Classifier systems in human languages are derived from human perception of entities in the world. Classifiers are considered linguistic correlates to perception. According to Allan, ``classifiers have meaning in the sense that a classifier denotes some salient perceived or imputed characteristic of the entity to which the associated noun refers (or may refer)'' [All77]. A classifier is a word or a particle, used in some cases with numeral-noun combinations and in other cases with certain types of verbs, that assigns entities to common categories on the basis of shared characteristics. According to Andersen [And78, page 344,], the primary distinction made in the classifier systems across languages of the world is between animate and inanimate. This distinction is based partly on perceptual properties, such as movement. Both animate and inanimate objects are often categorized on the basis of shape. In particular, the shapes round, long and less frequently flat are most often picked out, and are used either alone or with some secondary features such as rigidity, relative size, or orientation (i.e., vertical versus horizontal) to classify objects [AC73,All77].
In this paper we will discuss the concept of categorization or classification of entities by native speakers of Assamese based on perception of shape and discuss how the categorization impacts on the proper choice of classifiers.