|View Our Catalog||
from the Ordinary: Personal Experience Narratives in American Sign Language|
WHAT CONSTITUTES DATA FOR ANALYSIS?
Imagine someone making the following command to you: “Hand me that pen.” You look down and see three different pens in front of you, a red one, a blue one, and a black one. Which one do you hand to the speaker? Which pen does the phrase that pen refer to? It is not possible to determine which pen to give the person based solely on the words uttered. If, however, we saw the speaker point toward the blue pen while uttering the sentence, we would know which pen to hand over.
Traditionally in the field of linguistics a distinction is made between linguistic and paralinguistic information. In this view, words and how they are structured constitute linguistic data. Paralinguistic information includes variations in the intonation, stress, and articulation of words. For some analysts this term also includes any movement of the body that accompanies the uttered words. Thus all four words in “Hand me that pen” fall into the category of linguistic information, whereas the gesture accompanying the statement is paralinguistic. The analysis of vocally produced narratives, with few exceptions, restricts the data for analysis to linguistic elements such as words and morphemes and omits paralinguistic information.
Because of this focus on linguistic information and the exclusion of paralinguistic information, words are given prominence. Linguists commonly assumed that words alone are sufficient to determine the meaning and structures of a narrative. Other sources of meaning are minimized, such as how the voice is used. Does the narrator vary the volume of his or her voice? Is the pace varied? Is the pronunciation of words altered? Studies usually neglect to analyze the use of different facial expressions. Gestures that accompany any of the words are not seen as available to contribute meaning and enhance one’s understanding of the narrative. The paralinguistic aspects of how a narrative is conveyed (voice quality, facial expressions, gestures, other movements of the body) are excluded if they never become part of the transcript.
McNeill (1992) and Kendon (2004) describe how gestures are used with spoken languages. McNeill’s analysis includes different classifications of gestures. Imagistic gestures convey an image of what is being referred to. This image may be physical, such as the shape of the object or how the object moves. Other gestures are non-imagistic. These include deictic gestures such as pointing or rhythmic movements, which mark segments of the discourse. Gestures have different functions, and, as Kendon argues, they contribute to the meaning of the utterances in which they are part. Consider, for example, a person using the distance between his or her hands to specify size while saying, “I am looking for a piece this big.” The words this big introduce size without being specific about actual size, while the gesture demonstrates the information about size. The contributions of both parts allow the addressee to correctly interpret the utterance.
Gestures may be excluded from most spoken narrative analysis because they do not lend themselves to easy documentation. If the example above was part of a typical sociolinguistic analysis of narrative, it would be transcribed as “I am looking for a piece this big.” Documentation of the gestures might be added in parentheses, for example, “(hands held about four inches apart).” This description may not clearly describe the details of the gesture. That is, it does not specify the orientation of the hands, the extension of any fingers, etc. It also does not identify at which point in the utterance the gesture appeared. Was the gesture produced simultaneously with the utterance beginning with the word “I” or did the speaker raise his or her hands while speaking the words “this big”? The fact that analysts rely on written transcripts may limit the inclusion of gestural information in the analysis of spoken narratives.
Very detailed systems for recording the details of gestures have been developed by McNeill and his students. The transcription systems for documenting the use of gestures attempts to be as precise as possible in identifying the specific video frame in which the gesture first appears. The work of McNeill, Kendon, and others argues for the importance of gestures in language. It also provides a methodology for documenting its production.
The inconsistent appearance of paralinguistic information in sociolinguistic analyses may be due to the use of audiotape recordings to collect narrative data. Audiotape recorders preserve only what can be heard. In spite of this, the importance of including gestures in a sociolinguistic analysis has received some attention. Hinrichs and Polanyi (1986) argue that deictic gestures are necessary to the correct interpretation of discourse and therefore must be included in one’s explanation of how discourse is structured.
It is a given that the signs produced in an ASL narrative must be represented in a transcript. Evidence of this is the fact that in all the studies reviewed, authors recorded a gloss for every sign articulated in the transcript. After that point there is considerable variation in what other information is documented. Body posture; eye gaze; head placement; mouth, cheek, and eyebrow movement; and facial expressions are all behaviors that contribute to the meanings of ASL narratives. These behaviors are not consistently documented in transcripts across different analyses, and it is not clear how this information impacts the analysis.
Liddell and Metzger do include descriptions of gestures used in their transcript. They also describe how signs can be directed in space. The ability to direct signs in gradient ways requires that the transcript also reflect this information. Compare, for example, the signs in figure 1.6a and b. The signer articulates two instances of the sign there→L. In figure 1.6a, the sign there→L1 is directed ahead and to the left of his body. In figure 1.6b there→L2 is directed to an area near his left shoulder. The two instances of the sign there→L differ in their direction. A transcript that simply includes the gloss there will not provide the information needed for a correct interpretation of the discourse.
Sign languages are different from spoken languages in that they are perceived visually rather than aurally. The language modality requires that signing be recorded using videotape and this visual documentation must be viewed in order to conduct an analysis. A transcript in the form of English glosses traditionally does not include this visual component. The transcript may be supplemented with the information that the researcher feels is critical to the analysis or interpretation of the discourse. How this is accomplished varies somewhat, but a gloss such as there is typical. The gloss may be accompanied by nonmanual information such as eye gaze, furrowed brows, a shocked expression, etc.
Using English to describe what sign glosses do capture is useful but risks omission or misinterpretation of information that is important to the analysis. For instance, I have not seen details such as the following in published transcripts: there (hand started at neck and ended ahead of signer and at level of stomach). The description is intended to show that the sign began at a higher vertical location than it ended. This may be analogous to spoken language transcripts that do not include prosody. This detail may be treated as paralinguistic and therefore excluded from the linguistic analysis. So as not to exclude important information, it is necessary that photos and expanded glosses be included in the transcript of an ASL narrative. It is also critical to return to the video as an analysis is being conducted to minimize exclusion of paralinguistic information.
If we expand our concept of language from solely grammatical symbols to include how speakers express meaning through combinations of grammar, gestures, and other paralinguistic information, then we cannot exclude any aspect of what a speaker or signer produces. All aspects of the utterance—words (signs), gestures, tone of voice, pitch of voice, direction of a sign, facial expression—must be included in the transcript in order to not lose data important to understanding the meanings expressed in narrative.
3. The term kinesics has been used to refer to “articulation of the body, or movements resulting from muscular and skeletal shifts” (Key 1975, 10). The term co-speech gesture is also commonly used to label gestures produced while speaking.