Ali will be presenting his QP data this week and discussing his ICLC presentation: “Principle of Canonical Orientation, a Cross-Linguistic Study.”
This study presents evidence for language-specificity in the violability of the Principle of Canonical Orientation (POCO) (Levelt 1996). POCO claims that there is a restriction on the use of intrinsic frames of reference depending on the orientation of the entity that the intrinsic frames is derived from. Reference frames are conceptual coordinate systems that are projected onto ‘figures’ and ‘grounds’ (Talmy, 2000: 312) in order to orient the former and locate it with respect to the latter. In intrinsic frames of reference, the ground object is the anchor that the axes are derived from. It has been shown that different frames of reference strategies are more prevalent than others in different language communities (e.g. Levinson, 2003).
POCO states that for “the intrinsic system to refer to a relatum’s intrinsic dimension, that dimension must be in canonical position with respect to the perceptual frame of orientation of the referent” (Levelt 1996: 92). That is, for the axes of the coordinate system to be anchored to the ground object, the ground object must be in its canonical posture. This restriction stems from the disalignment of the vertical axes of the ground object and the figure’s perceptual frame of reference. Thus, POCO predicts that speakers would not produce descriptions such as (1) below to describe the picture in figure 1.
(1) The ball is above the chair
However, it has been found that such descriptions are indeed produced by speakers of different language communities such as English and Yucatec speakers (e.g. Carlson-Radvansky & Irwin 1993). Bohnemeyer & Tucker’s (2010) findings suggest that POCO is not an absolute constraint but rather a tendency. The data in the present study are from 7 languages (K’iche’, Yucatec, Zapotec, Spanish, Arabic, Mandarin, and Taiwanese). Participants produced spatial descriptions during a referential communication task in which a director describes photos so that a partner may select a match.
Initial results from the first five languages mentioned above show that speakers of all languages use the intrinsic frames significantly less when the ground object is in non-canonical orientations. It is, however, found that degree of adherence to POCO varies across different languages. For example, K’iche’ and Yucatec speakers violate POCO significantly more than Spanish or Hijazi Arabic speakers. This crosslinguistic variation could be explained in terms of a preference for egocentric virsus allocentric frames in small space-scale. Intrinsic reference in languages that prefer egocentric frames seems to only be available under specified conditions. This does not seem to be the case in languages where allocentric frames are more prevalent. These findings addresses possible cross-linguistic differences in utilizing the human body as model for spatial reference assignment.