tv.um.es

This study investigated whether the law of growing standardization applies to
both Chinese writings and translations. To achieve this goal, I retrieved the normalized
frequencies of colloquial and dialectal words (as features against standardization) and
classical Chinese words (as features towards standardization) from a comparable
diachronic corpus with Chinese writings and translations. I examined the distribution
of these features across four text types in five periods: fictional and non-fictional
Chinese writings and translations from 1900–1911, 1919–1930, 1931–1949, 1950–
1966, and 1978–2012, respectively. To identify the semantic categories of colloquial or
dialectal words and classical Chinese words, I have annotated the corpus with CILIN
(同义词词林), a thesaurus that classifies Chinese words based on their semantic
relationship in a five-level hierarchical tree structure. Adopting a multivariate method
called multidimensional scaling, I tracked the developmental path of different variants
and the similarities and differences among them. A significant advantage of MDS is
that it provides representations of the similarities and differences among various items
on a lower dimension space. The closer one point is to another point, the greater the
similarities. The further two points are, the greater the dissimilarities. The Chinese
writings and translations in each period are represented by a point on a two-dimensional
space, which allows us to visualize the similarities and differences between Chinese
writings and translations in different periods. Another advantage of using MDS is that
it sketches the developmental path of Chinese writings and translations from the early
1900s to the 2000s. By tracking the time variants from the 1900s to the 2000s, we may
infer whether the Chinese writings and translations follow the same or opposite
developmental paths. Since the MDS plot failed to show the underlying frequencies of
colloquial, dialectal, and classical Chinese words under each semantic category, the
correlation coefficients (Pearson’s r) between each dimension and the averaged
frequencies of linguistic features under each semantic category were calculated to infer
which semantic categories contribute most robustly to the diachronic variation.
Semantic categories with r values larger than 0.7 or smaller than –0.7 were shown for
colloquial and dialectal words. With regard to classical words in written Chinese, the
threshold for r was lowered to values larger than 0.6 or smaller than –0.6, to capture a
greater amount of variance. Semantic investigations of the linguistic features revealed
how the writings or translations are drifting towards or away from standardization.
According to the hypothesis presented in the introduction, the colloquial and dialectal
words would have undergone a decrease in frequencies, whereas classical Chinese
words would have undergone an increase if Chinese writings and translations in the
2000s are more standardized than in the 1900s. However, empirical data have shown
that the standardization of Chinese writings and translations is highly dependent on
periodization and text types. The results show that (i) in general, Chinese writings and
translations in the 2000s are becoming more standardized than in the 1900s, but degrees
of standardization in different periods vary with sociohistorical factors and linguistic
planning, and (ii) “agile” text types like fictional writings and translations are less
standardized than their “uptight” non-fictional counterparts. The findings provide a
fine-grained explanation for the law of standardization.