CHILDES German Weissenborn Corpus

Jürgen Weissenborn
Institute for German Language
Humboldt University
juergen.weissenborn@rz.hu-berlin.de
website

Participants:	50
Type of Study:	naturalistic
Location:	Germany
Media type:	no longer available
DOI:	doi:10.21415/T5301C

Citation information

Weissenborn, J. (1985). Ich weiss ja nicht von hier aus, wie weit es von da hinten ist: Mak-roräume in der kognitiven und sprachlichen Entwicklung des Kindes. In H. Schweizer (Ed.), Sprache und Raum. Stuttgart: Metzlersche.

Weissenborn, J. (1986). Learning how to become an interlocutor: The verbal negotiation of common frames of reference and actions in dyads of 7- to 14-year-old children. In J. Cook-Gumperz (Ed.), Children’s worlds and children’s language. The Hague: Mou-ton.

In accordance with TalkBank rules, any use of data from this corpus must be accompanied by at least one of the above references.

Project Description

This corpus is a set of protocols taken from older children by Jürgen Weissenborn of the Max-Planck-Institut in the context of experimental elicitations of route descriptions. This corpus contains verbal protocols taken from a route-description task administered to German children and adults. The experiment carried out consisted of a route description task with pairs of German children of the same age — 7, 8, 9, 10, 11, and 14 years. In each group, 6 to 10 pairs of participants were tested.

The participants could not see each other. Each had an identical model of a small town in front of him and the direction giver had to specify for the other participant the route of a toy car through the town. The task material consisted of two identical three-dimensional wooden models of towns (0.60m by 0.70m). The houses, with red or blue roofs and two different sizes, were organized symmetrically (mirror-image) around a central axis. Four different paths (A, B, C, and D) of equal difficulty (same number of subpaths and turning points) were defined and each was then successively described by one child to another un-der three different conditions.

with supplementary landmarks (trees, animals, cars) destroying the symmetry of the display and with gestures (the children were allowed to use their hands freely during the description);
without landmarks and with gestures; and
without landmarks and without gestures (the children were sitting on their hands)

These conditions were combined with paths A to C as follows: 1A-2B-3C; 2B-1C-3A; and so forth. Path D was always described by the child to the experimenter under condition 2. The descriptions were videotaped.

The symmetrical design of the model was chosen because the referential determinacy of any path description that refers to it is only guaranteed if these descriptions are embedded in a verbal reference frame that has jointly been defined by the participants. For example, a description like “You pass under the bridge” would not suffice given that there are two bridges. The same holds for every other building. In order to resolve this indeterminacy the use of relational expressions like “left,” “right,” “in front of,” and “behind” is required. But, the reference of these terms is itself indeterminate between the deictic and the intrinsic perspective when applied to oriented objects. Thus, when applied to the toy car that the child drives along the path, “left” and “right” coincide with the describer’s perspective as long as the car moves away from him; when the car moves towards him this is no longer the case so that, at least for this instance, the describer has to specify explicitly which per-spective he has chosen if he wants to avoid misunderstandings.

This is only possible if these alternative perspectives are discriminated and if the ensu-ing necessity to coordinate the speaker’s and listener’s perspective is recognized. Notice that the two perspectives or reference frames are not equivalent in terms of cognitive com-plexity. The deictic perspective is based on the projection of the body schema of the speaker onto the experimental display whereas in the intrinsic perspective it is first mentally trans-posed onto the oriented object (i.e., the toy car) and then projected onto the display thus necessitating the constant coordination between the original deictic and the transposed in-trinsic use. Thus the structure of the experimental display asking for the use of these spatial terms has necessary conversational implications in that it requires the negotiation of the rules of use of these terms in order to establish a shared frame of reference and action.

What has been said so far about the consequences of the experimental design for the task solution applies in particular to condition 2. The task requirements are obviously quite different in condition 1 where the symmetrical design is destroyed by the introduction of additional landmarks. In this condition an unambiguous description of the path could be achieved by relying mainly on the information provided by these elements without neces-sarily using relational terms like “left” and “right.” That is, these landmarks furnish a con-crete and fixed frame of reference, external to the describer. Condition 3 was designed to study the influence of the absence of gestures on the child’s descriptive abilities.

In order to evaluate the describer’s ability to establish a coherent frame of reference a certain number of parameters have been defined that are considered to characterize each individual describer’s performance, that is completeness of the path description defined in terms of adequate characterization of the turning points and the connections between them, prevailing perspective, perspective awareness, and so forth.