Anaphora, as an important linguistic phenomenon, represents a cohesive relationship concerning two parts, namely antecedent and anaphor. The choice of anaphoric forms is understudied in previous research. In this study, an annotation framework is built and a machine learning method is employed to analyse the influence of motivators on the choice of anaphoric forms. In addition, the framework of accessibility theory is modified, with causals as the study object. Results indicate that competition-, salience- and distance-related motivators, as well as text type, can significantly influence the variation of anaphoric forms. Among those motivators, predictability emerges as the most significant variable. Under the influence of these motivators, zero pronouns, noun phrases and pronouns exhibit significant differences in distribution. Pronouns have a broader distribution range and fewer restrictions compared to zero pronouns and noun phrases. Based on the results, we also modify the accessibility theory in terms of competition and salience.

Adopting a multifactorial analysis, this corpus-based comparative study examines the metaphorical conceptualizations of ENVY in American English and Chinese. All the metaphorical occurrences of ENVY in the two languages were collected from the corpora and then submitted to a detailed semantic and usage analysis. The qualitative and quantitative analyses show both striking similarities and differences in the metaphorical structuring of ENVY across the two languages. Both languages resort to largely the same common conceptual metaphors but use divergent elements from these metaphors’ source domains in their conceptualizations of ENVY. Moreover, a correspondence analysis of the most frequently used envy metaphors in relation to four important factors reveals some noticeable cross-language differences in their behavioural profiles. Based on the corpus findings, this study also explores the motivations underlying the similarities and differences in the metaphorical structuring of ENVY. The results indicate that the similarities are grounded mainly in common bodily experiences, whereas the differences are motivated by either differential cultural experiences or differential cognitive preferences.