See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/304041473

Conference Interpreting

Article · December 2006

DOI: 10.1016/B0-08-044854-2/04285-1

CITATIONS READS

5 13,687

1 author:

Daniel Gile
Université de la Sorbonne Nouvelle Paris 3
161 PUBLICATIONS 3,966 CITATIONS

SEE PROFILE

All content following this page was uploaded by Daniel Gile on 02 October 2018.

The user has requested enhancement of the downloaded file.

Conference & Simultaneous Interpreting – 3278 words

Interpreting is the oral translation of oral discourse, as opposed to the oral translation of
written texts. The latter is known as sight translation or translation-at-sight.
Interpreting as an official or professional function seems to have been in existence
since very early times; some studies have indicated its use in Ancient Egypt (Kurz 1985).
Interpreters have played important roles in history, inter alia during exploration and invasion
campaigns: for instance when the Spaniards arrived in Central and South America (Kurz
1991). Recent interest in the field is associated with the emergence of specialized forms of
professional interpreting, such as business interpreting, conference interpreting, #COURT
INTERPRETING##, #COMMUNITY INTERPRETING## and #SIGNED LANGUAGE
INTERPRETING##. This entry makes particular reference to conference and simultaneous
interpreting.
1. Types and modes of interpreting
Conference interpreting was born during the first World War. Until then, important
international meetings were held in French, the international language at the time. During
World War I, some high-ranking American and British negotiators did not speak French,
which made it necessary to resort to interpreters (Herbert 1978). With the advent of
simultaneous interpreting, and especially after the Nuremberg trials (1945-46) and Tokyo
trials (1946-48), conference interpreting became more widespread. It is now used widely, not
only at international conferences but also in radio and TV programmes, various courses and
lectures, and during State visits, which makes the term `conference interpreting' a misnomer.
What now distinguishes conference interpreting from other forms of interpreting are its

modes (consecutive and simultaneous), and its high performance level.

Most conference interpreters only have two or three working languages, divided as
follows:

A language(s): the native tongue(s) of the interpreter or language(s) of which s/he has
native or near-native command. Interpreters work into as well as out of their A
language(s).

1
B language(s): non-native language(s) of which the interpreter has sufficient
command but not to the same level as an A language. Interpreters work into as
well as out of their B language(s).
C language(s): these are passive languages. Interpreters work from a C language into
their A or B language, but they do not interpret into a C language.
In consecutive interpreting, the interpreter listens to a speech segment of a few
minutes or so, takes notes, and then delivers the whole segment in the target language; then
the speaker resumes for a few minutes, the interpreter delivers the next segment, and the
process continues until the end of the speech. The `sentence-by-sentence' interpreting often
found in liaison and community interpreting is not regarded by conference interpreters as
`true consecutive'.
In simultaneous interpreting, the interpreter sits in an interpreting booth, listens to
the speaker through a headset and interprets into a microphone while listening. Delegates in
the conference room listen to the target-language version through a headset.
Simultaneous interpreting is also done by signed language interpreters (or interpreters
for the deaf) from a spoken into a signed language and vice versa. Signed language
interpreters do not sit in the booth; they stand in the conference room where they can see the
speaker and be seen by the other participants.
Whispered interpreting (or chuchotage) is a form of simultaneous interpreting in
which the interpreter does not sit in a booth but in the conference room, next to the delegate
who needs the interpreting, and whispers the target-language version of the speech in the
delegate's ears.
None of these modes of interpreting is restricted to the conference setting.
Simultaneous interpreting, for instance, has been used in large multilingual trials, and
whispered interpreting may be used in a business meeting.
2. Differences between translation and interpreting
While most scholars stress that translation and interpreting essentially fulfill the same
function, many - especially interpreters - consider that the two are very different, even
incompatible professions. This assertion, as well as alleged personality differences between

2
translators and interpreters (Henderson 1987), have not been clearly documented in the
literature. However, as regards actual translation and interpreting practice, some differences
are not controversial. The most obvious of these arise from the fact that translators deal with
written language and have time to polish their work, while interpreters deal with oral
language and have no time to refine their output. The implications are:
- Translators need to be familiar with the rules of written language and be competent writers
in the target language; interpreters need to master the features of oral language and be good
speakers, which includes using their voice effectively and developing a `microphone
personality'.
- Any supplementary knowledge, for example terminological or world knowledge, can be
acquired during written translation but has to be acquired prior to interpreting.
- Interpreters have to make decisions much faster than translators.
A subtler level of analysis of the skills required in translation and interpreting must await
advances in psycholinguistics and cognitive psychology. Unlike translation, interpreting
requires attention sharing and involves severe time constraints. Many recurrent interpreting
errors may well prove to be the result of either saturation in or improper management of the
interpreter's processing capacity (see below). A detailed discussion of differences and
similarities between translation and interpreting, and of their implications for training, can be
found in Gile (1995b). See also #PSYCHOLINGUISTIC/COGNITIVE APPROACHES##.
3. History of research on conference interpreting
Historically, research in conference interpreting can be broken down into four periods (Gile
1994): early writings, the experimental period, the practitioner's period and the renewal
period.
The early writings period covers the fifties and early sixties. During this period, some
interpreters and interpreting teachers in Geneva (Herbert 1952, Rozan 1956, Ilg 1959) and
Brussels (Van Hoof 1962) started thinking and writing about their profession. These were
intuitive and personal publications with practical didactic and professional aims, but they did
identify most of the fundamental issues that are still debated today. The first academic study
on interpreting, an introspective MA thesis by Eva Paneth, was defended at the University of

3
London in 1957.
During the experimental period (in the sixties and early seventies), a few
psychologists and psycholinguists such as Treisman, Oléron and Nanpon, Goldman-Eisler,
Gerver, and Barik (see Gerver 1976) became interested in interpreting. They undertook a
number of experimental studies on specific psychological and psycholinguistic aspects of
simultaneous interpreting and studied the effect on performance of variables such as source
language, speed of delivery, ear-voice span (i.e. the interval between the moment a piece of
information is perceived and the moment it is reformulated in the TL), noise, pauses in
speech delivery, etc. Practitioners rejected both the methods and the results of such studies.
During the practitioner's period, which started in the late sixties and continued into the
seventies and early eighties, interpreters, and especially interpreting teachers, began to
develop an interest in interpreting research and theory. The first Ph.D. dissertation on
interpreting by an interpreter was defended in Vienna by Ingrid Pinter (now Ingrid Kurz) in
1969. Numerous papers, as well as more than twenty MA theses and dissertations, were
subsequently written by practising interpreters. The main thrust came from Paris, but there
was also much activity in West Germany, East Germany, Switzerland and other European
countries, as well as in the USSR, Czechoslovakia and Japan. Most of the research was
speculative or theoretical rather than empirical, and most Western authors, except for a group
at ESIT (Ecole Supérieure d'Interprètes et de Traducteurs) in Paris, worked in relative
isolation. In particular, relations with the scientific community of linguists, psycholinguists
and cognitive psychologists were virtually non-existent, possibly more because of the
interpreters' defensive attitude than because of a lack of interest from non-interpreters (Gile
1995a, Gerver and Sinaiko 1978).
This period also saw the so-called théorie du sens (see #INTERPRETIVE
APPROACH##) become dominant. This `theory of sense' was not new (Pöchhacker
1992:22), but it was adopted in Paris and strongly promoted by ESIT during the seventies and
eighties. Its basic tenet is that translation and interpreting are based on meaning (le sens) as
opposed to language, that they proceed by `extracting' the meaning from the source text or
utterance, deliberately getting rid of the linguistic form of the original, and eventually

4
reproducing a target text or utterance on the basis of the `deverbalized message'. Proponents
of this theory assert that translation and interpreting are language-independent in that text
comprehension and production are spontaneous and automatic, whatever the languages
involved, provided the translator/interpreter has the necessary command of the source and
target languages and the relevant world knowledge.
The renewal period began in the mid-eighties and is still in evidence today.
Towards the mid-eighties, a new generation of practitioners began to question the idealized
view of interpreting postulated by the théorie du sens and to call for a more scientific study of
interpreting, giving rise to an interdisciplinary approach to the subject. During a seminar on
the teaching of translation and interpreting held by the University of Trieste (Italy) in
November 1986 (Gran and Dodds 1989), the prevailing dogma was challenged in favour of
the new paradigm (Moser-Mercer 1991). This new paradigm has since continued to gain
ground. Research is still largely undertaken by practising interpreters, but they increasingly
draw on findings and ideas from other disciplines, in particular neurolinguistics (see Lambert
& Moser-Mercer 1994). There are more and more empirical studies being conducted,
although their proportion remains very low if compared to the total number of publications on
interpreting. And finally, communication between researchers is improving, in particular
through the The Interpreter's Newsletter (published by the Scuola Superiore di Lingue
Moderne per Interpreti e Traduttori at the University of Trieste) and the Bulletin of IRTIN,
the international Interpreting Research and Theory Information Network, based at ISIT
(Institut Supérieur d'Interprétation et de Traduction) in Paris. These trends have also
materialized in a conference held in Turku, Finland, in August 1994, organized by the
University of Turku, the SSLMIT of Trieste, and ISIT in Paris.
4. Theoretical issues
Most studies on interpreting have so far focused on the central processes of simultaneous
interpreting. An important question for early investigators was whether simultaneous
interpreters actually translated simultaneously, that is whether they actually listened and
spoke at the same time. Some contended that this only occurred rarely, and that most of the
interpreter's speech production was done during the speaker's pauses. Various empirical

5
studies, however, have shown that this is not the case (Gerver 1976).
The next important question pertained to the nature of the mental activities which take
place during simultaneous interpreting. While all researchers agree that speech perception
and production are part of the process, little work has been done on other activities which are
assumed or known to take place, and little is known about the similarities and differences
between speech production and speech comprehension in interpreting vs. other contexts. For
proponents of the théorie du sens, in particular Seleskovitch and Lederer of ESIT, there are
no such differences. Gerver's (1976) and Moser's (1978) models of simultaneous interpreting
are also based on psycholinguistic models of ordinary comprehension and production.
Dillinger (1989) similarly contends that comprehension in interpreting is basically the same
as comprehension in everyday life. However, many practitioners and teachers of interpreting
point to a number of phenomena which suggest otherwise. For instance, as regards
production, they stress that because of the risk of loss involved when lagging too far behind
the speaker, interpreters often have to start formulating their TL sentences before having a
full picture of the idea to be expressed. Some, especially Ilg (1978), advocate
counter-strategies which involve selecting `neutral' sentence beginnings that allow the
interpreter to steer the sentence more easily towards the speaker's conclusion, once that
conclusion has been grasped. Others stress that interpreters have to resist constant linguistic
interference from the source-language, sometimes by avoiding TL words and structures that
are too similar to those used in the source-language speech. With respect to speech
comprehension, many point out that the interpreter's knowledge of the subject and the
situation is inferior to that of the other participants, and that s/he still has to achieve
comprehension beyond the level generally expected from listeners not conversant with the
subject (Gile 1989). Other potential differences are thought to exist but have not been
investigated systematically. Another important aspect of the interpreter's mental activity
concerns crisis management, that is coping tactics that are selected and implemented when
interpreters face comprehension, production and other difficulties (Gile 1989, 1994, 1995a,
1995b). In consecutive mode, an important part of the interpreter's mental activity is related
to note-taking, that is selecting the information to be noted, and the mode of notation, as well

6
as the way the notes are used during the reformulation phase.
5. Processing capacity and the `Effort' models
Over the past few years, researchers have been focusing on the interpreter's processing
capacity and its role in interpreting. Cognitive psychologists have known for some time that
while some operations are `automatic', in the sense that they require no processing capacity,
others are `non-automatic' and take up processing capacity, which is available in a finite
amount. In the Effort models of interpreting, developed in an attempt to explain the recurrent
and very frequent errors and omissions found in the performance of beginners and seasoned
interpreters alike, Gile (1989) argues that the main components of the interpreting process are
non-automatic. Simultaneous interpreting is divided into three sets of `Efforts': (a) the
Listening and Analysis Effort, which aims at comprehension of the SL speech, (b) the
Production Effort, which aims at production of the TL speech, and (c) a Short-term Memory
Effort, which handles information between perception and production in the TL.
As far as consecutive interpreting is concerned, this is divided into a listening phase,
during which the interpreter listens to the speaker and takes notes, and a reformulation phase,
during which the interpreter reformulates the speech in the TL. During the listening phase,
the Efforts are the Listening and Analysis Effort, the Note Production Effort, and the
Short-term Memory Effort for the management of information between the time it is received
and the time it is taken down. During the reformulation phase, there is a Note-reading Effort,
a Long-term Memory Effort to remember the speech, and a speech Production Effort. Gile
argues that in competent interpreters only the first phase is critical, since the second is not
paced by the speaker and does not involve much attention-sharing.
For interpreting to proceed smoothly, two conditions have to be met in simultaneous
and in the (critical) listening phase in consecutive: first, the sum of the individual Efforts'
processing capacity requirements should not exceed the total available capacity; second, at
each point in time, the capacity available for each Effort should cover the requirements
associated with the task the Effort is engaged in. If either condition fails to be met, the quality
of interpreting deteriorates, resulting in errors, omissions, clumsy reformulation of the
speech, and so on.

7
 According to the Effort models, triggerers of interpreting difficulties fall into two
categories. The first includes those which increase processing capacity requirements either
because they require more processing per unit time (for example dense or fast speeches and
enumerations) or because their signal is noisy or distorted (for example heavily accented
speeches, speeches with unusual grammar or logic, noisy physical environment and
inadequate acoustic equipment). The second category includes speech segments which raise
difficulties for the Listening Effort because of their brevity and lack of redundance (for
example numbers, short words and names).
The Effort models also explain errors in the interpreting of seemingly easy speech
segments by attributing them to saturation or processing-capacity deficit involving earlier,
more difficult segments and leading to the transfer of processing capacity and to a chain
reaction in which the failure occurrs at some distance from the actual triggerers (Gile 1989).
The concept of processing capacity is also linked to the type of mastery of working
languages required from interpreters. Because of time constraints and limited processing
capacity, the interpreter not only has to know the words and linguistic rules in his/her
working languages, but their active use in comprehension or production must be fast and take
up little processing capacity; in other words, the interpreter's linguistic knowledge must be
highly `available'. This requirement is critical in interpreters, as opposed to translators, who
do not have to share attention and who can devote minutes, hours or more to the
comprehension of text segments or to the retrieval of words or linguistic rules for use in their
target text.
The concept of `processing capacity' can shed some light on the much debated issue
of the desirability of working from an A into a B language or vice-versa. Many West
European interpreters claim that the only language mastered well enough to produce
acceptable target language utterances is an A language, and that interpreters should therefore
only work into their A language. On the other hand, many interpreters from the former
Eastern block suggest the opposite, namely that an interpreter should work from an A
language, because this is the only one that s/he understands well enough to react to rapidly.
The question of whether or when interpreters do achieve the required level of competence in

8
their A and B languages is by no means an easy one to resolve, as no precise and reliable
tools for measuring such competence are yet available. That aside, however, the issue of the
#DIRECTION OF TRANSLATION## can be addressed in terms of time and processing
capacity requirements in the Listening Effort and in the Production Effort. If it can be shown
that the Listening Effort takes up much more processing capacity, the A into B argument
becomes more convincing. If, on the other hand, it turns out that it is the Production Effort
which takes up much more processing capacity, the B into A argument becomes more
plausible. If no major difference is found to exist between the two, both arguments would
have to be assessed by reference to other factors, such as the individual interpreter's
command of the languages involved, his/her flexibility in adapting the structure of the output
to accommodate the incoming input despite anticipation difficulties, and any lexical or
grammatical peculiarities of the source or target languages which might influence ease or
difficulty of comprehension (for example the level of redundancy).
The issue of source/target language `peculiarities' raises the question of whether
interpreting is language-specific, that is whether interpreting between two specific languages
is more difficult or involves different processes and/or strategies from those used in other
language combinations. Proponents of the théorie du sens assert that this is not the case, but
other scholars have pointed to a number of specific linguistic features that do or may
influence the level of difficulty in interpreting. For instance, production may be more or less
difficult depending on the lexical richness and syntactic flexibility of the TL. Ease and
reliability of reception may be influenced by internal grammatical and lexical redundancy
(short or long words, grammatical indicators). In languages such as Japanese and Chinese,
homophony may also increase the amount of processing capacity and/or time required for
decoding. Syntactic differences between source and target languages may also increase the
level of difficulty, mainly due to the mandatory storage of a larger amount of information
between comprehension and production: information needed to proceed with formulating the
TL sentence may only be given in the SL sentence after other information, which would
typically be reformulated at a later stage in the TL. These hypotheses, however, await
empirical testing by means of linguistic and psycholinguistic studies in the future.

9
 See also #COMMUNITY INTERPRETING##; #COURT INTERPRETING##;
#PSYCHOLINGUISTIC/COGNITIVE APPROACHES##; #SIGNED LANGUAGE
INTERPRETING##.
Further Reading: Dillinger 1989; Gile 1994, 1995a, 1995b; Gran & Dodds 1989; Lambert &
Moser-Mercer 1994; Pöchhacker 1994; Target 7(1) 1995; Tommola 1995.
Daniel Gile
[END]

10

View publication stats