Key points are not available for this paper at this time.
The article studies the issues of measuring derivational productivity of affixes. It considers the concept of productivity and its various aspects which can be quantified by corpus linguistics methods. In particular, it analyzes the measures of “expanding” and “potential” productivity by H. Baayen, and proposes a new formula for “universal” productivity on the basis of the analysis. These three productivity measures are applied to 80 Chinese suffix morphemes (suffixes and semi-suffixes) within a thematic collection of texts – Chinese media reports on military topics for 2019. The language material comprises 2066 texts (2.852.262 characters). The study resulted in three variants of the rating list for the most productive suffix morphemes of the Chinese military news discourse, which can be used for didactic and applied purposes. H. Baayen’s “expanding productivity” and the “universal productivity” proposed in the work showed a high correlation (0.85), while H. Baayen’s “potential productivity” correlates with them negatively. A meaningful analysis of the results also brings up a question of the appropriateness of using the “potential productivity” measure. Comparison of existing measures and the formula proposed in the work allow us to conclude that for an approximate assessment of the suffix morphemes productivity one can use the formula of “expanding productivity” by H. Baayen, which, despite relying only on the single word token (hapax legomenon) frequency, gives quite plausible results. To measure the suffix morphemes productivity more precisely, it is recommended to use the “universal productivity” formula proposed in this paper which takes into account a larger number of significant parameters (the number of hapax legomena with the estimated affix, the total type count with the given affix, and the average regular word token frequency of the estimated affix) and is more reasonable in this respect.
D. S. Korshunov (Fri,) studied this question.