finding Korean equivalents, he will not have enough time and processing capacity for the next incoming message. This will be especially true when the next incoming message is incidentally short enough to be in the TTS of the previous sentence. In this case, the interpreter might spend his entire processing capacity on the previous sentence, reserving a little or no capacity for the next short sentence. Under these circumstances, the next sentence might be omitted in SI.
Syntactic differences also play a part in this. As Ahn (2005, p. 696) maintains “In simultaneous interpreting, if the syntactic structure of the source language […] and the target language […] are very different, interpreters have to wait before being able to reformulate the SL segments into a meaningful utterance in TL. Let us now imagine an interpreter is working from English into Persian. It is known that English uses SVO while Persian is basically an SOV language. The interpreter has to keep in his memory the English verb, received early in the utterance, until the object of the utterance is fully expressed, whereupon the rendition in TL becomes possible. This is particularly problematic in cases where the noun phrase functioning as the object of the sentence is too long. Gile (1997, p. 200, as cited in Lee, 2002, p. 603) explains the phenomenon from the point of view of his Effort Model:
If additional processing capacity is then allocated to the memory effort, this may in turn deplete the capacity available for listening and analysis effort, leading to a potential problem in the comprehension of another SL speech segment.
Another problem with a long EVS is that in SI, unlike CI, the message is almost constantly being presented to the interpreter at a speed which is set by the speaker, and not in accord with the interpreter’s processing capacity. Therefore, “if the interpreter tries to lag far behind the speaker for complete understanding of the sentence, the next sentence will be flowing in and the EVS will exceed the length of one sentence” (Lee, 2002, p. 603).
In conclusion, Lee (2011, pp. 154-155) holds:
[…] many fundamental skills could be relevant, one of them is the control of attention. Speedy retrieval of equivalents will be one of the most important skills and, as mentioned earlier, failing to complete the current sentence as quickly as possible and move on to the next incoming sentence will lead to overloading of processing capacity. Therefore, successful interpreters are those who manage to maintain the delicate balance between their cognitive strengths and weaknesses and who have developed coping skills (Moser-Mercer 2000/2001). Interpreters also should be able to distribute their limited information processing capacity optimally to several modules of processing.
Another problem is related to the audience’s perception. An extended EVS means a rather long pause in the IT as the interpreter stops uttering his rendition while paying all his attention to listening to SL. In such cases, the audience assumes any silence in SI reflects a loss of information (Déjean Le Féal, 1990, cited in Lee, 2002). In line with this, is Moser-Mercer’s (1996, cited in Lee, 2002, p. 603) finding that “nearly 90% of users of conference interpretation found the interpreter’s long pauses and lagging far behind the original speech irritating”. Therefore, it stands to reason to claim that interpreters cannot increase their EVS at their own discretion. Yet, following too close to the original speech (i.e. a too short EVS) runs the risk of producing a word-for-word translation and may lead to a faulty interpretation. It is in this context that striking a balance and “maintaining optimum EVS would be key to ensure quality SI” (Lee, 2002, p. 603). In his concluding remarks on the length of EVS and quality in English into Korean SI (which can be rightly assumed to hold true for other language pairs too), Lee (2011, p. 158) states that “maintaining optimal EVS is extremely important since too short EVS might end up with a clumsy target language due to the lack of comprehensive understanding of source text, while a too long EVS will lower the quality of SI of the sentence as well as the incoming one too.”
From what was discussed above, it becomes clear that there is some sort of correspondence between a long EVS and the vertical approach and also a short EVS and the horizontal perspective.
With regards to interpreter training, the importance of the question of EVS becomes ever more evident. To be able to take calculated decisions and appropriate measures while performing SI, the interpreter must understand the nature as well as the cause of EVS and must be able to adjust the length of EVS while simultaneously interpreting. One of the best ways to do this is to introduce the concept of EVS in SI classes and to use it as a factor adjusting the difficulty level of preparatory tasks such as shadowing and on-line or simultaneous paraphrasing. These teaching techniques, along with others, will be discussed in the next section.
2.4 Interpreter-Training Techniques
The following paragraph reflects Gran’s perspective on SI training:
Gran (1998) […] advocates a gradual approach to the acquisition of simultaneous interpreting skills beginning with exercises training one skill at a time, such as text analysis, abstracting, paraphrasing, and subsequently moving over to the whole task. At a later stage, training will be devoted to particularly difficult or complex parts of the interpreting process – the speaker’s pronunciation, speed of delivery, density of information, specialized terminology, rhetoric. (Ribas, 2012, p. 816)
The present study adopts Gran’s gradual approach to the teaching of SI, that is, we believe that certain exercises, applied in a logical sequence, can prepare the trainees step by step for the task of SI. However, within the scope of the present study, we do not, actually cannot, reach the later stage mentioned above. Within the time period of the experiment, we only afforded to introduce the trainees into the SI training program.
The present study, through application of a number of interpreter-training-specific exercises, is an attempt to shed light on the effectiveness of these teaching activities to overcome the mental barriers. Such activities (e.g. shadowing, improvisation, divided attention exercises, memory-improving exercises, anticipation exercises, paraphrasing exercises, and the like) have been discussed at length in the literature on interpreting (see Chernov, 2004; Green et al., 1994; Kalina, 1992; Kurz, 1992; Moser-Mercer, 1997; Padilla & Martin, 1992; Pöchhacker, 2004). There are data-driven as well as theory-based arguments for and against each and every one of these exercises.
However, they are applied as a package in this experiment to see whether they can positively affect the outcome of trainee interpreters or not. The activities have many different variations to suit the trainees’ level of mental preparedness. However, we will only introduce them briefly here.
Shadowing involves repeating what the speaker says, in exactly the same way, just after him. The trainee listens (preferably through headphones) to someone talking in L1 or L2 and tries to imitate them. The time lag may vary at different stages to adjust the level of difficulty of the task. As “a skill that is sufficiently attention demanding,” (Corina & Vaid, 1994, p. 240) shadowing allows your mind to get accustomed to listening and speaking at the same time (Lambert, 1992; Seleskovitch, 1999). Although its “usefulness is strongly challenged by many […] There are still some scholars who believe that shadowing practice helps your interpretation sound smooth and natural in the target language.” (Bahri & Gholami, 2012, p. 29)
We will now look at this cognitive exercise more closely. Shadowing is defined in more technical words by Lambert (1988, p. 377; 1994, p. 321) as “a paced, auditory tracking task which involves the immediate vocalization of auditorily presented stimuli, in other words, repeating word-for-word, and in the same language, a message presented to a subject through headphones.” Yagi (2000, p. 527) writes: “Shadowing is a simultaneous task like SI but it suffers no language transfer hindrance.” There are two main types of the shadowing exercise distinguished by Norman (1976, as cited in Lambert, 1988; Lambert, 1994): phonemic shadowing and phrase shadowing.
In phonemic shadowing, “subjects repeat each sound as they hear it, without waiting for the completion of a propositional phrase or “chunk”, or even a completed word, so that the shadower remains right “on top of” the speaker” (Lambert, 1994, p. 321). This shows that in phonemic shadowing, comprehension plays no role at all. The shadower may even be asked to shadow a speaker who is speaking a totally new language or even nonsense. The task is quite mechanical, involving no cognitive struggle for understanding, but rather only training the subject to be vigilant to the incoming sounds, react quickly, and speak and listen concurrently.
On the other hand, phrase shadowing is a version of shadowing exercise “where subjects repeat the speech at longer latencies, more precisely from 250 milliseconds upwards, and where shadowers wait for a semantic chunk before vocalizing, in the same way that a simultaneous interpreter would lag behind the original speaker” (Lambert, 1994, p. 321). It is clear from the definition that this type of shadowing draws more heavily upon cognitive resources as it is less mechanical and more cognition-based than phonemic shadowing.
The rationale behind using shadowing as a preparatory exercise in SI training programs is made clear by Lambert (1992, p. 266; 1994, p. 321) who maintains that shadowing is “frequently used in various