This extract is taken from pages 191-201 of Daniel Gile's excellent "Basic Concepts and Models for Interpreter and Translator training", 1995, Benjamins of Amsterdam.
In spite of interpreters' preparation strategies, problems do arise in interpreting situations (see Gile 1989) because of processing capacity limitations (as explained in chapter 7), errors in processing capacity management, and gaps in the interpreters' Knowledge Base. Many of these problems can be said to be unavoidable, as shown by the fact that they are encountered regularly even by interpreters with a solid reputation and long professional experience. Interpretation has been referred to by some professionals as "crisis management," and in the light of interpreters' daily experience, these are apt words to describe an aspect of interpreting which is virtually unknown to the public at large.
Difficulties affect both comprehension and production, often through failure sequences as explained in chapter 7. When interpreters are aware of such problems, they tend to use a rather small set of tactics to limit their impact.
Coping tactics are a very fundamental practical skill in interpreting. Basically, they are taught within the framework of practical exercises. In most training programs, this is done by trial and correction, with trial on the student's part and corrections from the instructor. Such corrections are generally normative; instructors sometimes refer to the communication impact of the tactics in order to explain their preferences, but are not necessarily aware of other factors which influence them.
This chapter attempts to provide instructors with a list of basic coping tactics for a general view of the issue. It also presents a conceptual framework which spells out the advantages and drawbacks of each tactic, and discusses a few rules which may help explain what makes interpreters prefer one tactic over the other beyond their individual merits.
2. Tactics in simultaneous interpretation
2.1 Comprehension tactics
The following are the main tactics used when comprehension problems arise, and when they threaten to arise under time-related or processing capacityrelated pressure.
a. Delaying the response
When a comprehension difficulty arises, interpreters may respond immediately with one of the other tactics presented below. However, they may also delay their response for a while (a fraction of a second to a few seconds), so as to have some time for thought while they receive more information from the source-language speech. After a while, they may have solved the problem entirely, or else they may decide to resort to another tactic.
Because of its very nature, the Delay tactic involves an accumulation of information in short-term memory, and is associated with the risk of losing speech segments in a failure sequence as outlined in chapter 7.
b. Reconstructing the segment with the help of the context
When interpreters have not properly heard or understood a technical term, name, number, or other type of speech segment, they can try to reconstruct it in their mind using their knowledge of the language, the subject, and, the situation (their extralinguistic knowledge).
The reconstruction process is an integral part of speech comprehension in everyday situations as well. It is defined as a tactic in the present context when it becomes a conscious endeavor, as opposed to an ordinary, subconscious process.
If successful, reconstruction can result in full recovery of the information. However, it may entail some waiting until more information is available and require considerable time and processing capacity. Like the Delay tactic presented above, it is associated with a high risk of saturation and individual deficits. Reconstruction from the context can therefore not be considered a high-priority tactic.
c. Using the boothmate's help
In simultaneous interpretation, there are theoretically at least two interpreters in the booth at all times. One is active (producing a target-language speech), while the other is passive (listening, but not speaking). The passive colleague, who can devote full attention to listening, has a better chance of understanding difficult speech segments than the active interpreter, whose processing capacity is being shared by the three Efforts. Moreover, on the production side, the passive interpreter can consult a glossary or another document, which takes up much time and processing capacity, and then give the information to the active colleague, generally in writing. The presence of a passive interpreter in the booth is therefore a major asset to the active interpreter.
The active interpreter can ask for the passive colleague's help with a glance or a movement of the head. In teams that work well, the passive interpreter will sense a hesitation in the active colleague's speech and understand there is a problem. He or she can also anticipate problems and write down names, numbers, technical terms, etc., without even being asked for help. When the problem is terminological, the boothmate will generally indicate to the active interpreter the target-language term if possible, so that it can be used for reformulation. When the problem lies with a single word, name, or number, the passive boothmate can also write it down in the source language for the benefit of the active interpreter who did not hear it correctly. It is much more difficult, however, to explain an idea efficiently, because the active interpreter does not have time to read a long explanation.
This tactic is a very good one because it does not cost much in time and processing capacity, and pooling the knowledge and intelligence of two persons, one of whom does not have to divide attention between listening and other tasks, provides a better chance of finding the information than using the resources of only one person.
However, in order for the tactic to work, the passive interpreter must be not only physically present in the booth, but also available and willing to make the effort and help the active colleague. This situation does not always occur:
The efficiency of this tactic varies greatly: looking for a term in a commercial dictionary may require much time and processing capacity, but finding an important word in a document that was read and marked before the conference can be very fast. This is why it is important to pay attention to both the preparation of documents and their management in the booth. Instructors should show students how to make important names and terms stand out for quick reference, using highlighters or other means. Writing important technical terms and names on a sheet of paper in front of the interpreter (beside the glossary prepared for the conference) is another way of making them readily available. In particular, documents should be laid out in the booth, sorted, and marked in such a way as to minimize the time needed to access them and to recognize their identification numbers or titles, possibly with different stacks for each language, sorted by numerical sequence, type of document, etc.
2.2 Preventive tactics
The following tactics are used when time or processing capacity pressure is such that the interpreter believes a problem may arise or is about to occur. The idea is to limit the risks of failure.
a. Taking notes When the speech contains figures and names that interpreters feel they may forget and that they cannot reformulate right away for syntactic reasons, they may take them down in notes. While affording greater security as regards the items which are taken down, this tactic entails a high cost in time and processing capacity, which increases the risk of losing other items of information that come before or after those written down (this is an interference phenomenon, as explained in section 3). The risk is reduced significantly when it is the passive colleague who writes the information for the active colleague.
It is interesting to note that when translating in simultaneous from and into Japanese, some Japanese interpreters take down not only numbers and names, but also other information which Westerners generally do not write (in this case, it is often the passive interpreter who takes down the information for the active colleague). The reason given by them is that syntactic structures differ greatly between Japanese and other (mostly Western) languages, which leads to much waiting before the reformulation of any specific part of a sentence, hence a possible overload of short-term memory and an increased risk of losing information (see chapter 9).
b. Changing the Ear-Voice Span
By changing the Ear-Voice Span (EVS), that is the time lag between comprehension and reformulation, interpreters can control to a certain extent the processing capacity requirements for individual Efforts. By shortening the lag, they decrease short-term memory requirements, but deprive themselves of anticipation potential and run the risk of misunderstanding a sentence and driving themselves into target-language sentences which will be difficult to complete. By lagging further behind, interpreters increase comprehension potential, but may overload short-term memory.
Teachers sometimes advise students to try to lengthen or shorten their EVS in specific cases, but there does not exist a clear-cut, consistent theory or set of operational rules on the subject. It seems that EVS regulation is learned with experience; I believe that this is the single largest benefit derived from practice in simultaneous interpretation during initial training.
When faced with potential overload of memory, as with a source language and a target language that are syntactically very different, with embedded structures in the source language, or with unclear sentence structures, interpreters may choose to reformulate speech segments earlier than they would normally do, sometimes before they have a full picture of what the speaker wants to say. In such cases, they may resort to neutral sentence beginnings or segments in the target language that do not commit them one way or another. For instance, in a source-language sentence expressing a causal relationship such as:
Enumerations are high-density speech segments that impose a high load on short-term memory. One tactic often observed consists of reformulating the last elements first so as to free memory from the information, and then to move on to other elements. To my knowledge, no analysis has yet been performed as to why this should reduce Memory Effort load. One possible explanation is that by reformulating the last elements first, it is possible to pick them up before they have been processed in depth and integrated fully into the semantic network, thus saving processing capacity. This tactic may work best with names, which can be reproduced from echoic memory (memory of the sound), or with terms which are easily transcoded; it may not be very effective if such elements cannot be transcoded or reproduced phonetically and require more processing capacity anyway.
The following are tactics used in reformulation in order to eliminate the potential consequences of production problems or short-term memory problems. The first three are the same as presented in section 2.1 on comprehension tactics.
This is the same tactic as used in comprehension, the idea being that the waiting period is used for a subconscious (or conscious) search for the missing term or sentence structure. As with the case of comprehension, the waiting entails a risk of short-term memory overload, as well as a possible increase in processing capacity requirements in the Production Effort when the information is eventually reformulated-because of the backlog that has accumulated in the meantime.
As can be inferred from the descriptions in section 2.1, the boothmate's help is more often given in the form of indications for reformulation than as explanations of what was said, which is reasonable in view of the strict time constraints involved.
When interpreters find themselves incapable of understanding a speech segment or reformulating it in the target language, one possible solution is to reformulate the message in a less accurate manner by using a superordinate in the case of a single word, or by constructing a more general segment in the case of a whole clause or sentence: "la streptokinase" may be reformulated as "the enzyme," "Monsieur Stephen Wedgeworth" as "the speaker," "deux cent trente trois millions" as "about two hundred and thirty million," "DEC, IBM, Hewlett Packard et Texas Instruments" as "a number of computer vendors," etc.
This tactic, which requires little time, implies loss of information in the target-language speech. This, however, does not necessarily mean that the information is lost for the delegates; it may be repeated in another sentence in the speech, or be already known to the delegates.
Interpreters may understand a term but not know the appropriate equivalent in the target language, in which case they can explain it. For instance, in one conference, the data processing term "tableur" (spreadsheet) was interpreted as "the program which defines rows and columns and allows calculations to be made."
This tactic can be efficient informationally but has two drawbacks: one is the large amount of time and processing capacity it requires, and the other is the fact that it may draw the delegates' attention to the fact that the interpreter does not know the proper term in the target language, possibly lowering his or her credibility and reducing the impact of the speech accordingly.
When encountering a name or technical term which is not known or recognized, the interpreter may try to reproduce the sound as heard. This is not an "intelligent" tactic insofar as it does not call for complex cognitive operations, but it can be efficient: if they know the name or term, delegates may hear it as it should have been pronounced, without even noticing that the interpreter has a problem. On the other hand, the approximation may also be heard and perceived as a distortion of the information, which may not only generate loss of information, but also discredit the interpreter.
When interpreters do not know the appropriate term in the target language, they may naturalize the source-language term, adapting it to the morphological or phonological rules of the target language. For instance, in a conference, the term "télédétection" (remote sensing) was rendered in English as "teledetection." Similarly, the English computer term "driver," as applied to a software program that helps operate a device such as a printer from a computer, or as applied to the physical unit that runs floppy diskettes, was translated into French as "driver" (pronounced "dreevair"), and into Japanese as "doraibâ."
Transcoding consists of translating a source-language term or speech segment into the target language word for word. For example, in the field of accounting, the English term "maturity date," the equivalent of which is "date d'échéance ", was interpreted as "date de maturité".
This tactic can be very efficient in the same cases as naturalization. Like naturalization, it can also lead to existing target-language terms; in various fields, many terms have been created by such transcoding by experts, just as
many terms have been created by phonetic naturalization. Even when transcoding does not lead to an existing target-language term, it may facilitate comprehension for the delegates because of the semantic indications the newly created term carries. For instance, in the field of dentistry, the English term "mandibular block" (a type of anesthesia) was interpreted as "bloc mandibulaire", whereas the appropriate term was "tronculaire". Delegates said afterward they had no trouble understanding "bloc mandibulaire", even though it bore no similarity at all with the appropriate French term.
When interpreters believe they have missed an important piece of information, they may decide to inform the delegates of the loss by stepping out of their role as the speaker's alter ego and saying for instance "... and an author whose name the interpreter did not catch," or "... the interpreter is sorry, he missed the last number." When this happens, delegates may fail to react, but they can also ask the speaker to repeat the information, either during the session itself or during a break.
This tactic is not used very often. One of the problems is that it takes up much time and processing capacity, and may therefore jeopardize the reformulation of other speech segments. Moreover, it draws the delegates' attention to the interpreter's problems. This has two drawbacks: first, delegates are interested in the speech, not the interpreters and their problems; second, by drawing the delegates' attention to his or her problems, the interpreter may lose credibility, and therefore also indirectly weaken the impact of the speaker's message.
To sum up, if important information is missed, interpreters consider it their ethical duty to inform delegates rather than gloss over it, but if the information is insignificant, or if informing the delegates may do more harm than good, they choose another tactic.
In specialized conferences, much of the information is given not only by the speaker, but also in written handouts and on screen, via slides and overhead transparencies. When encountering comprehension or reformulation difficulties, the interpreter can refer delegates to "the figures/names/equation etc. on the screen/in your handout," etc.
Interpreters may miss information without even noticing it because they did not have enough processing capacity available for the Listening and Analysis Effort when the speech segment carrying it was being uttered. They may also omit it because it disappears from short-term memory. The omission tactic refers to the case where an interpreter deliberately decides not to reformulate a piece of information in the target-language speech.
Again, not all the information which was omitted in the target-language speech is necessarily lost as far as the delegates are concerned, since it may appear elsewhere or be known to the delegates anyway.
When working conditions are particularly bad, and when interpreters feel it is imperative to continue speaking despite inability to listen, understand, and reformulate properly, they may invent a speech segment compatible with the rest of the source-language speech but not a faithful reflection of the problematic source-language speech.
This tactic is obviously an extreme one, to be used exceptionally and with the uttermost caution. I believe it should not be taught at the same time as other tactics. It is probably best left to the very end of training, when it is introduced
This is another extreme tactic. Some purists advocate its use when working conditions are poor and interpreters feel they cannot do a decent job. In actual practice, this is a very rare attitude. For all intents and purposes, it can be said that this tactic is only implemented when working conditions are so bad that interpreters believe they can do no useful work at all, meaning that interpretation would be worse than non-interpretation.
|Top of Page||Contact ITR|