Although this study hypothesizes that smoothing will manifest itself to different degrees in translations by experienced and inexperienced translators, it should be pointed out that smoothing is possibly one of the more problematic features of translated language. The results of the different operationalizations are discussed and interpreted in relation to smoothing. None of the existing studies have relevantly included expertise as a factor in examining the smoothing hypothesis.
The results of Neumann's (2012) study also indicate that leveling can be influenced by register. Grabowski (2013) uses a methodology based on Principal Components Analysis and Cluster Analysis (two methods that have been widely used in stylometry and authorship attribution) to test the leveling hypothesis based on word frequency and word distribution. The results of these two analyzes confirm the leveling hypothesis that translated texts are more similar than non-translations.
The study by Kruger and Van Rooy (2012) is the first to specifically focus on leveling in terms of register variability. In their study, leveling is operationalized as a smaller degree of register variability in a translated subcorpus than in an untranslated subcorpus, which occurs due to the effects of translation. Against this background, the first hypothesis states that leveling will occur to a greater extent in the work of inexperienced translators than in the work of experienced translators.
If this is the case, translation can be assumed to be the cause of the leveling of variation.
Data collection and data processing
As shown in table 2, the translation subcorpora has a strong representation of Germanic source languages - especially Afrikaans. This overrepresentation is due to the fact that the majority of the student translations have Afrikaans as their source language, which had to be reflected in the ET subcorpus for reasons of comparability. The NT subcorpus consists mainly of examples of British and South African English with some American English.
British and American English were included because they are global standard varieties, and South African English was included to match its representation in the two translation subcorpora. The total number of tokens and the percentage of the three varieties in the NT subcorpus in terms of the five different registers are shown in Table 3.
Linguistic features studied
Findings
Omission of the complementiser that
As can be seen in Figure 1, the complementizer is almost never omitted in the learning register in any of the three sub-corpora and only slightly more often in the academic register. Overall, the occurrence of this-omission is most varied in the registers of the NT subcorpus, while the pattern of this-omission appears to be similar for both translated subcorpora, although the ET subcorpus omits it more often, overall. , as a subcorpus of IT. The register distribution of this feature also follows roughly the same pattern in the three sub-corpora.
There is some visual indication of a leveling effect in translation, which is more noticeable in the IT subcorpus, but statistical support for this effect is lacking. In the reporting register of the IT subcorpus, conjunctive markers are slightly more common than in the same register in the NT and ET subcorpora. In the popular writing and creative writing registers, the IT subcorpus has a higher incidence of conjunctive markers, in contrast to the academic and educational registers, where it has the lowest incidence.
It may well be that they consciously or unconsciously overcompensate in the translation process due to their inexperience. Overall, the ET and NT subcorpora use conjunctive markers in a similar way across the five registers, although the ET subcorpora tend to use them slightly more in reportage, creative writing, and academic writing registers. This can be taken to indicate that the register sensitivity in the ET subcorpus is similar to that in the NT subcorpus with respect to this feature.
In summary, the varied pattern of IT subcorpora and the similarity between ET and NT subcorpora suggest that translation-related smoothing in the use of conjunctive markers across the five registers in the translated subcorpora does not occur. Interestingly, the reportage, creative writing, and popular writing registers in the NT subcorpus have similar TTRs—a pattern mirrored by the ET subcorpus. The fact that these three registers in translations by experienced translators reflect the smoothed distribution of non-translations does not constitute smoothing.
However, it is clear that the popular write register is noticeably higher compared to the other registers in the subcorpus, while the instructional write corpus has a lower TTR. It is interesting to see the TTR of creative writing compared to the other registers in the IT corpus. The highest scores occur in the creative writing and popular writing registers, the most informal registers in the corpus and the registers where the highest degree of readability would be expected.
This strong effect of register appears in the translated sub-corpora to the same extent as in the non-translation sub-corpora, and there is no indication that a leveling off has occurred. However, the NT subcorpus behaves somewhat differently in the lecture and instructional writing registers compared to the two translated subcorpora in the sense that the mean word length values for these two registers in the NT subcorpus are very similar, while the two translated subcorpora both have a larger average word length in the instruction as a reporting register.
Contraction ratio
The results of the factorial ANOVA show that there is no interaction effect for the independent variables register and corpus (F p=0.20) and the null hypothesis of no difference in the register-related distribution of single-occurrence lexical innovations cannot therefore be confidently rejected . There is also little evidence to support the second hypothesis that translation itself results in a leveling out of register differences. For all the features studied, register variation was evident in all three subcorpora, with none of the translational subcorpora showing a significant "flattened" distribution across the five registers.
While little evidence was found for features of leveling, the study sheds some light on the relationship between translation expertise and register sensitivity, and also how translations differ from non-translations across different registers for two, namely conjunctive markers and standardized TTR, of the seven traits studied. It is possible that they overtly connect ideas not only because they misjudge register expectations, but also to help them cognitively keep track of the text's development by using conjunctions to convey logical coherence to themselves. Apparently, experienced translators have a better understanding of the purpose of conjunction markers and how these work across different registers than inexperienced translators.
In addition, some of the more subtle effects indicate the register sensitivity of experienced translators in that the features studied followed a similar pattern or distribution in the translations compared to those of non-translators. When interpreting the results of the study, it should be kept in mind that the corpus used for this study is relatively small. Additionally, while this study attempted to address the current gap for product-oriented research in investigations of translation expertise, and how translation expertise affects register, it has methodological limitations in terms of the comparative design it used.
To address the problematic nature of the comparable-corpus methodology, Bernardini and Saldanha and O'Brien recommend that the comparable-corpus approach should be combined with a parallel-corpus design that creates two-way corpuses. As such, this observation leads to a call for corpus-based studies of translation expertise to incorporate a two-way corpus design that will not only produce more reliable results but also provide a better understanding of the impact of translation expertise on registers. Bertus van Rooy for their help with the processing and interpretation of the data and for the development of the arguments presented.
Steiner (eds.) Cross-language corpora for the study of translations: Insights from the English-German language pair. Ferreira (ed.) The development of translation competence: theories and methods from psycholinguistics and cognitive science. Results of the validation of the PACTE translation competence model: translation project and dynamic translation index.
Results of the validation of the PACTE translation competence model: Translation problems and translation competence. Steiner (ed.) Cross-language corpora for the study of translations: Insights from the English-German language pair.
