A Corpus-Assisted Study of Nominalization in Translated and Non-translated Judgments

This exploratory study examines nominalizations in judgments by comparing three comparable corpora, the translated English judgment corpus (Hong Kong Translated Judgment Corpus), the non-translated English judgment corpus (Hong Kong Non-translated Judgment Corpus), and the non-translated Australian judgment corpus (Australian High Court Judgment Corpus). All the nominalizations (words ending in “-ion, -ment, -ity, -ness” and their plurals) were retrieved using an automated method, i.e., the Multidimensional Analysis Tagger (MAT) developed by Nini (Multi-dimensional analysis: Research methods and current issues. London: Bloomsbury Academic, pp. 67–94, 2019). The One-way ANOVA test shows a significant difference in the use of nominalizations between the three corpora. Specifically, nominalizations are most frequently found in the non-translated Australian judgment corpus, followed by the non-translated and translated Hong Kong judgment corpora. Further qualitative analysis shows that the types of nominalizations vary to a large extent between the three corpora and translation might be an important variable in affecting the use of nominalizations in the translated judgments. In light of the findings, we assume that the underuse of nominalizations in translated judgments can be attributed to a number of factors including translational simplification, translation purposes and the influence of Plain English Movement.