Professional Documents
Culture Documents
Future Development
Xiao Kairong
College of International Studies, Southwest University
kairongxiao@163.com
1
Introduction
A balanced development of translation studies as a discipline entails effective researches
in all the three branches of Descriptive Translation Studies mapped by James Holmes (1972),
namely the product-, function- and process-oriented studies. With more research findings in
the first two fields made in the latter part of the 20th century, the process-oriented research
acquires promising momentum in the first decade of the 21st century, as it borrows the latest
theoretical models and research methods from the neighboring cognitive disciplines, such as
cognitive psychology, cognitive linguistics, neurophysiology and computational linguistics.
New methods of gathering multi-media data about translator behaviour and large-scale joint
research projects have emerged with findings about the underlying nature of translation as a
cognitive activity. These findings are unimaginable previously with the traditional research
methods.
The exciting development in the process-oriented research is leading some researchers
to propose the emergence of a cognitive translation studies. However, before the favoured
new branch is established, some problems remain to be solved and some theoretical and
methodological issues remain to be considered. This present paper attempts to sort out the
problems and challenges in the cognitive research of translation process based on the review
of the present achievements in the empirical researches of translation so as to suggest the
potential direction of development in cognitive translation studies.
2 Theoretical models of translation process
Several theoretical models have been put forward by researchers to offer accounts of the
metal processes carried out by the translators or interpreters. Six of the most representative
models are classified into three groups according to their theoretic basis: the Interpretive
Theory of Translation, Psycholinguistic and Cognitive Psychology Models, and Cognitive
Pragmatic Model.
2.1 Interpretive Theory of Translation
As one of the earliest attempts to the account of the translation process, Seleskovitch
(1968) and Lederer (1981) were pioneering researchers in taking a cognitive approach to
translation process from the 1960s through the 1980s. They proposed the Interpretive Theory
of Translation (ITT) to identify three interrelated phases of translation/interpreting process:
understanding, deverbalization and re-expression. Understanding is the process of
generating sense, involving not only the linguistic knowledge, but also some other cognitive
inputs including encyclopedic knowledge and contextual knowledge. Memory plays an
important role with the immediate memory to store words for a short time and cognitive
memory to store the whole range of knowledge. The end product of understanding is
non-verbal synthesis. ITT postulates the existence of an intermediate phase of
deverbalization between understanding and re-expression. Re-expression is based on the
deverbalized sense, or the non-verbal synthesis rather than the linguistic form. It involves
also the work of both linguistic and non-linguistic knowledge, and is similar to monolingual
communication in that the intended meaning is expressed in the target language.
1
methods can be labeled as either soft as they are qualitative and subjective in nature, or
hard as they gather quantitative and objective data.
3.1.1 Soft methods
The so-called soft methods are mostly verbal reports, including TAPs, dialogue protocols,
retrospection, Integrated Problems and Decision Reporting (IPDR). These methods are often
combined with questionnaires (e.g. Yousself 1989), interviews (e.g. Shih 2006) and
translation journals or diaries (e.g. Bergen 2006; Fox 2000) to elicit informants verbal
responses or accounts. In terms of the time of eliciting the data, the verbal reporting methods
are distinguished between concurrent and retrospective reporting. The former takes place
simultaneously with the task performance, while the latter takes place after the task
performance. The time factor is now a great concern about the validity and reliability of the
date elicited in the experiment. Think-aloud was the first method used on the individual
translator to report their mental activities in translating, while retrospection has been
increasingly used, particularly together with other hard methods. Dialogue protocols began
to be used in the late 1980s on two or more translators working on the same text (e.g. House
1988; Kumaul 1995). Although dialogue protocols are reported to be superior to think-aloud
as the former yields richer data, the question whether the negotiating and debating between
the partner translators have any effect on the efficiency of the translation has been a concern.
Another frequently used verbal report is IPDR, which requires subjects to write down
comments on the problems they encounter during a translation task and how they have
solved them. It is more subjective as the subjects may write down what they consider to be
most important or relevant, and more often used for didactic purposes. Verbal reports include
also some methods borrowed from sociology or psychology, such as questionnaires,
interviews and translation journals or diaries.
3.1.2 Hard methods
The so-called hard methods emerge with the combination of the latest advancement of
technologies and development of new software. They are keystroke logging with Translog,
video and screen recording with Proxy and Camtasia, and eye-tracking with relevant
software.
In the late 1990s, the TRAP group employed Translog to record the keyboard and mouse
activities during a translation process. This marks the early introduction of keystroke logging
into the research of translation process (see Jackobsen 1998). One interest with the
keystroke logging research is in pauses, which have been seen as the indicators of cognitive
processes, especially planning processes. The length of pauses and the number of pauses
may indicate the complexity of the linguistic units that are processed in the pauses. Keystroke
logging data are often complemented by think-aloud or retrospective reports rather than
concurrent verbalization as the latter may impose interference on the translator, most
probably on the pause.
Video and screen recordings are used for their advantage of showing the subjects
actions, faces and the electronic sources or web-sites they are using in translation. These
data offer richer aspects of the translator behaviours and the information sources elicited with
the help of think-aloud and retrospection to provide account of the problems they encounter
and strategies they adopt to solve them.
Eye-tracking is a fairly recent method introduced in translation studies (see OBrien 2005;
Dragsted & Hansen 2007). Its basic hypothesis is that eye movements and pupil dilation
correlate with the perceptual and cognitive processes going on as the subjects doing
translation on the computer. Relevant researches have shown, for example, that the pupil
4
dilation creases with cognitive load (Gpferich 2008, 56). Therefore eye-tracking is used to
elicit data about the translators eye movements, which offer insight into the translators
mental activities. Eye-tracking is preferred by researchers as it is found to increase the
ecological validity for the on-intrusive feature of the equipment available today (Gpferich &
Jskelinen 2009, 173).
Apart from the use of these technologies and software, researches also introduce
methods from the neurosciences, such as EEG, and neuro-imaging techniques, such as
functional magnetic resonance imaging (fMRT) and positron emission tomography (PET).
Another trend of methodological innovation is found in the change from the use of single
method in a research to the integration of more methods, as each of the above-mentioned
methods has its own advantages, disadvantages and specific focus or emphasis. Recently,
triangulation has been gradually adopted by more researchers to elicit data of different types
and aspects so as to improve the quality, validity and reliability of the research findings (Alves
2003). Triangulation is a term that refers to the use of two or more than two methods in a
single study. According to Shreve & Angelone (2010, 6), the research of the past decade
shows an increasing reliance on multiple methodologies, and the majority of the empirical
studies in Translation and Cognition edited by him make use of one or more of the new
methods. Angelone (2010), for example, uses screen recording and think-aloud in the
research of the problem-solving behavior of professional and student translators. Dragsted
(2010) employs keystroke logging and eye-tracking to explore source text comprehension
and target text production. Fabio Alves and his colleagues are more innovative in integrating
process-based and product-based research methods, namely the combination of corpus
analysis and keystroke logging, eye-tracking and retrospection to the research translation
unit associated with cognitive effort during a translation task (Alves, et al. 2010).
3.2 Increasing interdisciplinarity
Research of translation process has enjoyed a close relationship with the neighbouring
disciplines such as psychology, linguistics, and psycholinguistics, in both model building and
methodology borrowing. The models discussed in part 2 have introduced such notions as
memory, information processing, relevance, effort, etc. from psychology and other cognitive
sciences. The methodologies in the previous part are also mostly borrowed from the
neighbouring cognitive sciences.
The integration of translation studies and cognitive sciences is increasingly remarkable
in the past decade. One chapter of the book Translation and Cognition (Shreve and Angelone
2010) is given to the discussion of interdisciplinary approaches, with the title Integration of
translation process research and the cognitive sciences. These interdisciplinary researches
include the expert-performance perspective exploration of expertise in interpreting, the
search of neuro-physiological correlates of expertise in interpretingand the study of the
prompting cognates in the bilingual lexicon. These research findings are the strong proof of
the productive forces of the sister disciplines, such as psycholinguistics, cognitive linguistics,
cognitive psychology, neuro-physiology when they are integrated with translation process
research. The interdisciplinary approaches have contributed to the research of translation
process some inspiring notions and concepts, namely expertise, bilingual ability, conceptual
representation, prompting cognates, metacognition, etc. Meanwhile, some traditional
concepts such as competence, translation unit and strategies have been redefined and
re-examined to their new vitality.
Another indicator of the interdisciplinary trend of translation process research is seen in
the fact that an increasing number of researchers from neighbouring subject fields extend
5
their interest into the field of translation studies. Among the 22 contributors to the book
Translation and Cognition, 7 of them are based in other cognitive sciences, e.g. psychology,
neuro-physiology, cognitive linguistics, and computational linguistics. They either cooperate
with translation researchers, or adopt translation process researches with the models and
methods in their own fields to probe into the mental mechanism of this particular bilingual
activity translation.
3.3 Large-scale collaboration in research design
The translation process researches in earlier years were characterized by a relative slow
development, isolatedness in the research design, and a small number of subjects confined
mostly to translation students or trainees rather than professional translators. Lack of more
scientific research methods and data acquiring measures are the main reasons for the
limitations of the process researches, which lead to the less convincing research findings. In
recent years, especially in the first decade of the 21st century, international collaboration
marks a new trend in establishing large-scale research projects. Researchers from different
countries and language background have formed a number of communities in carrying out
empirical investigation from different perspectives on subjects from different backgrounds.
These projects and communities include, among others, the TRAP and EYE-TO-IT at
Copenhagen Business School in Denmark, LETRA at the Universidade Federal de Minas
Gerais in Brezail, the PACTE group at the University of Barcelona and the PETRA project at
the University of Granada in Spain, the TransComp project at the University of Graz in
Australia, and the CTP project at the Zurich University of Applied Sciences in Swiss. Most of
these projects involve researches from more than one country, for example the EYE-TO-IT
project has the researchers from Australia, Bulgaria, Denmark, Finland, Norway and Sweden.
These projects carry out process researches of the both translation students and professional
translators with a wider range and bigger number of subjects, better control of variables and
more language pairs involved.
These large-scale collaborative research communities and projects have also yielded
valuable research findings, including Hansen (1999), Tirkkonen-Condit and Jskelinen
(2000), Alves (2003), Gpferich et al.(2008), Shreve & Angelone (2010), and OBrien (2011).
Research findings also appeared in special issues of influential journals, including a 2005
special edition of Meta with the title of Processes and Pathways in Translation and
Interpreting and a 2009 special edition of Across Languages and Cultures entitled Process
Research into Translation Competence. These research fruits have broadened our
theoretical and methodological horizons in translation process research. They show us not
only what is potential in human mind, but also how we can go about exploring the mind to
examine its contents. They represent the recent progress and latest trends in research design,
methods and increasing interdisciplinarity in cognitive translation process research (Shreve &
Angelone 2010, 2).
4 Challenges for cognitive approaches to translation process
Despite the remarkable progress in data acquiring methods, research design, variable
control, and project size in translation process research, the cognitive approaches to
translation process as a relatively new paradigm are facing some challenges and potential
problems are expected to emerge. A clear understanding of the challenges will give us a
picture of what might be the future direction to develop the cognitive translation studies as
theoretic framework. The challenges can be summarized as building of theoretic models,
evaluation of methodologies and establishment of data documentation systems.
6
think-aloud on the translating strategies, for example fewer formal correspondence at the
lexical level (Jskelinen 2000, 79) or increase of revisions in verbal reporting experiments
(Krings 2001, 229). The methods of translation journals or diaries might change the
translation process itself as the subjects would be more aware of the problem-solving
processes. Therefore, such questions remain to be answered as whether these methods will
change the research object, or whether the data reflect the object of the research accurately,
or to what degree will the methods interfere with the translation process. A general concern
with the methodological evaluation is whether these effects and interferences change
remarkably the translation product and the way the translators translate.
Evaluation of methodologies is to be made on a contrastive basis, or to compare different
methods for the choice of the most appropriate one in a specific research design. Another
benefit of the methodological evaluation is the combination of more methods in a single study.
Triangulation has already been adopted by some researchers, but there is still the question of
the number and nature of methods to be combined to produce what effect.
4.3 Data documentation
The use of new methods and technologies has yielded massive amounts of multi-media
data, including voices, videos, images, and figures about translator behaviour. Some
large-scale internationally collaborative projects, such as EYE-TO-IT, have acquired
experimental data covering multiple language pairs, a variety of cultural contexts and
translators from very different backgrounds. Some researchers also employ contrastive and
comparable corpora to verify the empirical findings. A challenge is how to go about analyzing
and interpreting these massive amounts of data available now through the use of new
methods (Shreve & Angelone 2010, 11-12). If these massive data could be stored, retrievable
and shared by other researchers, the empirical research will be of better value.
In view of this, some researchers (e.g. Gpferich & Jskelinen 2009, 185) suggest to
set up an Internet portal that provides access to a data archive so that all researchers could
share. In this archive, all data could be retrieved with the criteria by which they are stored.
The Internet portal could also provide access for the researchers to upload their own data to
be shared by others through the archive to carry out contrastive studies between different
groups of subjects. In view of this idea, it is of great concern to establish a translation process
research data archive, or the asset management systems (AMS) for storing, archiving,
annotating, and analyzing digital resources of any type, such as texts, graphics, videos, and
sound files. Gpferich and his colleagues in the TransComp Group have begun such a job
(Gpferich & Jskelinen 2009, 186). They made all the materials used in TransComp
available to the scientific community in the asset management system, such as the source
texts, the translation assignments, model translations, the questionnaires used, and all data
obtained in the experiments, such as the TAPs, the log files, and the screen recordings
(see http://gams.uni-graz.at/container:tc). Such Internet-based data storage and sharing
system is just beginning to appear as a special case. With the expansion of translation
process research projects, deepening of the research topics, this type of data storage and
sharing system on a global basis will be a welcome trend in the near future.
5 Conclusion
Different from the time when the imaginary theoretical thinking about translation process
prevailed, the first decade of the 21st century saw remarkable progress in innovative data
eliciting methods, increasing interdisciplinarity in research approaches and formation of
large-scale collaborative research communities. These exciting progress has led some
scholars (e.g. Halverson 2010; Muoz 2010) to argue in favour of establishing a new
8
References
Alves, Fabio. 2003.Triangulating Translation: Perspectives in Process-Oriented Research.
Amsterdam & Philadelphia: John Benjamins.
Alves, Fabio, Adriana Pagano, Stella Neumann, Erich Steiner & Silvia Hansen-Schirra. 2010.
Translation Units and Grammatical Shifts: Towards an Integration of Product- and
Process-based Translation Research. In Translation and Cognition, edited by Gregory
Shreve & Erik Angelone, 109-142. Amsterdam & Philadelphia: John Benjamins.
Bell, Roger. 1991. Translation and Translating: Theory and Practice. London and New York:
Longman.
Bergen, D. 2006. Learner Strategies and Learner Autonomy in Translator Training. In
Translation and Interpretation Training and Research, edited by Tommola, J. & Y.
Gambier. 119-126. Turku: University of Turku.
Dragsted, Babara. and Gorm Hansen. 2007. Speaking Your Translation: Exploiting
Synergies Between Translation and Interpreting. In Interpreting Studies and Beyond: A
Tribute to Miriam Shlesinger, edited by Pchhacker, F., A. L. Jakobsen & I. Mees,
251-274. Copenhagen: Samfundslitteratur.
Dragsted, Babara. 2010. Coordination of Reading and Writing Processes in Translation: An
Eye on Uncharted Territory. In Translation and Cognition, edited by Gregory Shreve &
Erik Angelone, 41-62. Amsterdam & Philadelphia: John Benjamins.
Ericsson, Anders & H. Simon. 1980. Verbal Reports as Data. Psychological Review. 3:
215-251.
Fox, O. 2000. The Use of Translation Diaries in a Process-oriented Translation Teaching
Methodology. In Developing Translation Competence, edited by Schffner, C. & B. Adab,
115-352. Amsterdam and Philadelphia: John Benjamins.
Gile, Daniel. 1995. Basic Concepts and Models for Interpreter and Translator Training.
Amsterdam and Philadelphia: John Benjamins.
Gpferich, Susanne. 2008. Translationsprozessforschung: Stand-Methoden-Perspektiven.
Tbingen: Narr, 2008.
Gpferich, Susanne, Arnt Lykke. Jakobsen, And I. M. Mees. eds. 2008. Looking at Eyes: Eye
Tracking Studies of Reading and Translation Processing. Copenhagen:
Sammfundslitteratur.
Gpferich, Susanne & Riitta Jskelinen. 2009. Process Research into the Development of
Translation Competence: Where Are We, and Where Do We Need to Go?. Across
Languages and Cultures, 10: 169-191.
Gutt, Ernst-August. 1991. Translation and Relevance: Cognition and Context. Oxford: Basil
Blackwell.
Halverson, Sandra. 2010. Cognitive Translation Studies: Developments in Theory and
Method. In Translation and Cognition, edited by Gregory Shreve & Erik Angelone,
349-369. Amsterdam & Philadelphia: John Benjamins.
9
Hansen, Gyde. ed. 1999. Probing the Process in Translation: Methods and Results.
Copenhagen: Sammfundslitteratur.
Holmes, James. 1972. The Name and Nature of Translation Studies Unpublished
manuscripts, Amsterdam: Translation Studies Section, Department of General Studies,
reprinted in Laurence Venuti. ed. 2000. The Translation Studies Reader, 172185.
London and New York: Routledge.
House, Julianne. 1988. Talking to Oneself or Thinking with Others? On Using Different
Thinking-aloud Methods in Translation. Fremdsprachen lehren und lernen. 17: 84-99.
Jakobsen, Arnt Lykke. 1998. Logging Time Delay in Translation. In LSP Texts and the
Process of Translation, edited by Gyde Hansen, 173-101. Copenhagen: Copenhagen
Business School.
Jskelinen, Riitta. 2000. Focus on Methodology in Think-aloud Studies on Translating. In
Tapping and Mapping the Processes of Translation and Interpreting. Outlooks of
Empirical Research, edited by Riitta Jskelinen & Son Tirkkonen-Condit, 71-82.
Amsterdam & Philadelphia: John Benjamins.
Kiraly, Donald. 1995. Pathways to Translation: Pedagogy and Process. Kent: Kent State
University Press.
Krings, Hans-Peter. 2001. Repairing Texts: Empirical Investigations of Machine Translation
Post-editing Processes. Kent: Kent State University Press.
Kumaul, P. 1995. Training the Translator. Amsterdam and Philadelphia: John Benjamins.
Lederer, M. 1981. La Traduction Simultane. Paris: Minard.
Muoz Martin, Ricardo. 2010. On Paradigm and Cognitive Translatology. In Translation and
Cognition, edited by Gregory Shreve & Erik Angelone, 169-187. Amsterdam &
Philadelphia: John Benjamins.
OBrien, Sharon. 2005. Methodologies for Measuring the Correlation between Post-editing
Effort and Machine Translatability. Machine Translation,1:37-58.
OBrien, Sharon. 2011. Cognitive Explorations of Translation: Eyes, Keys, Taps. London &
New York: Continuum.
Seleskovitch, D. 1968. Linterprte dans les confrences internatinales: problmes de
langage et de communication. Paris: Minard, 1968. trans. 1978. Interpreting for
International Conferences, Washington, DC: Pen and Booth.
Shih, C. Y. 2006. Revision from Translators Point of View. Target 2: 295-312.
Shreve, Gregory & Erik Angelone. 2010. Translation and Cognition: Recent Development.
In Translation and Cognition, edited by Gregory Shreve & Erik Angelone, 1-13.
Amsterdam & Philadelphia: John Benjamins.
Shreve, Gregory. & Erik Angelone. eds. 2010. Translation and Cognition. Amsterdam and
Philadelphia: John Benjamins.
Tirkkonen-Condit, Son and Riitta Jskelinen. eds. 2000.Tapping and Mapping the
Processes of Translation and Interpreting. Amsterdam and Philadelphia: John
Benjamins.
Sperber, D. & D. Wilson. 1986/1995. Relevance: Communication and Cognition. Oxford: Basil
Blackwell.
Wilss, Wolf. 1996. Knowledge and Skills in Translator Behaviour. Amsterdam and
Philadelphia: John Benjamins.
Youssef, A. F. 1989. Cognitive Processes in Written Translation. Unpublished PhD thesis.
Ann Arbor: University Microfilms International.
10