Meronymy across
Languages: Lexicalization,
Semantics, Morphosyntax September 27-28, 2013
Instituto de Investigaciones Antropológicas, Universidad Nacional Autónoma de México, Mexico City, Mexico The goal of this symposium is to contribute to the typology
of meronym systems in the languages of the world. Meronyms are terms that
describe entities as parts of larger entities. Terms for parts of the human
body are perhaps universally the prototypical meronyms, and semantic transfer
from body parts to object parts and spatial relations may well be universal
as well (Svorou 1994; Heine 1997). However,
evidence from Mesoamerican languages points to a number of features not
hitherto attested elsewhere. From the perspective of the available literature
on the typology of spatial descriptions, Mesoamerican meronymies
are unusual in two respects. First, they represent perhaps the most important
resource for the expression of place functions (as defined in Jackendoff 1983) in many Mesoamerican languages - in particular,
in languages without spatial case markers and with few or no adpositions.
Second, MA meronyms are systematically assigned on the basis of the geometry
of the object and the shapes of its parts, not on the basis of the parts’
functions. For example, in Western languages, the ‘blade’ and the ‘handle’ of
a knife are labeled by terms that apply to blades and handles of other
objects on the basis of their function, regardless of shape. In contrast, in
Yucatec Maya, the handle is the ‘leg’ of the knife. There is no word for the
blade as such; instead, the two planar surfaces of the blade are identified
as its ‘fronts’. These terms are applied to parts of similar shape in
arbitrary objects regardless of function. Two different proposals have been advanced to
account for the productivity of shape-based meronymy in MA. MacLaury (1989)
describes Ayoquesco Zapotec meronyms as body part terms that are
metaphorically extended to other entities on the basis of a global analogical
mapping process with the structure of an erect human body as its source
domain and the structure of the entity described by the ‘holonym’
in its actual orientation as the target domain. This mapping is
orientation-sensitive: the highest part of the object becomes the
metaphorical ‘head’ and the lowest part the ‘buttocks’ or ‘feet’, depending
on its shape. In contrast, Levinson (1994) describes meronym assignment in Tenejapan Tseltal as governed, not by a metaphorical
mapping process, but by an algorithm that takes as input the visually segmented
outline of the whole and labels parts on the basis of their shape and the
axis of the entity they occur on. Presently, the research project “Spatial language
and cognition beyond Mesoamerica” (MesoSpace; NSF Award No. BCS-1053123) is
examining the conceptual basis for meronym assignment, testing predictions
derived from the global-analogy account proposed by MacLaury for Zapotec and
the shape-analytical algorithm proposed by Levinson for Tseltal in their
field languages. Sponsored By:
©
2013 Spatial language and cognition beyond Mesoamerica project |