Interpreting research

What you never wanted to ask but may like to know.


An increasing number of AIIC members sail through life and the booths with a trawlfull of academic titles and hundreds of pages of prose about interpreting. They appear to be claiming to know more than others about what we really are and what we are actually doing when interpreting, and generate a fair amount of puzzlement and some hostility among colleagues. As a member of this trawlers' community, and one with a split personality to boot, I will try to both describe and show the limitations of this pursuit of ours, in an attempt to give colleagues a better idea of what to expect from interpreting research, and more importantly perhaps, what not to expect.

Trends in the evolution of research into interpreting

What first attracted researchers to conference interpreting seems to have been the mystery of the mental processes underlying the strange activity of speaking and listening at the same time. In the late sixties and up to the mid-seventies, a few psychologists and psycholinguists engaged in a first attempt to understand these processes. Understandably, they used their own theoretical and methodological tools, largely experimental and quantitative, which were not necessarily adequate for the task at hand. They neglected the strategic dimension of interpreting, that is, the fact that interpreters choose to a large extent what they say and how they say it using criteria which have to do with communication issues and social norms rather than with an attempt to preserve language structures or all the information content of the speaker's words. This first wave of actual research soon died out, and ever since, not more than a handful of non-interpreters have been active in research into interpreting at any time. To a large extent, this may be due to objective difficulties, as explained further down.

Be it as it may, in the early seventies, a few interpreters decided to take research into interpreting into their own hands. With much motivation and some institutional support, they wrote many essays and prescriptive texts based on their professional experience, on introspection and on some reading, and even obtained doctoral degrees. The movement (initially driven and inspired by Danica Seleskovitch of ESIT) gathered impetus, and hundreds of papers of a similar nature were published in the seventies and eighties. Meanwhile, there was little input from academics and researchers in other disciplines, who were shut out from the interpreters' literature, with a very few exceptions.

Eventually, in the late eighties, the voices of interpreters such as Laura Gran, Ingrid Kurz, Jennifer Mackintosh, Barbara Moser, Catherine Stenzl and myself, who spoke in favor of research more in line with the methods of established empirical scientific disciplines, were heard. Since then, the trend has been favorable to the use of concepts, theories and methods from other disciplines, and many studies conducted today rely on what is developed in psychology, linguistics, neurophysiology, etc. (two recent examples are Setton's (1997) and Shlesinger's (2000) doctoral dissertations, the former linking interpreting and pragmatics, a branch of linguistics, and the latter linking interpreting and cognitive psychology). Additionally, there has been a dramatic growth both in the number of studies published in the field, from about 180 in the seventies, to close to 500 in the eighties, and to more than 1000 in the nineties, and in the geographic extension of research into interpreting: While in the sixties and seventies most publications on interpreting came from France, Germany and Switzerland, in the nineties, they came from all over the world, the most 'productive' countries being Italy (see Gran and Viezzi 1995) and Japan (see Kondo and Mizuno 1995), followed by countries such as the Czech Republic (see Cenkova 1995), Finland and Spain. For a more detailed analysis of facts and figures, see Gile 2000. A strong driver of such production is the institutionalization of interpreter training programs in universities, and the ensuing requirements for graduation theses. The interpretation research scene has become quite different from what it was in the seventies (though I would happily contradict Seleskovitch's pessimistic assessment of its evolution (Seleskovitch 1997) and say that it has improved dramatically).


To most readers within AIIC, the question is probably what research has achieved in the field of interpreting. Has it taught us things that we do not know? Has it developed better interpreting methods or training methods?

The short, honest and conservative answer is that research cannot claim to have made any major discovery so far, or to have developed major applications in professional interpreting or training. But:

1. Research is a long-winded endeavor, and it does not make sense to expect major results after three decades of work, especially when it is conducted under difficult conditions, as is shown further down.

2. Research into interpreting also has other roles, in particular a social role: being associated with academia, it can help raise the social status of interpreters and support their claims for a remuneration similar to that of highly qualified professionals. This is probably where the largest contribution of Seleskovitch and her group in the seventies and eighties lies.

3. Research has stimulated reflection about interpreting, and helped develop and formulate central concepts which serve as explicit guidelines to trainers and practitioners (see Gile 1995a, Falbo et al. 1999). One important example is that of "deverbalization", a stage in interpreting postulated by Seleskovitch where the source speech in its linguistic form disappears completely from the interpreter's mind, to be replaced by some non-linguistic representation of its "sense". The existence of such a stage in the strict sense, meaning that no trace of the linguistic form of the source speech is present in the interpreter's memory, has never been demonstrated, but the concept is a major one nevertheless insofar as it asserts the legitimacy of meaning-based (as opposed to word-based) interpretation, and seems to be accepted and put into practice by interpreters and interpreter-trainers worldwide.

Such reflection about interpreting also helps develop more specific ideas, with theories, precise hypotheses, and support from research in other disciplines. One may rightly object that it has not been able to answer fundamental questions, such as:

  • Are interpreters born, not made, or made, not born?
  • Is it possible to devise very efficient admission selection tests in training programs?
  • Should interpreters work only into their A language, or can they provide good target speeches in their B language?
  • Is interpretation language-specific or language-independent?
  • Are interpreters more accurate in consecutive or in simultaneous?

Such questions have, however, been reformulated in new terms, taking into account what is known in linguistics and cognitive psychology, and specific hypotheses are being tested. For instance, we now know that cognitive operations, including speech production, speech comprehension, and decisions on what to keep and what to omit, take up resources out of a limited available amount, and we can attempt to find answers to questions and explain and predict problems, failures and strategies on this basis. Thus, progress is made possible, whereas in the past, claims and counterclaims tended to be holistic and categorical, and therefore static.

4. Research has led to some advances, albeit not revolutionary, in the exploration of interpreting. For instance, we     now have some documented knowledge about quality expectations by users of our services, and evidence about       the variability of quality assessment, about the lack of reliability of fidelity assessment in consecutive, about what       happens linguistically to a rather written-like/spoken-like source speech when it is interpreted, about the proportion       of numbers and names that are missed in interpreting, about variability in interpreting performance for the same       source speech under the same conditions, about the capacity of the interpreter's 'working memory' vs. that of non-interpreters, about patterns of intonation in target speeches, about the evolution of pauses during the interpretation learning process, about the relative accuracy of simultaneous vs. consecutive, about anticipation, and about language-specific strategies used in interpreting. We even have some evidence that seems to contradict some of our received wisdom, for instance about the importance of seeing the speaker, or of knowing about the speeches in advance.

We must understand that such findings are both very modest and tentative. No serious researcher will claim that       isolated findings are final. They need to be checked, replicated, and integrated into further research before their       status can be consolidated and before their implications on fundamental questions can be assessed. This careful,       systematic, and (relatively) objective nature of scientific findings is what makes them different from the results of       personal experience and introspection, and this is where their added value lies. In that respect, the modest findings contributed by interpreting research so far do provide signs that some progress is being made, in that we are starting to collect (relatively) solid information which should allow us to gradually build our knowledge and professional orientations on more than intuition.

Last, but possibly not least, for individual interpreters who engage in it, interpreting research has a valuable       contribution to make. Firstly, it makes them look at interpreting more systematically and more carefully, and       thus makes them aware of certain phenomena which they might otherwise overlook. This may influence the way       they work in the booth. Secondly, proper research is an exercise in analytical, rigorous thinking, and may be useful as such. Thirdly, research is an intellectually stimulating, enjoyable activity, which has given me and others much pleasure over the years (see Strolz 1995 for an interesting personal account). Let us not lose sight of this point, which may be the most important motivation for many of us interpreting researchers.


Interpreting as an object of scientific investigation combines the complexity of language comprehension and production with that of human behavior, which, in our case, is determined by many factors and is constrained by highly diverse environments. It follows that before major conclusions about interpreting are drawn from research, a very large number of situations, variables and interactions must be studied, probably more so than in many other fields of human behavior.

But how do you conduct hundreds of experiments on a population of less than 5000 individuals scattered over many countries, with different language combinations, different training and qualification backgrounds, of different ages, many of whom will refuse to take part in any study in which their output will be scrutinized? Mission impossible, clearly. At this point, it does not seem realistic to set as a goal for research a thorough, comprehensive exploration of interpreting. It is much more reasonable to aim at more humble endeavors designed to gain insight into specific phenomena.

Another problem has to do with the researchers themselves. Virtually all of us (a few hundred at any time, but only a few dozen are active over a period of several years and produce more than one study) are interpreters, or student interpreters where MA and graduation theses are concerned. We may well claim that interpreters are the only ones who understand what interpreting is all about, but how available and how qualified are we for research? Out there in the 'professional' academic world, researchers are trained over many years, through formal courses in research methods and through close work with supervisors who guide them and correct them. Once they have completed their training, they continue to devote much of their time to research, in a situation of competition which forces them to maintain quality in their studies. Many of us have earned doctoral degrees, but how many of us have gone through formal research training? How many of us have worked under the close supervision of qualified researchers who know about interpreting? How much of our time is devoted to research, and how much competition do we encounter when we seek to publish? In the field of interpreting research, the mechanisms that promote quality and filter out poor texts are less effective than they should be, and the methodological level of our research is correspondingly lower. This is why in the literature, there are many prescriptive, repetitive, non-innovative essays with a low value-to-length ratio, as well as many scientifically flawed research texts.

This problem is made worse by the fact that our academic titles and publications generate prestige, and we rapidly reach positions probably higher than those we would reach in a more competitive, more selective environment. Once up there, it is only human to hold on to the status, to be reluctant to admit our weaknesses, to be defensive, and sometimes aggressive. Many developments on the interpreting research scene over the years can probably be accounted for by socio-psychological considerations.

This does not mean that all of our research should be discarded. Interesting, innovative, methodologically sound studies are also part of the literature, but probably in a smaller proportion than in most established disciplines.

Interpreting research and professional issues

Interpreting research is expected to contribute to the profession in several ways. One is to offer information which will allow us to gain better insight into interpreting and its environment and develop better working methods and training methods. Yet another is to support our professional claims and demands by showing that they are justified. But this is where problems may arise, when findings are inconclusive or suggest that our claims as professionals are not justified.

For instance, several studies over the years failed to find significant differences in the performance of interpreters who could not see the speaker as opposed to those who could, and one recent study on the effects of prolonged turns in interpreting (Moser-Mercer et al. 1998) failed to demonstrate convincingly that quality dropped after an hour or so in the booth. How do we, as interpreters interested in defending our professional interests, react to such results? Do we accept to change our professional claims and demands, do we disregard the data, or do we quote only those findings that are favorable to our positions, at the risk of seeing opponents (employers, the FTC, non-AIIC interpreters) using the evidence (or lack thereof) against us?

Going further, much evidence from empirical research indicates that target speeches we produce may contain many errors. Are we willing to acknowledge that fact? How does it affect the very image of conference interpreters that we seek to project, that of high-level professionals who are capable of reformulating faithfully high-level, complex speeches? Are we interpreters willing to follow Pearl's courageous analysis of our fallibility (Pearl 1999) and change our professional policy accordingly?

Research strives to be objective. Professional associations do not, and their respective interests may clash. This problem should not be overlooked.


By way of a conclusion:

For objective reasons, demographic, geographic, economic and institutional, it is unlikely that research into interpreting will present major discoveries to the community in the immediate future. I will admit to some puzzlement when I hear colleagues make strong claims about their theories and findings, and about their immediate contribution to interpreting, but, as I hope to have shown convincingly:

  • Some contribution has been forthcoming and can be reasonably expected to keep coming in the form of modest findings, orientations for training, social status for the profession, and personal development and pleasure for individual researchers.
  • The cost of interpreting research to the community is virtually nil, so that even modest contributions can be seen as positive.

However, with respect to professional interests, in particular those defended by AIIC, research is a double-edged sword, to be handled with caution.

I hope this short paper can help induce realistic views and expectations with respect to research into interpreting, and possibly generate some interest in the field.

A Selection of References

This list includes all publications mentioned in the body of the article, with a few additions. It has been kept short for obvious reasons. Except for Setton's and Shlesinger's doctoral dissertations, which were mentioned in the body of the text, only texts which can be read rather easily by non-experts have been selected.

Cenkova, Ivana. 1995. La recherche en interprétation dans les pays d'Europe de l'Est: une perspective personnelle. Target 7:1. 75-89.

Falbo, Caterina, Mariachiara Russo & Francesco Straniero Sergio (eds). 1999. Interpretazione Simultanea e Consecutiva. Problemi teorici e metodologie didattiche. Milano: HOEPLI. (An example of how the findings of research can be integrated into interpreter training.)

Gambier, Yves, Daniel Gile and Christopher Taylor (eds). 1997. Conference Interpreting: Current Trends in Research. Proceedings of the International Conference on Interpreting: What do we know and how? Amsterdam/Philadelphia: John Benjamins.

Gile, Daniel. 1995a. Basic Concepts and Models for Interpreter and Translator Training. Amsterdam/Philadelphia: John Benjamins. (Another example of how the findings of research can be integrated into interpreter training.)

Gile, Daniel. 1995b. Regards sur la recherche en interprétation de conférence. Lille: Presses Universitaires de Lille. (A qualitative analysis of the evolution of interpreting research.)

Gile, Daniel. 2000. The history of research into conference interpreting: a scientometric approach. Target 10:1. 69-93.

Gran, Laura and Maurizio Viezzi. 1995. Development of Research Work at SSLM, Trieste. Target 7:1. 107-118.

Kondo, Masaomi and Akira Mizuno. 1995. Interpretation Research in Japan. Target 7:1. 91-106.

Kurz, Ingrid. 1996. Simultandolmetschen als Gegenstand der interdisziplinären Forschung. Wien: WUV-Universitätsverlag. (A readable account of interdisciplinary research into interpreting.)

Moser-Mercer, Barbara, Alexander Künzli, Marina Korac. 1998. Prolonged turns in interpreting: Effects on quality, physiological and psychological stress (Pilot study). Interpreting 3:1. 47-64.

Pearl, Stephen. 1999. The Other Three Eights and the Four F's. Finiteness, Fallibility, Freedom of Speech and Fair Competition in the simultaneous interpretation environment. The Interpreter's Newsletter n°9. 3-28.

Seleskovitch, Danica. 1997. Statement made during the Prix Danica Seleskovitch 1996 ceremony. AIIC Bulletin 25:2. 27-28.

Setton, Robin. 1999. Simultaneous Interpretation: A cognitive-pragmatic analysis. Amsterdam/Philadelphia: John Benjamins. (Highly technical; published version of Setton's 1997 doctoral dissertation.)

Shlesinger, Miriam. 2000. Strategic Allocation of Working Memory and Other Attentional Resources in Simultaneous Interpreting. Doctoral dissertation, Bar Ilan University, Israel. (Highly technical.)

Strolz, Birgit. 1995. Une approche asymptotique de la recherche sur l'interprétation. Target 7:1. 65-74.

Electronic sources of bibliographical and other information on interpreting research

The CIRIN Bulletin

The AIIC Bibliography

The web site of Akira Mizuno and the Japan Association for Interpretation Studies: (a wealth of information for those who read Japanese).

Recommended citation format:
Daniel GILE. "Interpreting research". April 11, 2001. Accessed July 12, 2020. <>.