Can

1
Can translation universals survivein Mandarin?
Idioms, word clusters, and reformulation markers
in translational Chinese

• Richard Xiao

2
TU A focus of CBTS

• An important area of corpus-based TS over the
  past decade
• Baker (1993, 1996) Chesterman (2004) Kenny
  (1998, 1999, 2000, 2001) Laviosa (1998)
  Mauranen Kujamaki 2004) McEnery Xiao (2002,
  2007) Olohan (2004) Olohan Bakers (2000)
  Øverås (1998) Pym (2005) Xiao and Yue (2008),
  Xiao (2010)
• Most prominent and pioneering work based on
  English data
• Translation universals
• Simplification
• Normalization
• Explicitation
• Sanitization
• Convergence / levelling out

3
TU A target of debate

• Is translational language different from target
  native language?
• Translational language is at best an
  unrepresentative special variant of the target
  language because translations cannot possibly
  avoid the effect of translationese
• e.g. Baker 1993 Gellerstam 1996 Hartmann 1985
  Laviosa 1997 McEnery Wilson 2001 McEnery
  Xiao (2002, 2007) Teubert 1996 Xiao (2010)

4
TU A target of debate

• Are the features uncovered on the basis of
  translational English generalizable to other
  translated languages?
• Existing evidence has largely come from
  translational English and related European
  languages
• If such features are to be generalized as
  translation universals, the language pairs
  involved must not be restricted to English and
  closely related languages
• Cheongs (2006) study of English-Korean
  translation contradicts even the least
  controversial explicitation hypothesis
• Evidence from genetically distinct language
  pairs such as English and Chinese is undoubtedly
  more convincing if not indispensable

5
The ZCTC corpus

• The ZJU Corpus of Translational Chinese (ZCTC)
  was created with the explicit aim of studying the
  features of translated Chinese
• A translational counterpart of the Lancaster
  Corpus of Mandarin Chinese (LCMC), a
  one-million-word balanced corpus of native
  Chinese (McEnery Xiao 2004)
• www.ling.lancs.ac.uk/corplang/lcmc/
• Five hundred 2,000-word text samples taken
  proportionally from fifteen written text
  categories published in China in the 1990s
• www.ling.lancs.ac.uk/corplang/ZCTC/

6
LCMC / ZCTC corpus design
7
Translated Chinese Some observations

• Laviosas (1998) classic findings of the four
  core patterns of lexical use in translational
  English are generally also applicable in
  translated Chinese (Xiao 2010)
• Beyond the lexical level
• Mean sentence length is sensitive to genre
  variation and may not be reliable as an indicator
  of simplification
• A comparison of frequent conjunctions in native
  and translated Chinese shows that simpler forms
  tend to be used in translations
• In spite of some genre-based subtleties,
  conjunctions are more frequently used in
  translational Chinese, providing evidence in
  favour of the explicitation hypothesis
• Idiom, word cluster, reformulation marker

8
Previous findings of 3 features

• Bakers (2007) observations of idioms paint a
  mixed picture, pointing in two opposite
  directions
• Translations are expected to make heavier use of
  idioms to confirm to the target language norm
• Idioms, and opaque idioms in particular, are
  expected to be avoided in translations because of
  their informal flavour
• Baker (2004) and Nevalainen (2005) find that
  recurring word clusters are more common in
  translations in comparison with non-translated
  texts
• According to Baker (2004), RMs are substantially
  more frequent in translational than native
  English fiction Chen (2006) also notes that RMs
  are more common in translated than native Chinese
  in the genres of popular science books

9
The issues

• Previous findings are interesting but largely
  based on specific genres such as fiction and
  popular science writing while language can vary
  substantially across genres (cf. Biber 1995)
• Scientific writing is the least diversified of
  all genres across various varieties of English
  (Xiao 2009)
• Likely that what has been observed of idioms,
  word clusters and reformulation markers in the
  previous studies might be specific to particular
  genres rather than applicable to translational
  English or translational Chinese as a whole
• It is also debatable whether the features
  uncovered on the basis of translational English
  can be generalized to other translated languages
  because of the nature of existing evidence

10
Idioms in LCMC ZCTC

• With a few exceptions (E, L, N - but the
  differences in E and N are not significant),
  idioms in native Chinese (LCMC) are considerably
  more frequent than in translated Chinese (ZCTC)
• The overall frequency of idioms in LCMC (7979) is
  also much higher than in ZCTC (6265)

11
Idioms in translated Chinese

• Balanced comparable corpora of translational and
  native Chinese appear to support Bakers (2007)
  observation of the second tendency in translation
  to avoid idioms
• This cross-linguistic contrast is probably a
  result of the different sources and functions of
  idioms in English and Chinese

12
Idioms in translated Chinese

• Many Chinese idioms, especially the so-called ??
  chengyu (typically four-character mould idioms),
  are historically allusive in origin, with stories
  from ancient times behind them, which often
  render them archaic in flavour and highly opaque
  in meaning (i.e. with their actual meaning
  different from their surface meaning)
• Hence Chinese idioms other than those called ??
  suyu (common saying) tend to carry a formal
  tone in style
• In contrast, English idioms, especially those
  with an opaque meaning (e.g. kick the bucket),
  tend to have an informal flavour of slangs

13
Word clusters in LCMC ZCTC
Word clusters LCMC ZCTC LL p
2-word clusters 21002 23006 103.44 0.000
3-word clusters 4015 5523 248.36 0.000
4-word clusters 580 732 16.58 0.000
5-word clusters 160 197 4.06 0.044
6-word clusters 70 105 7.25 0.007

• Focusing on word clusters of 2-to-6 words
• Longer clusters are infrequent in million word
  corpora
• Word clusters of all types are much more frequent
  in ZCTC than in LCMC
• All differences are statistically significant

14
Word clusters in LCMC ZCTC

• The more frequent use of word clusters in
  translational Chinese is also evidenced by a
  keyword cluster analysis
• For 2-to-6-word clusters, 958 clusters are
  significantly more frequent in ZCTC in contrast
  to 59 such clusters which are significantly more
  frequent in LCMC
• For 3-to-5-word clusters, 123 clusters are
  significantly more frequent in ZCTC as opposed to
  just one such cluster which is significantly more
  frequent in LCMC

15
Word clusters in LCMC ZCTC

• In addition to their significantly higher
  frequencies in translational Chinese, word
  clusters also demonstrate two other interesting
  characteristics
• High-frequency word clusters (gt0.01 of total
  corpus) are more common in translated Chinese
  (413 vs. 291, LL21.96, plt0.001)
• Word clusters have a much wider coverage in
  translated Chinese in comparison with native
  Chinese

16
A marked contrast showing an accelerating
tendency as the coverage rate drops
2-word clusters
3-word clusters
Similar because of the low overall frequencies
17
Word clusters in LCMC ZCTC

• The higher frequency and wider coverage of word
  clusters in translational Chinese suggest that
  translators demonstrate a higher propensity for
  striving for fluency than writers of native
  Chinese texts
• There is also a sharp contrast in highest
  coverage rate between the two corpora
• 2-word clusters 69.8 in LCMC (? ? DE one) vs.
  79.8 in ZCTC (? ? not be)
• 3-word clusters 19.6 in LCMC (? ? ? be one
  kind) vs. 27.6 in ZCTC (? ? ? but not be)

18
Word clusters in LCMC ZCTC

• Apart from the macro-level quantitative analysis
  above of frequency and coverage, a qualitative
  analysis of high-frequency word clusters at micro
  level also yields equally interesting findings
• Some 2-word and 3-word clusters are frequently
  used in both native and translated Chinese
• But there are a much greater number of
  high-frequency word clusters which are unique in
  ZCTC than those which are unique in LCMC

19
Word clusters in LCMC ZCTC

• High-frequency 2- and 3-word clusters which are
  unique in ZCTC are mostly demonstrative
  structures (e.g. ?? ? this DE) and modifying
  structures (e.g. ? ?? ? most important DE)
• Indeed, many of those demonstrative structures
  are also modifying structures
• In contrast, high-frequency word clusters which
  are unique in LCMC seem to be mainly head
  structures (e.g. ? ?? DE development, the
  development of)

20
Reformulation markers in LCMC ZCTC
Style Reformulation mark LCMC ZCTC
Formal (archaic) ? 267 274
Formal (archaic) ??? 5 8
Informal (colloquial) ???? 27 28
Informal (colloquial) ??? 14 25
Informal (colloquial) ????/???? 8 18
Informal (colloquial) ???? 15 10
Informal (colloquial) ????? 1 20
Informal (colloquial) ????? 1 10
Informal (colloquial) ???/??/???? 2 7
Total 340 400
Only RMs in a narrow sense, i.e. paraphrastic RMs
are considered, which are generally more
frequent in ZCTC than LCMC (LL 4.52, p0.033)
21
Formal vs. informal RMs

• Both formal and informal RMs are more common in
  ZCTC
• Frequencies of the formal forms of RMs in the two
  corpora are very close and not significant
  (LL0.127, p0.772)
• Informal colloquial items are significantly more
  frequent in ZCTC than LCMC (LL13.31, plt0.001)
• In line with previous observations of formal and
  informal conjunctions in native and translational
  Chinese (Xiao Yue 2009)

22
Conclusions

• Idioms are significantly more common in native
  Chinese as a whole and also in nearly all genres,
  a finding different from Bakers (2007) mixed
  findings based on English data
• As the distribution patterns of idioms tend to be
  language-specific, the heavy use of idioms may
  not be a universal feature of translational
  language
• The statistically significant quantitative
  contrast in the use of idioms in native and
  translational Chinese also tells a different
  story from the translation universal hypothesis
  of normalization

23
Conclusions

• Word clusters are substantially more common in
  translational Chinese in terms of frequency,
  coverage, keyword clusters and as well nature
• Translators tend to use fixed and semi-fixed
  recurring patterns which are purely structurally
  defined on the basis of their collocational
  behaviour in an attempt to achieve improved
  fluency
• Translated texts also tend to be more similar to
  each other than to non-translated texts, which
  means that the universal hypothesis of
  convergence or levelling out is upheld in the
  light of evidence from Chinese

24
Conclusions

• Our finding based on balanced comparable corpora
  of native and translational Chinese supports
  previous observations in some specific genres
  that RMs can function as an explicitation
  strategy
• Translational Chinese generally makes more
  frequent use of informal colloquial RMs, but the
  distribution of formal markers seems to interact
  with the formality of genres, suggesting that
  translations are stylistically simpler than
  native Chinese texts
• While explicitation is supported in translational
  Chinese, the patterns of RM use in translated
  Chinese also provide evidence in support of
  simplification and convergence but against the
  normalization hypothesis

25

• Thank you!
• Richard.Xiao_at_edgehill.ac.uk