ABSTRACT
There are 20 basic categories in the system of parts of speech, among which 19 categories are conventional and only one is special. The parts of speech in modern Chinese are classified into 20 basic categories, different from the other classifications, which have 11 or 13 categories. The statistical research on parts of speech proves the fact that the relationship between word distribution and parts of speech is very intricate. Word frequency is significantly related with some grammatical functions of parts of speech. The system of parts of speech in modern Chinese is characterised by its "flexibleness" and "rigidness", according to the Amsterdam model in typology of part of speech. In Chinese, a permanent property of degree is coded as a predicate; a permanent property of a non-degree is coded as a modifier. The analyses of distribution of parts of speech based on the data corpus prove the intricate relationship between word distribution and parts of speech.
