Language Choice in Note-taking for C-E Consecutive Interpreting—An Empirical Study on Trainee Interpreters in China’s Mainland and Taiwan

Based on features of language in note-taking for C-E consecutive interpreting, this research studies and analyzes possible reasons behind the language choice of trainee interpreters in consecutive interpreting. The research combines quantitative and qualitative methods, and conducts a questionnaire survey, consecutive interpreting experiments and semi-structured interviews with 12 trainee interpreters in China’s Mainland and Taiwan. The research shows that notes in language forms account for the largest percentage of notes taken in consecutive interpreting, and that language choice in note-taking cannot be totally explained using the “source language vs. target language” or “A language vs. B language” dichotomy, but is also influenced by efficiency of handwriting and interpreters’ training profile.

The language used by interpreters to take notes has always been the subject of academic debates, but has little support from empirical data. Xu and Chai (2008, p. 123) believe that a significant difficulty of note-taking in consecutive interpreting is that the interpreter's writing and memory of audio messages interfere with each other, thus suppressing the phonological loop in working memory, affecting the accuracy of notes. Gile (1995) speculates that interpreters make language choice in note-taking to reduce the cognitive load in information processing.
In terms of what governs interpreters' choice of language in consecutive interpreting, there are two categories for discussion: the status of the language in the task (i.e., source languagetarget language), and the status of the language in the interpreter (i.e., A language -B language). Traditionally, scholars have used the language direction associated with the interpreting task to describe and explain the language choice of the interpreter's notes, i.e., using the source language or using the target language in note-taking. Dam's (2004) research, however, uses a new categorythe natural status of the language.
Through a study of five Danish-Spanish professional interpreters, Dam argues that interpreters tend to use their A language to take notes. Szabó's (2006) research on eight Hungarian-English professional interpreters, while supporting Dam's A-B language categorization, concludes that interpreters tend to take notes in their B language.
The above-mentioned literature has aroused more questions. One cannot help but ask: What is the language choice for interpreters for the English-Chinese language pair that European scholars have rarely paid attention to? One may use either the "source-target language" or "A-B language" category to describe note-taking in consecutive interpreting between English and Chinese, but will the different writing systems in English and Chinese influence interpreters' choice of language? Chinese researcher Chen (2015) conducted an empirical study based on Dam (2004) and Szabó (2006), and concludes that her subjects show a tendency of taking notes in the source language. This conclusion still fails to escape from the "source-target language" category, and thus does not live up to her claim to "break away from the debate on the conventional categories of note-taking language choice as either being source-target language or A-B language" (Chen, 2015, p. 146).

Rationale and Research Questions
In order to more accurately analyze the language choice of note-taking in consecutive interpreting between English and Chinese, and break away from the debate on conventional categories of note-taking language choice as either being source-target language or A-B language, we would argue for the need to approach this topic from the natural attributes of the language of the interpreter. Here in this research we shall introduce a new parameterthe writing systems of the Chinese language, ie.
traditional Chinese characters, which are mainly used in Taiwan, Hong Kong and Macau, and simplified Chinese characters, which are mainly used in China's Mainland and Singapore. As the writing system on China's Mainland and Taiwan differ, we hypothesize that interpreters across the Taiwan Straits show certain characteristics in the choice of language in note-taking, which is likely to be not totally relevant to the conventional demarcation of SL-TL or A-B.
This study therefore compares notes from interpreters of China's Mainland and Taiwan in a given consecutive interpreting task, describes, analyzes and tries to explain their language choice. The study aims to answer these questions: 1. What are the characteristics of CI note-taking by Mainland interpreters and Taiwanese interpreters?
For example, will they use more language-based notes or symbols?
2. In terms of language-based notes, do the two groups of interpreters show a general tendency of language choice for CI note-taking? To be more specific, do they choose Chinese or English? 3. What are the possible reasons for such language choice of the two groups of interpreters? In other words, we want to find out what possibly governs their language choice.

Research Method
This study combines qualitative methods with quantitative ones, using questionnaires, experiments, and semi-structured interviews. The questionnaire is mainly used to collect basic information about the subjects, including their personal information, educational background, and training and work experience of interpreting. All subjects take part in an English-Chinese consecutive interpreting test, after which semi-structured interviews are conducted to find out more about their note-taking habits and approach.

Subjects
A total of 12 trainee interpreters participated in the study. They are from Mainland Chinese institutions (6 students from Guangdong University of Foreign Studies) and Taiwanese institutions (3 from National Taiwan Normal University and 3 from Fu Jen University). Among the 6 Mainland students, 2 are male and 4 are female, with an average age of 24 years; 4 of the 6 Taiwanese students are male and 2 female, with an average age of 25 years. All of them are native speakers of Chinese and have English as their B language.
All subjects have passed rigorous entrance tests and have received professional training in a Master's program of conference interpreting at their respective institutions. By the time of the experiment, they had had three to five semesters of CI training, and all had certain interpreting work experience.
According to the questionnaire results, the six Taiwanese students had passed one or more international English proficiency tests; all six Mainland students had passed at least one international English proficiency test and interpretation level two of CATTI (China Accreditation Test for Translators and Interpreters).

Questionnaire
The questionnaire consists of three parts. The first part contains 6 questions on the training background and language pair of the subjects; the second part contains 2 questions on the education background and English proficiency of the subjects; the third part contains 4 questions on the CI training and work experience of the subjects.

Test Materials
The selection of the consecutive interpreting test materials followed four principles: (1) the material should have a certain length and difficulty, so that the subjects have the need to take notes; (2) the material should be on a general theme, with no technical contents; (3) The material should be an audio-visual recording of a live speech, and be as close to a real-life interpreting task as possible; (4) the speech should have a moderate delivery rate of 100 to 120 English words per minute (Gerver, 1969;Seleskovitch, 1978).
The material selected was Paragraph 1 to 6 of former U.S. Secretary of State Hillary Clinton's speech at the 2012 UN Conference on Sustainable Development. The excerpt was 5 minutes and 7 seconds, 531 words, and the delivery rate was 104 words per minute. The speech is divided into two parts. The first part is 278 words, 2 minutes and 36 seconds; the second part is 253 words, 2 minutes and 31 seconds. Each part is divided into three segments.

Data Collection and Analysis
The experiment was conducted in the interpretation laboratory at the three institutions. It took each subject about 50 minutes to complete the experiment. The whole process is divided into pre-interpreting stage, interpreting and post-interpreting stage. In the pre-interpreting stage, each subject was briefed on the details about the task and filled out a questionnaire. Then, the background information and vocabulary of interpreting task was distributed, as well as blank A4-size paper for note-taking. In the interpreting stage, each student completed the test one by one, and the whole process was recorded. In the post-interpreting stage, the subjects participated in a semi-structured interview and the entire process was recorded. The interview recordings were transcribed, and together with the collected notes, were repeatedly studied and analyzed in two groups, namely the Mainland Group and the Taiwan Group.

General Features of Notes
We first counted the notes contents of each of the 12 trainees. After reading their notes, we found that the contents of the notes generally consist of six types of elements: complete English words, complete Chinese characters, symbols, English abbreviations, meaningless scribbles, and unrecognizable notes.
The above six types, according to the functions of notes, can be categorized as: language-based notes (including complete characters, words and abbreviations), logic links (a dividing line indicating coherence of discourse, an arrow, a mark, etc.), drawings (general or personalized symbols or codes with signifying functions), scribbles (they are meaningless, but many subjects in the interview said that sometimes they will scribble something that they do not understand, simply as a way to relieve the cognitive pressure in the process of interpreting), and unrecognizable notes (they are not scribbles, but the researcher cannot recognize them, nor are the subjects able to read them in interpreting or recall them in the follow-up interview, and therefore can be considered notes with no actual functions. In their study on the difficulty of note-taking in consecutive interpreting, Xu and Chai (2008) classify this type of notes as "ill format notes"). We take each information unit with independent and complete meaning as a unit of notes, and count the number and distribution of various types of notes. As can be seen in the two tables, in general, Mainland students take more notes than Taiwanese students. In the semi-structured interviews, the Taiwan Group reported that their interpreting trainers seldom assigned whole classroom sessions to note-taking teaching, but mentioned from time to time in the training that taking notes is only a tool to assist memory and information processing, and not imperative. Interpreters may or may not take notes. In contrast, trainers in institutions on China's Mainland, as the interviews show, not only teach the skills of note-taking, analyze samples of notes, but also ask students to demonstrate and discuss their notes in class, and encourage students to develop their most suited note-taking system. We therefore tentatively conclude that the Mainland Group show a tendency of taking more notes than the Taiwan Group, probably because of some differences in ideas and operations in consecutive interpreting teaching across the Taiwan Straits.
Despite the differences in the number of notes, the two groups show similar characteristics. The lowest proportion of language-based notes is 40.7% (S12) and the highest is 86.2% (S5). Except one subject, all are close to or above 50%. This shows that notes of most subjects are predominantly language-based.
Logic links are in second place. In the interview, some of the subjects reported that the logic links in their notes depend mainly on the speaker's speed of delivery and their understanding of the speech. To be specific, if the speaker's delivery is moderate (or even slow) for the subject, they will have time to note down the logic links of the presentation. But if the speaker is fast, they will use most of their effort for comprehension, and write down anything they are able to, without additional effort to note down the logic relations in the speech, and will only re-organize the speech in the target language output stage. On the other hand, if the subjects do not have any difficulty in understanding the speech, they will have the effort to write down the logic relations in note-taking. But if the information is dense or too difficult, they will focus on listening and comprehension, and have little effort for note-taking of the logic links. This characteristic can be explained with Gile's Effort Model (1995).

Language Choice in Notes
We further studied these language-based notes and divided them into English words and abbreviations and Chinese characters for a general tendency of the language choice of note-taking by the subjects. Statistics show that the proportion of English notes is clearly higher with the Taiwan Group than with the Mainland Group. Even the lowest proportion of 51.6% (S4) is larger than the highest proportion in Mainland interpreter of 50% (S10). If the source language -target language" categorization is used, Taiwanese interpreters tend to take notes more in the source language, while the Mainland interpreters show a tendency of taking notes in the target language. If the "A language -B language" category that Dam (2004) proposes is used, then Taiwanese interpreters tend to take notes more in their B language, while the Mainland interpreters show a tendency of taking notes in their A language. As both the status of the language in the task and the status of the language in the interpreter are identical for the two groups, the fact that they show a different tendency of language choice is something of interest to the researchers.
We then inquired about the subjects' tendency of language choice in note-taking with open-ended questions in the subsequent semi-structured interviews. In the interviews, we avoided using the expressions such as "source language", "target language", "native language" and "foreign language", but only used "Chinese" and "English", so that the subjects can make their own choice of the categorization of the two languages. The subjects' answers are summarized in the table below.

No. Language choice Reason
Taiwanese interpreters S1 EN Because the language of the speaker is what I heard directly.

S2
EN I need to focus on listening and comprehension.

S3
EN Chinese characters have too many strokes to be written fast.

N/A
No matter what language, I tend to write down what first comes to me.

EN
It is faster to write in English, so I can take more notes.

S6
N/A I don't have a clear preference, for both SL and TL have their pros and cons.

S7
ZH If possible, I tend to convert the language and then take notes in the target language.

S8
ZH I try to use the target language as much as possible. But if I'm not quick enough, I will turn to the source language.

S9
ZH I take notes in my mother language, for I can react more quickly.

S10 N/A I don't have a clear preference.
S11 ZH I guess one should use the language that's closer to the interpreter's production, so I use the Chinese language. S12 N/A I don't know, but I think symbols are more effective than characters.
It can be seen from the answers of the subjects' self-reported language choice is consistent with the actual language tendency observed in their notes. Four subjects report that they don't have a clear preference, and their notes show more or less the same proportion in English and Chinese, which indicate that their choice of language in note-taking is quite random. S12 reports that he was more inclined to use symbols. Statistics in Table 1 show that he did use the largest number of symbols among all subjects (29.1%), which, together with logic links (30.2%), exceed language-based notes (40.7%).
For the subjects that show a tendency toward certain language choice, when asked why they use Chinese or English in note-taking, they gave answers that point to different categories. It can be seen that some subjects approach their language choice from the "source language -target language" categorization. For example, S1 mentions "the language of the speaker", and S11 mentions "the language closer to the interpreter's production". Some subjects even use the terms "source language" and "target language" (S6, S7, S8). Some subjects, on the other hand, approach their language choice from the "A language -B language" categorization, for example, "mother language" (S9). It should be noted that the reasons that S3 and S5 give relate to the complexity of the writing system of the language.
We then asked the subjects open questions, to inquire about the advice they had from their interpreting trainers, if any.

S8
TL There is some influence. I force myself to use TL more.

S9
TL I'm not influenced by the trainer's advice. I take notes in my own way. S10 N/A At first, the trainer suggested we use more symbols. Later, we were advised to choose a language that works for us. S11 TL I am influenced. I use the target language as much as possible.
S12 TL There is certain influence. However, the influence is minimal in interpreting from English to Chinese. The reason is that I tend to use lots of symbols in E to C consecutive interpreting to improve the efficiency. As for C-E interpreting, I prefer note-taking in the target language, for listening and comprehension are not challenging to me.
It can be seen that more emphasis is given to the teaching of note-taking for consecutive interpreting in professional programs on China's Mainland. Trainers generally recommend note-taking in the target language, which is consistent with the views of scholars such as Herbert (1952), Rozan (1956), Seleskovitch and Lederer (1989), Mikkleson (1983), and Jones (1998), and AIIC recommendations (1994). All the subjects in the Mainland Group are students of the same institution, thus we do not know from this study whether interpreting trainers of other Mainland institutions also recommend note-taking in the target language. What is worth noticing is that "target language" here has nothing to do with the natural status of the language. In other words, it may be the A language of the students (Chinese), or the B language of them (English), depending on the actual direction of language conversion in consecutive interpreting. In contrast, interpreting trainers in Taiwanese institutions lay less stress on note-taking in consecutive interpreting. Students form a habit of note-taking more out of their experience-based intuition, rather than from principles or recommendations from their trainers.

Conclusion
From the above discussion, we can conclude that generally speaking, notes taken by subjects in their consecutive interpreting are predominantly language-based, followed by logic links and symbols.
Mainland interpreters tend to take more notes than Taiwanese interpreters. In the E-C direction, Mainland interpreters show an overall tendency of taking notes in Chinese, while Taiwanese interpreters take notes more in English.
Such findings seem to suggest a phenomenon that cannot be explained by previous researchers. In this study, the two groups of subjects have exactly the same directionality (E-C) and language combination (A-ZH and B-EN), and should therefore show the same tendency of language choice in consecutive interpreting. The explanations from the subjects in the interviews also mix the two categories of "source language -target language" and "A language -B language". We therefore argue that interpreters' language choice cannot be explained totally with the "source language -target language" language choice may be factors other than the status of the language in the task or the status of the language in the interpreter.
We suggest that there are two other factors that may influence the interpreters' language choice: (1) Ease of writing. One of the parameters introduced to this study is different writing systems across the Taiwan Straits. As mentioned in the semi-structured interviews by the Taiwanese Group, with more strokes than simplified Chinese characters, the same traditional Chinese characters are more complicated to write. In order to save cognitive efforts and processing time, interpreters may drop Chinese characters and use English words, abbreviations or symbols instead. (2) Training profile. This study shows that interpreting trainers in Taiwan do not lay much stress on note-taking in consecutive interpreting, while Mainland trainers emphasize note-taking much more. Language choice of interpreters may well be influenced by their training, which can be seen in their notes as well as answers to the interview questions. The Taiwanese Group therefore shows greater arbitrariness and intuition in their note-taking.
This study has a certain significance for the interpreting teaching. Not concerned with the optimum choice or recommendation, the study may have offered a new way of explaining language choice that differs from the convention categories. The study is limited in its subjects, sample size and interpreting directionality. A few possible directions for future research include: (1) enlisting the cooperation of professional interpreters in the Chinese Mainland and Taiwan; (2) testing the other direction of C-E; and (3) increasing the number of subjects for a larger sample size.