McNeill and the gestural manifestations of viewpoints (1992)

In his Hand and Mind (1992), McNeill suggested that gestures, unconstrained by language-like conventions, can not only convey semantic meanings, but also express pragmatic content. In examining the relationship of gesture to ongoing, real-time storytelling, McNeill also suggested that viewpoint is one area of meanings where gesture can co-express with speech. Since McNeill’s main focus is on narrative data, he denoted the notion of viewpoint as “the feeling of distance from the narrative”

(1992:118). A given narrative can be portrayed as if it were being experienced, or as if it were being seen from a distance. Two viewpoints—the character viewpoint (C-VPT) and observer viewpoint (O-VPT) are therefore recognized. In addition, definitions of C-VPT and O-VPT are described on the basis of the different performances of iconic gestures. In representing C-VPT, the iconic gesture may seem to re-enact the character, and the depiction is dispersed over the narrator’s body in an appropriate way. The narrator’s hand plays the part of the character’s hand, and the narrator’s body enacts the part of the character’s body. On the other hand, in representing O-VPT, the gesture may appear to display an event such as showing that the narrator seeks to keep some distance from the story. The depiction of characters is concentrated in the hand, and the narrator’s body is an onlooker outside the narrative. In addition to how the characters of a narrative are represented through iconic gestures could be a criterion in distinguishing C-VPT or O-VPT, gestural space is also a key element in distinguishing these two viewpoints. A C-VPT gesture incorporates the speaker’s body into the gestural space, and the speaker’s hands represent the hands of the character. An O-VPT gesture in contrast, excludes the speaker’s body from the gestural space and his hands play the part of the character as a whole. McNeill further suggested that in the narrative data he studied, 60% of iconic gestures have an O-VPT, and 40% a C-VPT.

Despite the fact that McNeill defined viewpoints from a gestural perspective, his notion of character viewpoint is equal to that of Koven’s (2002) character role

inhabitance defined from a linguistic perspective, in which both are concerned with

the speakers’ attempts to enact the characters of the narratives whether through the speech or gestural channel. On the other hand, McNeill’s observer viewpoint also can be equated to Koven’s authorial role inhabitance. In Koven’s authorial role

inhabitance, speakers are storytellers or narrators outside of the event, which is also

the core value of McNeill’s observer viewpoint despite the fact that the channels used to express the viewpoints are different.

Indeed, McNeill’s purpose in studying gestural viewpoints is to see how gesture can reflect the discourse structure as narrators proceed through the narrative. Different kinds of gesture or gesture performances might reveal the narrative structure, and the representation of gestural viewpoints also serves as a clue. According to McNeill, storytelling refers to the entire set of events that make up the conveying of a story by one person to another, and it is structured on multiple levels. References to events or incidents from the fictive world of the story are called the narrative level. Sentences at this level follow a certain order that together compose the story line, which is called the temporal constraint, so that listeners are able to understand that those sentences are part of the story. In addition to narrate the story plot, narrators also make explicit

references to the structure of the story. Clauses present the story about the story that are interwoven within the narrative level constitute the meta-narrative level. For example, in McNeill’s narrative data, a narrator says and uh the first scene you see is

uh to comment on the story structure and manipulate the story as whole a unit.

Storytellers also make references to their own experience to the event of storytelling itself, and this is called the para-narrative level. At this level, narrators speak for themselves and try to focus on their relationship to listeners. For example, they might ask listeners whether they have ever heard any story similar to the one that is being told before.

In identifying these three narrative levels—narrative level, meta-narrative level, and para-narrative level, McNeill pointed out that different kinds of gesture tend to appear in one level over others. Iconic gestures in most of the cases appear at the narrative level, where they exhibit a series of story events. C-VPT and O-VPT thus are most often observed in the narrative level with the frequent occurrences of iconic gesture.

In addition to the observation that C-VPT tends to appear in the narrative level and events of greater importance, the shifts between C-VPT and O-VPT within a narration are not at all random. McNeill followed Church’s (Church et al. 1989) classifications of event as central or peripheral, and suggested that more central events

are given the C-VPT; while more peripheral ones are given the O-VPT. Central events can be either the initiation of goal actions, main goal actions, or outcomes of goal actions; while peripheral ones include setting statements, subordinate actions, responses to actions and outcomes. Two viewpoints appear in different contexts, on the basis of these definitions regarding central or peripheral events. McNeill provided statistics to show this—71% of C-VPT gestures appeared with central events, and 93% percent of O-VPT gestures appeared with peripheral events. A speaker’s narration which comes from McNeill’s narrative data can serve as an example to illustrate. The narrator is describing a series of events of a scene from a cartoon, in which one character Sylvester, tries to reach a second character Tweety, by climbing up the inside of a drainpipe. Two successive events about one of the characters Sylvester—he [comes out the bot]tom of the drainpipe, and he’s [got this] big bowling

ball inside him are described from different viewpoints. As the narrator talked about

‘come out’, he made an O-VPT gesture. Since the first event was considered as the

intervening event that was used to support the more important upcoming event, it got the O-VPT. As for the second event, it concerned the cause and effect of the whole story and was considered as the main point of the scene. Thus, it got the C-VPT.

Many other examples in McNeill’s narrative data support this finding. In McNeill’s data, though all of the gestures were iconics and all of the clauses were at the

narrative level, the events were not equal in terms of their importance for the advancing of the story line. Gestures with C-VPT or O-VPT can suggest this difference and the study of these gestures can be used to study the narrative structure.

While McNeill studied gestural viewpoints within narrative data, he also provided linguistic manifestations of viewpoints. In his quantitative study, he generalized that the C-VPT gesture tends to appear with transitive verbs and single clause sentences, and the O-VPT gesture with intransitive or stative verbs and multi-clause sentences (complex sentence). He further suggested that it is through gestural manifestation that linguistic manifestation may be made obvious and clear.

Without the speech-accompanying gestures that represent different viewpoints, these linguistic forms that are also capable of conveying viewpoints are unobvious and it is hard to draw any attention to them.

According to McNeill’s study, therefore, gestural and linguistic manifestations achieve equivalent results in terms of ‘distance’ for a given viewpoint. In the C-VPT gesture, speakers walk into the scene of the narrative and insert themselves into the gestural space. In the linguistic parallel there are single clauses and transitive verbs that also function to represent the speakers as walking into the events narrated. With a single clause, “there is a minimal grammatical separation of the event from the speaker” (1992:120). With a transitive verb, the events being narrated are as if under

magnification. A typical example given by McNeill as an illustration of a C-VPT is

and drops it down the drainpipe. In the gesture, the narrator’s hand appeared to grasp

the bowling ball and shove it down the drainpipe, which description is provided from

the viewpoint of the character, and the action is accompanied in language by a single clause and a transitive verb ‘drop’. In terms of an O-VPT in gesture, speakers exclude

themselves from the gestural space to express that they are outside of the events narrated. Linguistically, the O-VPT correspondingly appears with complex sentences and intransitive or stative verbs which allow speakers to keep some distance from the events described. Complex sentences, according to McNeill, “interpose distance from the action in its own configuration” (1992:120). One example cited by McNeill is he

tries climbing up the side of the building, where the act of climbing is presented in the

embedded clause and the upper clause (he tries) implies that the action is seen from the point of view of an outside observer of the story. While McNeill acknowledged that the correspondence between gestural and linguistic manifestations with regard to the same viewpoint is not always definite, the speech-gesture co-expressions of the same viewpoint are pretty much correspondent in terms of the distance from the narratives.

The present study does not analyze transitive/instransitive or stative verbs and single clause/multi-clause sentences as linguistic manifestations of viewpoints, as

suggested by McNeill. In McNeill’s study, linguistic and gestural representations of viewpoints are not independently analyzed. In addition, linguistic manifestations of certain viewpoints are distinguishable and obvious only when the gestural representations are first identified. The present study, different from McNeill’s analysis, analyzes linguistic manifestations of viewpoints by observing and analyzing various linguistic structures and paralinguistic devices that could possibly represent different viewpoints that speakers make use of in talking about third-person past events within conversational contexts, rather than deducing them from gestural representations.

To sum up, McNeill brought up the notion of viewpoint in his study on gestures within narratives, and proved that viewpoint can not only be conveyed through language, but also can be manifested through embodied channels like gestures.

However, since McNeill’s focus is on gestures, his definition of viewpoint is primarily and purely seen only in regard to the gestural aspects. In other words, he provided a definition of the notions of character and observer viewpoint by describing different gestural performances. The problem in this approach is that linguistic manifestations are presented as merely additive correspondents with speech-accompanying gestures, rather than independent realizations of a certain viewpoint. McNeill is using gestural definition to deduce linguistic manifestations, instead of directly analyzing linguistic

devices that could possibly convey viewpoints. In other words, McNeill has not systematically studied the collaboration between linguistic and gestural representations in expressing viewpoints.

In addition, McNeill focused his study of viewpoint only on iconic gestures, overlooking the capabilities of other gesture types in expressing viewpoints. Despite the fact the performance of other gesture types to represent viewpoints might not be as multi-dimensional as iconic gestures, other gesture types like metaphoric, deictic and spatial gestures are also potential in expressing viewpoints. The study of the use of iconic gestures, indeed, might also be more fruitful for the finding of examples of expression of viewpoints.

McNeill’s study is insightful in that it recognizes the multi-modal nature of the representations of viewpoints. He has shown that the representations of viewpoints are not confined to language, and that the embodied, non-verbal devices like gestures can also powerfully express viewpoints. His identifications of character and observer viewpoint are also important, since they suggest the existence of different viewpoints that can also be inferred from gestural aspects. Even though the definitions of these two terms are illustrated from a purely gestural perspective, the present study follows McNeill’s concept of these two viewpoints and attempts to supply a definition that can better account for both the linguistic and gestural viewpoints.