Abstract
Natural languages often have elements with meaning but not sound, restricted in their distribution and interpretation. Grammar captures some of such restrictions by distinguishing different types of empty categories (NP-trace, variable, empty pronoun) -ECs, each of which is subject to interpretive and licensing rules. Chinese is prominent in the presence of empty elements. The restrictions on interpreting empty elements in Chinese were the topics of a vast literature in the last several decades. This work reviews the major proposals and shows that an important subject/object asymmetry in interpretive possibilities has not been properly recognized, which underlies many of the descriptive problems challenging the available accounts. It will be demonstrated that the problems can be solved if we allow the existence of a true empty position (TEP) -an empty element that cannot be an EC. It is truly empty and contains no features except the categorical ones, which can be obtained from the linguistic contexts. Its existence is forced because grammar prohibits an EC to occur in a certain position but the position is necessary to fulfill subcategorization requirements. It is a last resort strategy. This provides important clues to understanding the different behavior of empty nominals in Chinese and Japanese-Chinese seems to exhibit an asymmetry in the interpretive possibilities of empty subjects and objects much more strictly than Japanese. The difference is traced to structures: Chinese projects nominal expressions in argument positions as DPs and Japanese, as NPs.
