CHAPTER 3 DATA AND METHODOLOGY
3.7 Interim summery
A gesture is synthetic in that several features composing a single gesture should interact with each other to represent a certain viewpoint. Five gestural features have been identified as crucial criteria, including gestural space, handedness, stroke duration, frequency, and the involvement of other parts of the body. How these gestural features indeed collaborate with each other to represent different gestural viewpoints and instantiations of each viewpoint will be discussed in Chapter 5.
3.8 Summary
In this chapter, data and methodology adopted in this study are introduced. With regard to the data, it is from five conversations in the sub-corpus of Mandarin of NCCU Corpus of Spoken Chinese. Clausal events are selected from the five conversations, where each clausal event conveys a complete proposition concerning events that have happened to someone. Only third-person past events are chosen to accommodate the purpose of this study. The framework of this study utilizes Koven’s (2002) framework of speaker role inhabitance and McNeill’s (1992) notion of character and observer viewpoint, and therefore defines three viewpoints—speaker, character, and observer viewpoint that can arise in a description of third-person past events within conversational contexts. Linguistic and gestural representations of three
viewpoints are then examined qualitatively as coding principles for quantitative study.
With respect to linguistic representations, syntactic structures, including sentence forms, phrasal expressions and lexical items, and paralinguistic devices, such as discourse markers, prosodic features and laughter, could all serve as linguistic resources to express a certain viewpoint. For gestural representations, five gestural features are identified as influencing factors of gestural viewpoints, including gestural space, handedness, stroke duration, frequency and the involvement of other parts of the body. Through different performances of the interaction and collaboration between these gestural features, a gesture might also be able to represent the three viewpoints.
In Chapter 4, the next chapter, we shall see how speakers represent speaker, observer, and character viewpoint in talking about third-person past events by making use of different linguistic representations which shall be introduced in this chapter first. In Chapter 5, we will then see how the three viewpoints are represented through the gestural channel in the interaction and collaboration of the different gestural features identified in this chapter for use as criteria for analysis of the quantitative study of the gestural viewpoints within the description of third-person past events.
95
LINGUISTIC REPRESENTATIONS OF VIEWPOINTS
The present study analyzes the linguistic and gestural viewpoints of 119 clauses.
Based on a framework which identifies speaker, observer and character viewpoint developed in the present study, and the qualitative study of the linguistic and gestural representations of each viewpoint as coding principles, this chapter presents the results of the quantitative study of linguistic viewpoints. The distribution of linguistic viewpoints in the descriptions of third-person past events within conversational contexts is shown in 4.1. The distribution of different linguistic or paralinguistic devices that speakers use to represent speaker, observer and character viewpoint is discussed in sections 4.2, 4.3, and 4.4. Section 4.5 is a brief summary.
4.1 Quantitative study of linguistic viewpoints
The quantitative study of linguistic viewpoints suggests how speakers in talking about third-person past events within a conversation often make use of speaker, observer and character viewpoint. The distribution of each viewpoint in language is presented in Table 1.
Table 1. The distribution of linguistic viewpoints
Types of viewpoint Number of each viewpoint Total
S-VPT 40 33.6%
O-VPT 72 60.5%
C-VPT 7 5.9%
Total 119 100%
From Table 1, we can see that when talking about third-person past events within a conversation, speakers most often represent an observer viewpoint. Speakers frequently choose to simply describe events about others by making use of plain statements that reveal the spatio-temporal information, the persons involved (characters of the events), and the characters’ actions or deeds concerning the events, or indirect reported speech that plainly reports what characters in the event have said.
Despite the fact that most of the time speakers act purely as outsiders who are not involved in the original events, speakers are also concerned about their current role as a speaker in an ongoing conversation while describing the events. Therefore, speakers on occasion also convey speaker viewpoint and reveal their here-and-now relationships with other co-conversationalists by interacting with other speakers or making personal comments (S-VPT, 33.6%). What speakers do only infrequently is to step into the scene of the original events, enacting the roles of the characters of the events and performing their speech or thoughts. This again suggests that within a conversation, speakers usually want to talk about a third-person past event from the perspective of an outside-event observer, and keep some distance from the described
events rather than step into it.
Speakers in talking about these events also produce speech-accompanying gestures. Following McNeill’s idea that “one area of meaning where speech and gesture are coexpressive is the point of view” (1992:118), we might hypothesize that gestures often collaborate with the accompanying speech in expressing the same viewpoint of the same event. Therefore, we might expect that observer viewpoint will also be commonly adopted in gesture, as the same for language in describing the same event. On the other hand, since speakers rarely enact characters’ speech in talking
about third-person events, we would also expect that a speaker’s enactment of a character’s actions or deeds represented in gesture to be infrequent. Whether this
hypothesis concerning the distribution of gestural viewpoints conforms to the actual findings of the current study will be seen in the discussion in the following chapter on the gestural representations of viewpoints.
In the following sections, we will see how speakers make use of different linguistic representations in representing the three viewpoints in language. In other words, we will see the distribution of the linguistic representations in each viewpoint.
The distribution of linguistic representations for speaker viewpoint is discussed in section 4.2, followed by a discussion on observer and character viewpoint in section 4.3 and 4.4.
4.2 The distribution of linguistic representations of speaker viewpoint
In representing the speaker viewpoint in the descriptions of a third-person past event within an ongoing conversation, speakers are showing their concern about the maintenance of the conversation. They notice the fact that they are not only describing an event to others, but at the same time are also involved in a conversation that must be carried on. Therefore, speakers might reveal their here-and-now relationship with other co-conversationalists in the conversations by interacting with other speakers, revealing their attitudes, or making personal comments in their current role as speaker to show their concern in maintaining the conversation. In terms of the linguistic representations speakers can adopt to represent the speaker viewpoint, speakers can make use of interrogative sentences, impersonal use of second-person pronouns, or explicit appeal to another speaker to interact with or gain identifications from their co-conversationalists. They can also use speculative or suggestive expressions, parenthetical remarks and evaluative or emotive expressions to show their attitudes toward or make comments on the past events. Utterance-final discourse markers and laughter also serve to indicate speakers’ feelings on the events being talked about. The distribution of these linguistic representations for speaker viewpoint is shown in Table 2.
Table 2. The distribution of linguistic representations of speaker viewpoint
Speaker viewpoint
Linguistic representations Number Example
Interrogative sentences 5 10.4% Example (1) in Chapter 3
Speculative expressions 10 20.8% Example (2) in Chapter 3
Suggestive expressions 2 4.2% Example (3) in Chapter 3
Parenthetical remarks 3 6.2% Example (4) in Chapter 3
Evaluative and emotive expressions 8 16.7% Example (5) in Chapter 3 Impersonal use of second-person pronouns 2 4.2% Example (6) in Chapter 3 Explicit appeal to another speaker 1 2.1% Example (7) in Chapter 3
Discourse markers 11 22.9% Example (8) in Chapter 3
Laughter 6 12.5% Example (9) in Chapter 3
Total 48 100%
Two devices used within the same event 8 16.7%
From Table 2, we can see that the representations of speaker viewpoint that show speakers’ current status as a speaker within an ongoing conversation and their concern about enabling the conversation to continue are mainly achieved by expressing their attitudes or comments while talking about others’ events. Despite the fact that speakers can also maintain a conversation by interacting with other speakers by using interrogative sentences, explicit or impersonal appeals to gain identification from other speakers, they seldom do so.
Speakers also make use of two kinds of linguistic or paralinguistic devices within the descriptions of the same event to mark their representations of speaker viewpoint.
Most of the combinations include a discourse marker (four cases) and an evaluative or emotive expression (four cases) that explicitly show speakers’ attitudes or evaluations
of the events.
From the distribution of the linguistic representations of speaker viewpoint, we can then hypothesize that speakers’ gestures in representing speaker viewpoint will also collaborate with the accompanying speech mainly to express speakers’ attitudes or comments rather than be those used for interacting with other speakers within the conversations.
4.3 The distribution of linguistic representations of observer viewpoint
Concerning the linguistic representations of observer viewpoint, speakers in talking about third-person past events make use of indirect reported speech and plain statements to simply report related information concerning the events. Speakers describe the events as an outside-the-event observer, without involving themselves in the descriptions of the events or making personal comments. Table 3 presents how speakers make use of these two linguistic structures to convey observer viewpoint.
Table 3. The distribution of linguistic representations of observer viewpoint
Observer viewpoint
Linguistic representations Number Example
Indirect reported speech 5 6.9% Example (12) in Chapter 3
Plain statements 67 93.1% ---
Total 72 100%
From Table 3, we can see that the use of plain statements accounts for nearly all
linguistic representations of observer viewpoint. Moreover, among all linguistic representations of three kinds of viewpoints, plain statements are also the most commonly adopted linguistic structures in the descriptions of third-person past events.
This suggests that while speakers in talking about third-person past events have various linguistic structures or paralinguistic devices that can be used as a way of describing a third-person past event (which therefore results in different viewpoint representations), such as enacting characters’ speech, injecting personal emotions, or making explicit comments, speakers nevertheless merely use language to plainly describe the past events as an outsider.
In McNeill’s gestural study on viewpoints in narrative data, he found that 60% of iconic gestures have observer viewpoint (O-VPT) and 40% have character viewpoint (C-VPT). The percentage of the occurrences of observer viewpoint in gesture is higher than character viewpoint. From McNeill’s study and the distribution of linguistic representations of viewpoints in the current study, we might then expect that observer viewpoint might also be the most frequently adopted in gesture as it is in language. Collaborating with the accompanying speech, the use of gesture provides another channel for speakers to plainly describe the past events like an outsider without stepping into the original scene.
4.4 The distribution of linguistic representations of character viewpoint
Speakers can make use of direct speech, voiced direct reported speech, or inner speech to enact different characters’ speech or thoughts in talking about third-person past events. In this case, speakers act as characters in the original scene of the past events, performing their speech or thoughts. The distribution of the linguistic representations of character viewpoint is shown in Table 4.
Table 4. The distribution of linguistic representations of character viewpoint
Character viewpoint
Linguistic representations Number Example
Direct speech 3 42.9% Example (15) in Chapter 3
Voiced direct reported speech 3 42.9% Example (16) in Chapter 3
Inner speech 1 14.2% Example (17) in Chapter 3
Total 7 100%
From Table 4, we can see that linguistic representations of character viewpoint are infrequent in the current data. Speakers in talking about third-person past events seldom attempt to enact characters’ speech or thought as a way of making a description. If gestures often collaborate with the accompanying speech in expressing the same viewpoint, we would then hypothesize that the expression of character viewpoint in gesture is also rare. If this is the case in the current data, we might suggest that the situation when speakers are talking about past events in conversations to be the same as found by McNeill in that observer viewpoint is more frequently seen than character viewpoint in gesture when people are narrating stories. Whether the
hypothesis that character viewpoint in gesture is infrequent and uncommon is right or wrong will be looked into and discussed in the following chapter on gestural representations of viewpoints.
4.5 Summary
This chapter presents the quantitative analysis of linguistic viewpoints. The distribution of the three viewpoints suggests that observer viewpoint is the most common choice for speakers to use to talk about third-person past events in ongoing conversations, while character viewpoint is the most infrequent. The distribution of the linguistic representations concerning each viewpoint not only suggests how speakers make use of different linguistic or paralinguistic devices as resources to represent each viewpoint, but also suggests that plain statements are the most unmarked way for speakers to talk about third-person past events. With the distribution of each linguistic viewpoint and of the different linguistic representations used within each viewpoint, this chapter presents the performance of viewpoints in language when speakers are talking about third-person past events within conversational contexts. In hypothesizing that gestures might collaborate with the accompanying speech in expressing the same viewpoint, we would expect a similar pattern for the distribution of each viewpoint in gesture. Whether this hypothesis
holds for the current data will be presented and discussed in the following chapter on gestural representations of viewpoints.
105
CHAPTER 5
GESTURAL REPRESENTATIONS OF VIEWPOINTS
This chapter presents a quantitative study of gestural viewpoints and the collaborative expressions of viewpoints in language and gesture. In the interaction and collaboration between different gestural features, a speaker’s use of embodied gestures can express speaker, observer and character viewpoint. The ways in which the five gestural features identified in the current study—gestural space, handedness, stroke duration, frequency, and the involvement of other parts of the body within the performance of a gesture interact and collaborate with each other in representing different viewpoints could serve as important criteria in identifying and determining gestural viewpoints are introduced in section 5.1. The distributions of the three viewpoints in the gestural channel in the descriptions of third-person past events in conversational contexts are then presented in section 5.2. Three gestural instantiations of each viewpoint are then given in section 5.3. Gestures types produced in the current data in representing viewpoints are discussed in 5.4. A brief summary is given in section 5.5. Section 5.6 discusses the collaborative expressions of viewpoints in linguistic and gestural channel. A summary for this chapter is given in section 5.7.
5.1 Gestural features and gestural viewpoints
In the current study, we have recognized five gestural features—gestural space, handedness, stroke duration, frequency, and the involvement of other parts of the body as five criteria for use in analysis in determining different gestural viewpoints.
How the five gestural features interact and collaborate with each other in representing the three viewpoints, and how these gestural features could serve as distinctive criteria in identifying different viewpoints are discussed in this section. In particular, we will see how the performance of each gestural feature can be used to distinguish between observer and character viewpoint, due to the limited data of speaker viewpoint in gesture in the current data. Table 5 shows how gestures of observer and character viewpoint are performed with reference to the five gestural features when each gestural feature is analyzed as showing more gestural complexity: the hand movement involves Center and Periphery space, use of both hands, longer stroke duration, repetition of the same stroke, and the involvement of other parts of the body. In Chapter 3, it was suggested that a speaker’s gesture often shows more gestural complexity when representing character viewpoint, since speakers are using their hand movements to enact characters’ actions or deeds. We will see whether C-VPT gestures produced in the current data indeed perform more complex gestural features than O-VPT gestures. Table 6 further shows the number of complex gestural features
used in C-VPT and in O-VPT gestures.
Table 5. Gestural features that show more gestural complexity
Viewpoint
*The percentages that are shaded in grey refer to each (complex) gestural feature used between C-VPT and O-VPT gestures; the percentages shown without the grey shading refer to the five (complex) gestural features used within a C-VPT or an O-VPT gesture.
Table 6.
Number of complex gestural features used in C-VPT and O-VPT gestures Viewpoint between C-VPT and O-VPT gestures; the percentages shown without the grey shading refer to the five (complex) gestural features used within a C-VPT or an O-VPT gesture.From Table 5, we can see that the percentage of each gestural feature which
shows more gestural complexity in C-VPT gestures is much higher than that in O-VPT. Table 6 further shows the number of complex gestural features that are used in two types of gestures, and suggests that the percentage of C-VPT gestures that are composed of two or more than two complex gestural features is higher than that of O-VPT gestures. In addition, the use of three or more than three complex gestural features is only seen in C-VPT gestures but not in O-VPT gestures. This suggests that C-VPT gesture indeed is more possible to be composed of gestural features that show greater gestural complexity than O-VPT gesture, and that a C-VPT gesture is composed of more of these complex gestural features than O-VPT gesture to be identified as representing character viewpoint. O-VPT gesture might show none of these gestural features that show greater gestural complexity (12 out of total 52 cases of O-VPT), or at most two complex gestural features. To sum up, speakers in representing character viewpoint through gesture perform gestural features that show more gestural complexity and make use of a greater number of these complex gestural features within a C-VPT gesture than an O-VPT gesture.
At the same time, we can also see how gestures of observer and character viewpoint are performed with reference to the five gestural features when each gestural feature is analyzed as showing less gestural complexity. It was suggested in Chapter 3 that speakers plainly depict the past events like an observer that does not
attempt to walk into the original scene of the past events in representing observer viewpoint. As a result, speakers tend to produce a “simpler” gesture, which means that the gesture shows less gestural complexity, when representing observer viewpoint. A simple gesture that shows less gestural complexity might have the following gestural features: the gestural space is confined to the Center-Center space, use of a single hand, shorter stroke duration, no repetition of the same stroke, and no involvement of other body parts. Table 7 shows the percentage of each gestural feature that shows less gestural complexity used in C-VPT and O-VPT gestures:
Table 7. Gestural features that show less gestural complexity
Viewpoint
In Table 5, we have seen that the percentage of each gestural feature that shows more gestural complexity in C-VPT gestures is apparently higher than that in O-VPT gestures. However, the distinctive difference between C-VPT and O-VPT gesture on
the percentage of the use of gestural features that show simpler and less gestural complexity is not at all obvious, as shown in Table 7 (for example, 49.5% of C-VPT and 50.5% of O-VPT concerning the feature “no involvement of other parts of the body”). Only the results for the performance of shorter stroke duration show a
the percentage of the use of gestural features that show simpler and less gestural complexity is not at all obvious, as shown in Table 7 (for example, 49.5% of C-VPT and 50.5% of O-VPT concerning the feature “no involvement of other parts of the body”). Only the results for the performance of shorter stroke duration show a