Pauses in interpretation output

Researches on pauses in interpreting output have mostly focused on the functions and meanings of pause. Similar to spontaneous speech, pauses in interpretation are thought to be closely linked to cognitive efforts needed in the translation process. Due to the complexity of interpreting process, increase in interpretation hesitancy can likely be attributed to various reasons, ranging from difficulties of the task, directionality of interpretation, to the experience of the interpreter.

Manipulation of the source text has been commonly adopted in interpretation studies to control task difficulties and examine its effect on production fluency. Extra background noises were added to the source speech in SI task, adding stress to the interpreters as it would be harder to listen and comprehend the content for interpretation. As a result, pauses in the output were longer compared with those from output performed under normal circumstances (Piccaluga, Nespoulous, & Harmegnies, 2005). Increasing density of information delivered in the source speech within a certain timeframe may also be cognitively demanding, as interpreters need to comprehend, reformulate and produce densely packed information within a short period of time. Fastening of the source speech was shown to increase the number of pauses in interpretation output (Piccaluga et al., 2005). Increase in

25

the speaker’s speech proportion, meaning the ratio of speaking time to the total time of SI, also increase the density of information in the source text, and results in more pauses and errors in the interpreted output (Lee, 1999).

Directionality is another factor affecting cognitive efforts unique to interpretation tasks. It is commonly believed that since interpreters have less control over their B language, their first foreign language, they would need to allocate more effort to the reformulation and production stages of interpretation when translating into their B language, thus making the task more difficult than translating into their A language (mother tongue), where most of the effort would be allocated to comprehending the source text. determining interpretation fluency and quality. Disfluency indices such as pause, fillers and repetition are used as quantitative measurements to analyze the quality of delivery in interpretation (Nation, 1989; Riggenbach, 1991; 楊承 淑, 2000). The existence of abundant or frequent pauses is considered a negative element of fluency, as pauses infer difficulties encountered in performing the complex task of interpreting.

Using disfluency parameters to measure interpretation quality is well justified as researches have shown that hesitation in interpretation output affect the listener’s perception of quality. Macias (2006) tested the theory in her experiment, where listeners were asked to rate the fluency of several videos with different levels of disfluencies. The control video was a German

26

speech and its Spanish simultaneous interpretation given by a professional interpreter. Test Video 1 and 2 each were inserted with an extra 13 and 20 silent pauses, and the silent pauses were categorized into three groups: 0.25 to 2 seconds, 2 to 4 seconds, and 4 to 6 second. Three groups of participants listened to one of the three videos each, then rated quality-related features of the interpretation in a questionnaire. The findings showed that when compared with the control video, videos with longer total pause duration and more 2 to 4 seconds pauses were given lower ratings for their fluency, indicating that pause is certainly considered a negative element of interpretation output by listeners.

Studies on pauses in interpretation employ similar global pause indices and methodologies with researches on psycholinguistic functions of pauses in spontaneous speech, which means they also share similar flaws. Distribution of pause is seldom discussed in interpreting studies, which may affect results of interpretation quality obtained through the calculation of production disfluencies. For instance, researchers have acknowledged that pauses in interpreting may also hold rhetorical and communicative values, and can be used by the interpreters tactically (Mead, 2000; Tissi, 2000). These pauses, which fit the description of juncture pause, may improve the listener’s comprehension, and in fact enhance the quality of the interpreting production.

However, when used as sub-parameter of interpretation fluency, such pauses are not differentiated from those that truly indicate production difficulties.

Thus the claim that frequent pausing in interpreting output directly indicates negative quality may not hold true without further probing and qualitative research.

27

To sum up, various pause studies have defined the location and functions of juncture pause and hesitation pause. Juncture pause tends to appear at syntactic clause boundaries in the speaker’s output. The speaker inserts juncture pause to emphasize the end of a syntactic unit and clarify the structure for the listeners. Juncture pauses also provide the speaker ample time to formulate the upcoming segment. Hesitation pause tends to appear elsewhere in the sentences. The occurrence of hesitation pause signifies the speaker is thinking on one’s feet and encountering difficulties in production.

Review of pause studies in spontaneous speech and interpretation reveals that examination of the speaker / interpreter’s immediate cognitive response at the moment of pause onset has yet to be explored; the location of pauses and their psycholinguistic functions are still in need of being analyzed together under empirical experiment. Furthermore, studies in both areas so far have only examined the pauses and their surrounding oral output contents for analyses. Therefore, incorporating another set of data for triangulation would be a good place to start on the journey towards unveiling the speaker / interpreter’s cognitive processes right at the moment of juncture and hesitation pauses.

This research proposes to employ the eye tracker, a cognitive psychology tool, in the analyses of pause phenomena during sight translation. Eye tracking data recorded during the onset of pauses could provide indications of cognitive activities in addition to the information obtained from the oral output of interpretation. Such triangulation of data can hopefully produce the empirical data to support the respective functions of juncture and hesitation pauses in interpretation.

28