4.2.2 Observations
For each translated or interpreted version of sentences in our corpus, descriptive data was gathered for three independent variables – mode, target language and sentence complexity – as well as three dependent variables identified as indicators of difficulty in translation or interpretation – reordering, nesting changes and changes in semantic relations. Summary statistics for the rates of each dependent variable corresponding to each pair of independent variables are presented below. For nesting changes, observations on changes in double and triple nestings are presented, but not included in the frequency comparisons because of the low values involved. For changes in triple nestings, few non-zero values were observed, so only mean rates are shown.
This section simply reports descriptive observations on data in the corpus. The next section will present the results of statistical tests predicting the mean response of each dependent variable to groups of independent variables in interaction, as well as the statistical significance of those responses.
Let’s start our observations with summary statistics for the three indicators of difficulty observed for various combinations of mode and target language. The figures are shown in table 7.
Table 7
Statistics for indicators of difficulty observed per sentence
for combinations of mode and target language
(R = Reordering, N1/N2/N3 = Changes in single/double/triple nestings, S = Changes in semantic relations)
| Mode | Target language | Mini- mum | First quartile | Median | Mean | Third quartile | Maximum | Standard deviation | Number of obser- vations |
||||||||||||
| R | N1 / N2 | S | R | N1 / N2 | S | R | N1 / N2 / N3 | S | R | N1 / N2 | S | R | N1 / N2 | S | R | N1 / N2 | S | ||||
| legal trans- lation | Russian | 0 | 0 | 0 / 0 | 0 | 0 | 0 / 0 | 0 | 0.31 | 0.26 / 0.01 / 0.00 | 0.30 | 0.00 | 0 / 0 | 0 | 7 | 5 / 1 | 8 | 0.87 | 0.73 / 0.08 | 1.04 | 498 |
| Hungarian | 0 | 0 | 0 / 0 | 0 | 1 | 1 / 0 | 0 | 2.31 | 1.29 / 0.10 / 0.00 | 0.55 | 3.00 | 2 / 0 | 1 | 38 | 12 / 5 | 11 | 3.57 | 1.68 / 0.50 | 1.31 | ||
| Turkish | 0 | 1 | 0 / 0 | 0 | 5 | 2 / 0 | 1 | |8.33| | |3.07| / 0.27 / 0.01 | 1.58 | 9.75 | 4 / 0 | 2 | 140 | 31 / 11 | 25 | 13.41 | 3.82 / 0.96 | 2.63 | ||
| Mandarin | 0 | 0 | 0 / 0 | 0 | 1 | 1 / 0 | 0 | 3.00 | 1.99 / 0.36 / 0.03 | 0.75 | 3.00 | 3 / 0 | 1 | 39 | 17 / 11 | 10 | 4.90 | 2.38 / 0.93 | 1.26 | ||
| Japanese | 0 | 1 | 1 / 0 | 0 | 2 | 2 / 0 | 0 | |9.13| | |3.22| / 0.35 / 0.02 | 0.79 | 10.00 | 5 / 0 | 1 | 170 | 27 / 10 | 17 | 15.25 | 3.75 / 1.16 | 1.56 | ||
| subtitle trans- lation | Russian | 0 | 0 | 0 / 0 | 0 | 0 | 0 / 0 | 0 | 0.12 | 0.12 / 0.00 / 0.00 | 0.44 | 0.00 | 0 / 0 | 0 | 4 | 3 / 0 | 5 | 0.47 | 0.41 / 0.00 | 0.91 | 413 |
| Hungarian | 0 | 0 | 0 / 0 | 0 | 0 | 0 / 0 | 0 | 0.17 | 0.21 / 0.00 / 0.00 | 0.32 | 0.00 | 0 / 0 | 0 | 6 | 4 / 1 | 4 | 0.55 | 0.58 / 0.07 | 0.75 | ||
| Turkish | 0 | 0 | 0 / 0 | 0 | 1 | 0 / 0 | 0 | 1.97 | 0.97 / 0.27 / 0.00 | 0.46 | 3.00 | 1 / 0 | 1 | 18 | 10 / 5 | 5 | 2.58 | 1.46 / 0.39 | 0.92 | ||
| Mandarin | 0 | 0 | 0 / 0 | 0 | 0 | 0 / 0 | 0 | 0.38 | 0.44 / 0.01 / 0.00 | 0.62 | 0.00 | 1 / 0 | 1 | 10 | 6 / 1 | 6 | 0.94 | 0.85 / 0.09 | 1.08 | ||
| Japanese | 0 | 0 | 0 / 0 | 0 | 1 | 0 / 0 | 0 | 1.59 | 0.71 / 0.01 / 0.00 | 1.15 | 2.00 | 1 / 0 | 2 | 27 | 8 / 2 | 10 | 2.38 | 1.24 / 0.14 | 1.59 | ||
| simul- taneous interpre- tation | Russian | 0 | 0 | 0 / 0 | 0 | 0 | 0 / 0 | 0 | 0.02 | 0.11 / 0.00 / 0.00 | 0.72 | 0.00 | 0 / 0 | 1 | 1 | 3 / 0 | 9 | 0.13 | 0.40 / 0.00 | 1.28 | 225 |
| Hungarian | 0 | 0 | 0 / 0 | 0 | 0 | 0 / 0 | 1 | 0.20 | 0.21 / 0.00 / 0.00 | 1.50 | 0.00 | 0 / 0 | 2 | 4 | 2 / 0 | 9 | 0.54 | 0.51 / 0.00 | 1.92 | ||
| Turkish | 0 | 0 | 0 / 0 | 0 | 1 | 0 / 0 | 1 | 1.19 | 0.64 / 0.02 / 0.00 | |1.72| | 1.00 | 1 / 0 | 3 | 11 | 5 / 2 | 8 | 1.78 | 1.00 / 0.18 | 1.94 | ||
| Mandarin | 0 | 0 | 0 / 0 | 0 | 0 | 0 / 0 | 0 | 0.27 | 0.47 / 0.01 / 0.00 | 1.07 | 0.00 | 1 / 0 | 2 | 3 | 5 / 1 | 7 | 0.63 | 0.89 / 0.09 | 1.58 | ||
| Japanese | 0 | 0 | 0 / 0 | 0 | 0 | 0 / 0 | 1 | 0.98 | 0.68 / 0.03 / 0.00 | |1.76| | 1.00 | 1 / 0 | 3 | 8 | 7 / 1 | 9 | 1.41 | 1.05 / 0.17 | 1.90 | ||
In table 7, the two highest values in the columns showing the mean rates for each indicator of difficulty (R, N1 and S) corresponding to each mode are underlined. Those underlined values show that, among the five target languages for translation or interpretation in this study, the highest mean rates for all three indicators were generally observed in Turkish and Japanese.
Of the underlined values in the columns showing the mean rates for each indicator of difficulty, the two highest values are also in boxes. Those boxed values show that the highest mean rates for reordering were observed in legal translation, where the mean rate for reordering was more than 8 place shifts per sentence in both Turkish and Japanese. (Counts for reordering in individual sentences reached 140 in Turkish and 170 in Japanese.) The highest mean rates for nesting changes were also observed in legal translation, where the mean rate for changes in single nestings was more than 3 changes per sentence in both Turkish and Japanese. (Counts for changes in single nestings in individual sentences reached 31 in Turkish and 27 in Japanese.) The highest mean rates for changes in semantic relations were observed in simultaneous interpretation, where the mean rate was over 1.7 changes per sentence in both Turkish and Japanese. (Counts for changes in semantic relations in individual sentences reached 17 in both languages.)
The mean rates for each indicator of difficulty observed for various combinations of mode and target language are visualized in chart 1.
Chart 1. Mean rates for indicators of difficulty observed per sentence
for combinations of mode and target language
Reordering:
highest in Turkish and Japanese,
especially in legal translation
[If charts don’t appear, refresh page.]
Changes in single nestings:
highest in Turkish and Japanese,
especially in legal translation
Changes in semantic relations:
highest in Turkish and Japanese,
especially in simultaneous interpretation
[Hover mouse over bars to see values.]
Similar trends were found in all three legal translation texts (the UDHR, the Paris Agreement and the FCPA). Table 8 shows summary statistics for the three indicators of difficulty observed for various combinations of legal text and target language.
Table 8
Statistics for indicators of difficulty observed per sentence
for combinations of legal text and target language
(R = Reordering, N1/N2/N3 = Changes in single/double/triple nestings, S = Changes in semantic relations)
| Legal text | Target language | Mini- mum | First quartile | Median | Mean | Third quartile | Maximum | Standard deviation | Number of obser- vations |
||||||||||||
| R | N1 / N2 | S | R | N1 / N2 | S | R | N1 / N2 / N3 | S | R | N1 / N2 | S | R | N1 / N2 | S | R | N1 / N2 | S | ||||
| UDHR | Russian | 0 | 0 | 0 / 0 | 0 | 0.0 | 0 / 0 | 0 | 0.40 | 0.32 / 0.03 / 0.00 | 0.10 | 0.00 | 0.00 / 0 | 0 | 6 | 5 / 1 | 3 | 1.16 | 0.97 / 0.17 | 0.46 | 68 |
| Hungarian | 0 | 0 | 0 / 0 | 0 | 1.0 | 1 / 0 | 0 | 2.04 | 1.56 / 0.15 / 0.00 | 0.32 | 3.00 | 2.00 / 0 | 0 | 10 | 9 / 5 | 4 | 2.38 | 1.97 / 0.72 | 0.82 | ||
| Turkish | 0 | 1 | 0 / 0 | 0 | 3.0 | 3 / 0 | 0 | 5.82 | 2.81 / 0.46 / 0.01 | 0.24 | 8.00 | 4.00 / 0 | 0 | 61 | 16 / 11 | 4 | 8.61 | 3.82 / 0.96 | 0.77 | ||
| Mandarin | 0 | 0 | 0 / 0 | 0 | 1.0 | 1 / 0 | 0 | 2.18 | 1.46 / 0.24 / 0.03 | 0.41 | 3.00 | 2.00 / 0 | 1 | 13 | 11 / 6 | 3 | 3.42 | 3.02 / 1.65 | 0.76 | ||
| Japanese | 0 | 1 | 0 / 0 | 0 | 3.0 | 3 / 0 | 0 | 5.72 | 2.82 / 0.21 / 0.00 | 0.21 | 8.25 | 4.00 / 0 | 0 | 52 | 13 / 7 | 7 | 7.78 | 2.93 / 1.04 | 0.91 | ||
| Paris Agree- ment | Russian | 0 | 0 | 0 / 0 | 0 | 0.0 | 0 / 0 | 0 | 0.26 | 0.26 / 0.00 / 0.00 | 0.19 | 0.00 | 0.00 / 0 | 0 | 5 | 5 / 1 | 4 | 0.79 | 0.68 / 0.07 | 0.64 | 225 |
| Hungarian | 0 | 0 | 0 / 0 | 0 | 1.0 | 1 / 0 | 0 | 2.80 | 1.49 / 0.04 / 0.00 | 0.57 | 4.00 | 2.00 / 0 | 1 | 38 | 12 / 2 | 11 | 4.24 | 1.80 / 0.22 | 1.28 | ||
| Turkish | 0 | 1 | 0 / 0 | 0 | 6.0 | 2 / 0 | 0 | |10.16| | 2.99 / 0.22 / 0.00 | 0.72 | 13.00 | 4.00 / 0 | 1 | 140 | 22 / 8 | 7 | 15.99 | 1.46 / 0.39 | 1.17 | ||
| Mandarin | 0 | 0 | 0 / 0 | 0 | 1.0 | 2 / 0 | 0 | 3.28 | 2.32 / 0.52 / 0.05 | 0.58 | 4.00 | 3.00 / 1 | 1 | 27 | 12 / 5 | 7 | 5.07 | 3.63 / 0.76 | 1.10 | ||
| Japanese | 0 | 1 | 1 / 0 | 0 | 6.0 | 2 / 0 | 0 | |11.00| | |3.57| / 0.63 / 0.04 | 0.66 | 14.00 | 5.00 / 1 | 1 | 170 | 25 / 10 | 7 | 18.53 | 3.87 / 1.54 | 1.11 | ||
| FCPA | Russian | 0 | 0 | 0 / 0 | 0 | 0.0 | 0 / 0 | 0 | 0.33 | 0.24 / 0.00 / 0.00 | 0.49 | 0.00 | 0.00 / 0 | 0 | 7 | 5 / 0 | 8 | 0.84 | 0.71 / 0.00 | 1.42 | 205 |
| Hungarian | 0 | 0 | 0 / 0 | 0 | 1.0 | 1 / 0 | 0 | 1.86 | 0.98 / 0.16 / 0.00 | 0.60 | 2.00 | 1.00 / 0 | 1 | 19 | 11 / 5 | 10 | 2.98 | 1.38 / 0.61 | 1.46 | ||
| Turkish | 0 | 1 | 0 / 0 | 1 | 5.0 | 2 / 0 | 2 | 7.16 | |3.25| / 0.27 / 0.01 | |2.98| | 9.00 | 5.00 / 0 | 4 | 116 | 31 / 7 | 25 | 11.23 | 1.00 / 0.18 | 3.43 | ||
| Mandarin | 0 | 0 | 0 / 0 | 0 | 1.0 | 1 / 0 | 0 | 2.97 | 1.79 / 0.23 / 0.00 | 1.06 | 3.00 | 3.00 / 0 | 2 | 39 | 17 / 11 | 10 | 5.11 | 4.26 / 0.84 | 1.48 | ||
| Japanese | 0 | 1 | 0 / 0 | 0 | 6.0 | 1 / 0 | 0 | 8.21 | 2.98 / 0.09 / 0.00 | |1.11| | 9.00 | 4.00 / 0 | 2 | 117 | 27 / 3 | 17 | 12.71 | 3.85 / 0.35 | 2.02 | ||
In table 8, the two highest values in the columns showing the mean rates for each indicator of difficulty (R, N1 and S) corresponding to each legal text are underlined. Those underlined values show that, among the three legal translation texts as in the entire corpus, the highest mean rates for all three indicators were generally observed in Turkish and Japanese.
Of the underlined values in the columns showing the mean rates for each indicator of difficulty, the two highest values are also in boxes. Those boxed values show that, among the three legal texts, the highest mean rates for reordering were observed in the Paris Agreement, where the mean rate for reordering was more than 10 place shifts per sentence in Turkish and Japanese. The mean rates for nesting changes were similar in all three legal texts. The highest mean rates for changes in semantic relations were observed in the FCPA, where the mean rate was nearly 3 changes per sentence in Turkish.
The mean rates for each indicator of difficulty observed for various combinations of legal text and target language are visualized in chart 2.
Chart 2. Mean rates for indicators of difficulty observed per sentence
for combinations of legal text and target language
Reordering:
highest in Turkish and Japanese,
especially in the Paris Agreement
Changes in single nestings:
similar in all three legal texts
Changes in semantic relations:
highest in Turkish,
especially in the FCPA
[Hover mouse over bars to see values.]
Similar trends were also found in all five subtitle translation talks (the talks by Sir Ken Robinson, Amy Cuddy, Simon Sinek, Brené Brown and Tim Urban). Table 9 shows summary statistics for the three indicators of difficulty observed for various combinations of subtitled talk and target language. The table doesn’t show any data on changes in triple nestings, since there weren’t any triple nestings in the talks.
Table 9
Statistics on indicators of difficulty observed per sentence
for combinations of subtitled talk and target language
(R = Reordering, N1/N2 = Changes in single/double nestings, S = Changes in semantic relations)
| Subtitled talk by | Target language | Mini- mum | First quartile | Median | Mean | Third quartile | Maximum | Standard deviation | Number of obser- vations |
||||||||||||
| R | N1 / N2 | S | R | N1 / N2 | S | R | N1 / N2 | S | R | N1 / N2 | S | R | N1 / N2 | S | R | N1 / N2 | S | ||||
| Sir Ken Robinson | Russian | 0 | 0 | 0 / 0 | 0 | 0.0 | 0 / 0 | 0 | 0.19 | 0.14 / 0.03 | 0.48 | 0.00 | 0.00 / 0 | 1 | 3 | 2 / 0 | 4 | 0.55 | 0.43 / 0.00 | 0.85 | 69 |
| Hungarian | 0 | 0 | 0 / 0 | 0 | 0.0 | 0 / 0 | 0 | 0.23 | 0.19 / 0.00 | 0.20 | 0.00 | 0.00 / 0 | 0 | 2 | 2 / 0 | 3 | 0.49 | 0.43 / 0.00 | 0.61 | ||
| Turkish | 0 | 0 | 0 / 0 | 0 | 1.0 | 0 / 0 | 0 | 1.32 | 0.54 / 0.03 | 0.48 | 2.00 | 1.00 / 0 | 0 | 7 | 4 / 1 | 5 | 1.49 | 0.81 / 0.17 | 1.08 | ||
| Mandarin | 0 | 0 | 0 / 0 | 0 | 0.0 | 0 / 0 | 0 | 0.64 | 0.36 / 0.00 | 0.55 | 1.00 | 1.00 / 0 | 1 | 7 | 5 / 0 | 5 | 1.21 | 0.75 / 0.00 | 1.01 | ||
| Japanese | 0 | 0 | 0 / 0 | 1 | 1.0 | 0 / 0 | 1 | 1.19 | 0.64 / 0.01 | |1.23| | 2.00 | 1.00 / 0 | 2 | 7 | 6 / 1 | 7 | 1.40 | 1.18 / 00.12 | 1.39 | ||
| Amy Cuddy | Russian | 0 | 0 | 0 / 0 | 0 | 0.0 | 0 / 0 | 0 | 0.21 | 0.19 / 0.00 | 0.54 | 0.00 | 0.00 / 0 | 1 | 4 | 3 / 0 | 5 | 0.65 | 0.54 / 0.00 | 1.00 | 102 |
| Hungarian | 0 | 0 | 0 / 0 | 0 | 0.0 | 0 / 0 | 0 | 0.16 | 0.25 / 0.00 | 0.18 | 0.00 | 0.00 / 0 | 0 | 3 | 4 / 0 | 4 | 0.52 | 0.67 / 0.00 | 0.53 | ||
| Turkish | 0 | 0 | 0 / 0 | 0 | 1.5 | 1 / 0 | 0 | 2.04 | |1.25| / 0.13 | 0.34 | 3.00 | 2.00 / 0 | 0 | 11 | 10 / 5 | 4 | 2.16 | 1.67 / 0.59 | 0.76 | ||
| Mandarin | 0 | 0 | 0 / 0 | 0 | 0.0 | 0 / 0 | 0 | 0.29 | 0.39 / 0.00 | 0.57 | 0.00 | 0.75 / 0 | 1 | 3 | 4 / 0 | 4 | 0.67 | 0.77 / 0.00 | 0.97 | ||
| Japanese | 0 | 0 | 0 / 0 | 1 | 1.0 | 0 / 0 | 1 | 1.67 | 0.72 / 0.00 | |1.23| | 3.00 | 1.00 / 0 | 2 | 8 | 5 / 0 | 5 | 1.81 | 1.21 / 0.00 | 1.50 | ||
| Simon Sinek | Russian | 0 | 0 | 0 / 0 | 0 | 0.0 | 0 / 0 | 0 | 0.04 | 0.07 / 0.00 | 0.36 | 0.00 | 0.00 / 0 | 0 | 3 | 2 / 0 | 5 | 0.33 | 0.33 / 0.00 | 0.84 | 91 |
| Hungarian | 0 | 0 | 0 / 0 | 0 | 0.0 | 0 / 0 | 0 | 0.16 | 0.18 / 0.10 | 0.46 | 0.00 | 0.00 / 0 | 1 | 2 | 3 / 1 | 3 | 0.40 | 0.55 / 0.10 | 0.86 | ||
| Turkish | 0 | 0 | 0 / 0 | 0 | 1.0 | 1 / 0 | 0 | 1.52 | 0.96 / 0.04 | 0.67 | 2.50 | 1.00 / 0 | 1 | 9 | 5 / 2 | 4 | 1.88 | 1.23 / 0.25 | 0.99 | ||
| Mandarin | 0 | 0 | 0 / 0 | 0 | 0.0 | 0 / 0 | 0 | 0.36 | 0.55 / 0.01 | 0.46 | 0.00 | 1.00 / 0 | 1 | 4 | 6 / 1 | 4 | 0.77 | 0.99 / 0.10 | 0.87 | ||
| Japanese | 0 | 0 | 0 / 0 | 0 | 1.0 | 0 / 0 | 0 | 1.38 | 0.84 / 0.01 | 0.95 | 2.00 | 2.00 / 0 | 2 | 7 | 6 / 1 | 9 | 1.73 | 1.28 / 0.10 | 1.46 | ||
| Brené Brown | Russian | 0 | 0 | 0 / 0 | 0 | 0.0 | 1 / 0 | 0 | 0.06 | 0.12 / 0.09 | 0.40 | 0.00 | 0.00 / 0 | 0 | 2 | 2 / 0 | 5 | 0.29 | 0.37 / 0.00 | 0.92 | 80 |
| Hungarian | 0 | 0 | 0 / 0 | 0 | 0.0 | 0 / 0 | 0 | 0.24 | 0.29 / 0.10 | 0.59 | 0.00 | 0.00 / 0 | 1 | 6 | 3 / 1 | 4 | 0.83 | 0.72 / 0.11 | 1.01 | ||
| Turkish | 0 | 0 | 0 / 0 | 0 | 2.0 | 0 / 0 | 0 | |3.01| | |1.18| / 0.06 | 0.51 | 4.00 | 2.00 / 0 | 1 | 18 | 8 / 1 | 4 | 3.84 | 1.79 / 0.24 | 0.99 | ||
| Mandarin | 0 | 0 | 0 / 0 | 0 | 0.0 | 0 / 0 | 0 | 0.48 | 0.51 / 0.03 | 1.05 | 0.00 | 1.00 / 0 | 2 | 10 | 4 / 1 | 5 | 1.36 | 1.01 / 0.16 | 1.40 | ||
| Japanese | 0 | 0 | 0 / 0 | 1 | 1.0 | 0 / 0 | 1 | |2.44| | 0.92 / 0.03 | |1.68| | 3.25 | 2.00 / 0 | 3 | 27 | 8 / 1 | 10 | 4.01 | 1.59 / 0.16 | 2.11 | ||
| Tim Urban | Russian | 0 | 0 | 0 / 0 | 0 | 0.0 | 1 / 0 | 0 | 0.08 | 0.08 / 0.09 | 0.38 | 0.00 | 0.00 / 0 | 0 | 2 | 1 / 0 | 4 | 0.37 | 0.28 / 0.00 | 0.90 | 71 |
| Hungarian | 0 | 0 | 0 / 0 | 0 | 0.0 | 0 / 0 | 0 | 0.08 | 0.14 / 0.00 | 0.18 | 0.00 | 0.00 / 0 | 1 | 2 | 2 / 0 | 3 | 0.37 | 0.39 / 0.00 | 0.54 | ||
| Turkish | 0 | 0 | 0 / 0 | 0 | 1.0 | 0 / 0 | 0 | 1.93 | 0.76 / 0.08 | 0.27 | 3.00 | 1.00 / 0 | 0 | 13 | 7 / 3 | 5 | 2.67 | 1.37 / 0.44 | 0.74 | ||
| Mandarin | 0 | 0 | 0 / 0 | 0 | 0.0 | 0 / 0 | 0 | 0.15 | 0.34 / 0.00 | 0.45 | 0.00 | 0.50 / 0 | 0 | 2 | 2 / 0 | 6 | 0.40 | 0.63 / 0.00 | 1.03 | ||
| Japanese | 0 | 0 | 0 / 0 | 0 | 1.0 | 0 / 0 | 0 | 1.17 | 0.37 / 0.03 | 0.66 | 1.00 | 1.00 / 0 | 1 | 10 | 4 / 0 | 5 | 1.90 | 0.68 / 0.24 | 1.12 | ||
In table 9, the two highest values in the columns showing the mean rates for each indicator of difficulty (R, N1 and S) corresponding to each subtitled talk are underlined. Those underlined values show that, among the five subtitle translation talks as in the entire corpus, the highest mean rates for reordering and nesting changes were observed in Turkish and Japanese. The highest mean rates for changes in semantic relations were generally observed in Mandarin and Japanese.
Of the underlined values in the columns showing the mean rates for each indicator of difficulty, the two highest values are also in boxes. Those boxed values show that, among the five subtitled talks, the highest mean rates for reordering were observed in the talk by Brené Brown, where the mean rate was more than 3 place shifts per sentence in Turkish, and nearly 2½ place shifts per sentence in Japanese. The highest mean rates for nesting changes were observed in the talks by Amy Cuddy and Brené Brown, where the mean rate for changes in single nestings was more than 1 change per sentence in Turkish. The highest mean rates for changes in semantic relations were observed in the talks by Sir Ken Robinson, Amy Cuddy and Brené Brown, where the mean rate was well over 1 change per sentence in Japanese.
The mean rates for each indicator of difficulty observed for various combinations of subtitled talk and target language are visualized in chart 3.
Chart 3. Mean rates for indicators of difficulty observed per sentence
for combinations of subtitled talk and target language
Reordering:
highest in Turkish and Japanese,
especially in the talk by Brené Brown
Changes in single nestings:
highest in Turkish and Japanese,
especially in the talks by
Amy Cuddy and Brené Brown
Changes in semantic relations:
highest in Mandarin and Japnese,
especially in the talks by Sir Ken
Robinson, Amy Cuddy and Brené Brown
[Hover mouse over bars to see values.]
Now let’s look at sentence complexity, as measured by the number of functionally subordinate or reported propositions per sentence. Table 10 shows the calculations for the mean number of subordinate propositions per original sentence observed in each mode, each legal text and each subtitled talk in the corpus.
Table 10
Mean number of subordinate or reported propositions per original sentence observed
in each mode, each legal text and each subtitled talk
| Mode | Legal text | Subtitled talk | Minimum | First quartile | Median | Mean | Third quartile | Maximum | Standard deviation | No of observations |
| legal translation | 0 | 2 | 4 | |4.72| | 6 | 30 | 3.88 | 498 | ||
| subtitle translation | 0 | 1 | 2 | 2.41 | 3 | 12 | 1.99 | 413 | ||
| interpretation | 0 | 1 | 2 | 2.08 | 3 | 10 | 1.91 | 225 | ||
| UDHR | 0 | 1 | 3 | 3.18 | 4.0 | 14 | 2.73 | 68 | ||
| Paris Agreement | 0 | 2 | 4 | 4.46 | 6.0 | 22 | 3.54 | 225 | ||
| FCPA | 0 | 3 | 5 | |5.52| | 7.5 | 30 | 4.35 | 205 | ||
| Sir Ken Robinson | 0 | 1 | 2 | 2.22 | 3 | 12 | 1.86 | 69 | ||
| Amy Cuddy | 0 | 1 | 2 | 2.32 | 3 | 6 | 1.57 | 102 | ||
| Simon Sinek | 0 | 1 | 2 | 2.40 | 3 | 12 | 1.87 | 91 | ||
| Brené Brown | 0 | 1 | 3 | |3.33| | 5 | 11 | 2.52 | 80 | ||
| Tim Urban | 0 | 1 | 1 | 1.70 | 2 | 8 | 1.71 | 71 |
In table 10, the highest value in the column showing the mean degree of sentence complexity within each mode, each legal text and each subtitled talk is in a box. Those boxed values show that the mode with the most complex sentences on average was legal translation, where there were nearly 5 functionally subordinate or reported propositions on average per original sentence. The most complex legal text was the FCPA, where there were more than 5½ subordinate or reported propositions on average per original sentence. (Counts for sentence complexity in individual sentences reached 30 subordinate or reported propositions.) The most complex subtitled talk was the one by Brené Brown, where there were more than 3 subordinate or reported propositions on average per original sentence. (Counts for sentence complexity in individual sentences reached 11 subordinate or reported propositions.)
The mean number of subordinate or reported propositions per original sentence observed in each mode, each legal text and each subtitled talk are visualized in chart 4.
Chart 4. Mean number of subordinate or reported propositions observed per original sentence in each:
Sentence complexity among modes:
highest in legal translation
Sentence complexity
similar among legal texts,
highest in FCPA
Sentence complexity
similar among subtitled talks,
highest in talk by Brené Brown
[Hover mouse over bars to see values.]
Finally, let’s look at data involving structural variables. The first structural variable considered was difference in the branching direction of relative clauses. Table 11 shows summary statistics for the three indicators of difficulty observed for various combinations of mode and difference in the branching direction of relative clauses.
Table 11
Statistics on indicators of difficulty observed per sentence
for combinations of mode and difference in branching direction of relative clauses
(R = Reordering, N1/N2/N3 = Changes in single/double/triple nestings, S = Changes in semantic relations)
| Mode | Branching direction of relative clauses | Mini- mum | First quartile | Median | Mean | Third quartile | Maximum | Standard deviation | Number of obser- vations |
||||||||||||
| R | N1 / N2 | S | R | N1 / N2 | S | R | N1 / N2 / N3 | S | R | N1 / N2 | S | R | N1 / N2 | S | R | N1 / N2 | S | ||||
| legal trans- lation | same | 0 | 0 | 0 / 0 | 0 | 0 | 0 / 0 | 0 | 0.31 | 0.26 / 0.01 / 0.00 | 0.30 | 0 | 0 / 0 | 0 | 7 | 5 / 1 | 8 | 0.87 | 0.73 / 0.08 | 1.04 | 498 |
| moderately different | 0 | 0 | 0 / 0 | 0 | 1 | 1 / 0 | 0 | 2.31 | 1.29 / 0.10 / 0.00 | 0.55 | 3 | 2 / 0 | 1 | 38 | 12 / 5 | 11 | 3.57 | 1.68 / 0.50 | 1.31 | 498 | |
| opposite | 0 | 1 | 0 / 0 | 0 | 3 | 2 / 0 | 0 | |6.82| | |2.73| / 0.33 / 0.05 | 1.04 | 8 | 4 / 0 | 1 | 170 | 31 / 11 | 25 | 12.36 | 3.43 / 1.02 | 1.95 | 1494 | |
| subtitle trans- lation | same | 0 | 0 | 0 / 0 | 0 | 0 | 0 / 0 | 0 | 0.12 | 0.12 / 0.00 / 0.00 | 0.44 | 0 | 0 / 0 | 0 | 4 | 3 / 0 | 5 | 0.47 | 0.41 / 0.00 | 0.91 | 413 |
| moderately different | 0 | 0 | 0 / 0 | 0 | 0 | 0 / 0 | 0 | 0.17 | 0.21 / 0.00 / 0.00 | 0.32 | 0 | 0 / 0 | 0 | 6 | 4 / 1 | 4 | 0.55 | 0.58 / 0.07 | 0.75 | 413 | |
| opposite | 0 | 0 | 0 / 0 | 0 | 1 | 0 / 0 | 0 | 1.31 | 0.70 / 0.03 / 0.00 | 0.74 | 2 | 1 / 0 | 1 | 27 | 10 / 5 | 10 | 2.21 | 1.23 / 0.24 | 1.26 | 1239 | |
| simulta- neous interpre- tation | same | 0 | 0 | 0 / 0 | 0 | 0 | 0 / 0 | 0 | 0.02 | 0.59 / 0.00 / 0.00 | 0.72 | 0 | 0 / 0 | 1 | 1 | 3 / 0 | 9 | 0.13 | 0.40 / 0.00 | 1.28 | 225 |
| moderately different | 0 | 0 | 0 / 0 | 0 | 0 | 0 / 0 | 1 | 0.20 | 0.21 / 0.00 / 0.00 | 1.50 | 0 | 0 / 0 | 2 | 4 | 2 / 0 | 9 | 0.54 | 0.51 / 0.00 | 1.92 | 225 | |
| opposite | 0 | 0 | 0 / 0 | 0 | 0 | 0 / 0 | 1 | 0.81 | 0.64 / 0.02 / 0.00 | |1.52| | 1 | 1 / 0 | 2 | 11 | 7 / 2 | 9 | 1.42 | 0.99 / 0.15 | 1.84 | 675 | |
In table 11, the highest value in the column showing the mean rates for each indicator of difficulty (R, N1 and S) corresponding to each mode is underlined. Those underlined values show that, in all three modes, the highest mean rates for all three indicators were observed when the branching direction of relative clauses was opposite.
Of the underlined values in the column showing the mean rates for each indicator of difficulty, the highest value is also in a box. Those boxed values show that the highest mean rates for reordering and nesting changes were again observed in legal translation, where the mean rate for reordering was nearly 7 place shifts per sentence when the branching direction of relative clauses was opposite, and the mean rate for changes in single nestings was nearly 3 changes per sentence when the branching direction of relative clauses was opposite. The highest mean rate for changes in semantic relations was again observed in simultaneous interpretation, where the mean rate was more than 1½ changes per sentence when the branching direction of relative clauses was opposite.
The mean rates for each indicator of difficulty observed for various combinations of mode and difference in the branching direction of relative clauses are visualized in chart 5.
Chart 5. Mean rates for indicators of difficulty observed per sentence
for combinations of mode and difference in branching direction of relative clauses
Reordering:
highest with opposite branching
direction of relative clauses,
especially in legal translation
Changes in single nestings:
highest with opposite branching
direction of relative clauses,
especially in legal translation
Changes in semantic relations:
highest with opposite branching
direction of relative clauses,
especially in simultaneous interpretation
[Hover mouse over bars to see values.]
The other structural variable considered was difference in the branching direction of complement clauses. Table 12 shows summary statistics for the three indicators of difficulty observed for various combinations of mode and difference in the branching direction of complement clauses.
Table 12
Statistics on indicators of difficulty observed per sentence
for combinations of mode and difference in branching direction of complement clauses
(R = Reordering, N1/N2/N3 = Changes in single/double/triple nestings, S = Changes in semantic relations)
| Mode | Branching direction of complement clauses | Mini- mum | First quartile | Median | Mean | Third quartile | Maximum | Standard deviation | Number of obser- vations |
||||||||||||
| R | N1 / N2 | S | R | N1 / N2 | S | R | N1 / N2 / N3 | S | R | N1 / N2 | S | R | N1 / N2 | S | R | N1 / N2 | S | ||||
| legal trans- lation | same | 0 | 0 | 0 / 0 | 0 | 0 | 0 / 0 | 0 | 1.87 | 1.18 / 0.16 / 0.03 | 0.53 | 2 | 2 / 0 | 1 | 39 | 17 / 11 | 11 | 3.71 | 1.87 / 0.63 | 1.22 | 1494 |
| opposite | 0 | 1 | 0 / 0 | 0 | 5 | 2 / 0 | 0 | |8.73| | |3.15| / 0.31 / 0.03 | 1.18 | 10 | 4 / 0 | 2 | 170 | 31 / 11 | 25 | 14.36 | 3.79 / 1.06 | 2.20 | 996 | |
| subtitle trans- lation | same | 0 | 0 | 0 / 0 | 0 | 0 | 0 / 0 | 0 | 0.22 | 0.26 / 0.00 / 0.00 | 0.46 | 0 | 0 / 0 | 1 | 10 | 6 / 1 | 6 | 0.69 | 0.65 / 0.06 | 0.93 | 1239 |
| opposite | 0 | 0 | 0 / 0 | 0 | 1 | 0 / 0 | 0 | 1.78 | 0.84 / 0.04 / 0.00 | 0.81 | 3 | 1 / 0 | 1 | 27 | 10 / 5 | 10 | 2.49 | 1.36 / 0.29 | 1.34 | 826 | |
| simulta- neous interpre- tation | same | 0 | 0 | 0 / 0 | 0 | 0 | 0 / 0 | 0 | 0.16 | 0.26 / 0.00 / 0.00 | 1.10 | 0 | 0 / 0 | 2 | 4 | 5 / 1 | 9 | 0.50 | 0.65 / 0.05 | 1.64 | 675 |
| opposite | 0 | 0 | 0 / 0 | 0 | 1 | 0 / 0 | 1 | 1.08 | 0.66 / 0.03 / 0.00 | |1.74| | 1 | 1 / 0 | 3 | 11 | 7 / 2 | 9 | 1.61 | 1.03 / 0.17 | 1.92 | 450 | |
In table 12, the highest value in the column showing the mean rate for each indicator of difficulty (R, N1 and S) corresponding to each mode is underlined. Those underlined values show that, in all three modes, the highest mean rates for all three indicators were observed when the branching direction of complement clauses was opposite.
Of the underlined values in the column showing the mean rates for each indicator of difficulty, the highest value is also in a box. Those boxed values show that reordering and nesting changes were again observed most frequently in legal translation, where the mean rate for reordering reached nearly 9 place shifts per sentence when the branching direction of complement clauses was opposite, and the mean rate for changes in single nestings reached more than 3 changes per sentence when the branching direction of complement clauses was opposite. Changes in semantic relations were again characteristic of all three modes, especially simultaneous interpretation, where the mean rate reached nearly 2 changes per sentence when the branching direction of complement clauses was opposite.
The mean rates for each indicator of difficulty observed for various combinations of mode and difference in the branching direction of complement clauses are visualized in chart 6.
Chart 6. Mean rates for indicators of difficulty observed per sentence
for combinations of mode and difference in branching direction of complement clauses
Reordering:
highest with opposite branching
direction of complement clauses,
especially in legal translation
Changes in single nestings:
highest with opposite branching
direction of complement clauses,
especially in legal translation
Changes in semantic relations:
highest with opposite branching
direction of complement clauses,
especially in simultaneous interpretation
[Hover mouse over bars to see values.]
This section has presented the main findings produced by analyzing the observed corpus data for this study. Based on that data, a statistical model was estimated and applied, to predict the mean response of each dependent variable to different groups of independent variables in interaction. Those predictions – which can be applied to similar texts, talks and speeches not in the corpus – are presented in the next section.
← 4.2.1 Procedure
→ 4.2.3 Predictions