 1: Evolution of the Translation –​

1.1 Decisions and choices in the translation process 10

3.1 Methods used to gather data on the different phases of the

ATS alternative translation solution

The glamour of the finished translation hides the depths of a text’s

Research questions and structure of the book

i) How does a literary translation come into being?

The structure of the book is as follows. Chapter 1 provides a theoretical

This chapter outlines the concepts, definitions and models underpinning

1.1 Cognitive Translation Studies: An overview

1.2 The three phases of the translation process

1.3 Decisions in the translation process

• involves extensive and continuous decision-

• involves making choices

• deciding between alternatives

Figure 1.1 Decisions and choices in the translation process.

1.3.2 The structure of decisions

Identification of the problem

Description of the problem

Thinking of a way forward

Figure 1.2 My rearrangement of Wilss’s six-​stage decision-​making process.

1.3.3 Conscious and unconscious decisions

In the translation process of any individual, there are segments which

Translators thus make use of both deliberate and intuitive decision-​

1.4 Alternative translation solutions, postponed

Discussions dealing with the presence of several ATSs in the written

1.4.1 Choosing between ATSs

1.4.2 Written ATSs: Empirical evidence from the literature

1.5 Revision in CTS

Extended Translation Process

Translation process Post-translation

Post- Other- Publication

Figure 1.3 The extended translation process.

1.5.2.1 How is self-​revision defined in the literature?

1.5.2.2 Empirical studies of self-​revision

1.5.2.3 The effect of self-​revision on the translation process:

1.5.3 The post-​drafting phase

1.5.4 The post-​translation process and other-​revision

there is no way of knowing how many different persons were actu-

For more information about the series, please visit: https://www.routledge.

6 Decision-Making and Underlying Motivations 151

Appendix 1: Evolution of the Translation –

Figure 1.1 Decisions and choices in the translation process.

Figure 1.2 My rearrangement of Wilss’s six-stage decision-making process.

Translators thus make use of both deliberate and intuitive decision-

Figure 1.3 The extended translation process.

1.5.2.1 How is self-revision defined in the literature?

1.5.2.2 Empirical studies of self-revision

1.5.2.3 The effect of self-revision on the translation process:

1.5.3 The post-drafting phase

1.5.4 The post-translation process and other-revision

would illustrate specific parts of the “coming-into-being” of a trans-