Paragraph Translation Task

Table 4.8 exhibited the frequency of null properties in the paragraph translation data by the two learner groups. As a whole, the LP group dropped more pronouns than the HP group in terms of all the pro-drop cases. When we further investigate their

performance on each sentence, we found several interesting patterns.

Table 4.8 The percentage of each null property performed by two groups

LP (n=56) HP (n=76)

Null properties

N % N %

null subject (*5) 13/195 6.67 8/254 3.15

null object (*5) 57/270 21.11 47/349 13.47

null it-time (*2) 14/91 15.38 7/129 5.43

null it-raising (*1) 2/11 18.18 1/28 3.57

NST (*1) 19/52 36.54 10/68 14.71

null there (*1) 4/51 7.84 1/74 1.35

that-trace effect (*1) 15/19 78.95 12/41 29.27

S-V inversion (*2) 8/63 8.60 6/138 4.35

Note 1: N= X / X+Y (where X=number of instances of null subjects, and Y= numbers of cases where null subjects could have occurred in L1 but did not)

Note 2: (*possible occurrence in the original text)

First of all, learners seem to be affected by their L1 knowledge on their CIL grammars. We can see that both groups performed worst at translating the sentence (34a). High proportion of EFL learners produced sentences like (34b) with that-trace effects. Most participants seem to have difficulty in coping with this kind of sentence pattern in English. In fact, there are only 4 learners in the LP group and 29 learners in the HP group whose translations are grammatical sentences like (34c), while many other learners tend to be affected by the Chinese syntactic structure and translated it into (34d).

(34) a. 你覺得誰掉了這隻狗?

b. *Who do you think that lost this dog?

c. Who do you think lost this puppy?

d. *Do you think who lost the dog?

Besides, both groups also performed worse at the sentence (35a). None of them started this sentence with a verb, but they seem to transfer the topic-prominent

structure in Chinese to their translation in English, which brings about the ungrammatical sentence like (35b). It is similar to the findings in the GJ task, where they fail to detect the ungrammatical sentence with a non-subject topic.

(35) a. 高雄下了很大的雨

b. *Kaohsiung rained cats and dogs in the afternoon

Secondly, as we mentioned before, if EFL learners succeed in resetting a parameter to the new L2 value, they are expected to disallow both null subjects and null objects. In our design, there are five possible places that null subjects may occur in Chinese. Most learners had no problem in producing sentences with overt subjects;

however, it should be noticed that objects seem to be dropped more easily than subjects. There is an obvious asymmetry of null subjects and null objects for both groups, which is identical to the result in the GJ task.

Thirdly, the learners performed differently on the three kinds of expletives: there, it-time, and it-raising. Both groups performed better at sentences with there than

with it-time and it-raising. Most learners had no problem to use ‘there be V’ sentence pattern to translate Chinese sentence (36a) into English, like (36b), but still some learners dropped there be or use have instead, like the example (36c, d).

(36) a. 公園裡只有幾個人

b. There were a few people in the park.

c. *Only few people in the park.

d. *The park only have few people.

Compared with the good performance of expletive there, some learners still had problems with the expression of time. The asymmetric result of null expletive it and there is similar to those in GJ task. In the translation data of Chinese sentence (37a), most learners produced grammatical sentences with expletive it, but there are 8 LP

participants that dropped expletives it or it+be V in the English, as shown in the examples (37b-e):

(37) a. 約翰很開心，看看時間，已經 8 點了

b. *John was so happly, and looked at the clock, __ was already 8 o’clock.

c. *John is happy, look the time __ is already eight o’clock.

d. ?John was happy, and found __ already 8 o’clock.

e. ?John was happy. He saw the watch, already 8 clock.

As for the sentence (38a), almost all participants produced grammatical sentences with overt it, like (38b), except that nine participants drop both it and be V as in (38c).

However, this kind of utterance can also be found in the colloquial English, like the examples (39a, b) in BNC:

(38) a. 太晚了 b. it’s too late c. too late

(39) a. Too late for dinner, I know, but not, I hope, for a drink.

b. Neither of them back until eight o'clock, too late even to see the children.

When it comes to the expletive it-raising, there are two ways to express the underlined part of this sentence (40), as shown in the sentences (41a, b):

(40) 後來，突然出現了一隻小狗，似乎已經迷路了 (41) a. It seems that S+ V

It(expletive) seems that it got lost.

b. Sb. seems to do sth / Adj.

It(the puppy) seems

Thirty-nine participants produced sentence like (40a), while forty-eight participants produced sentences like (40b). Among those who produced sentences like (40a), only three participants dropped it in the sentences, indicating that the EFL learners still

to get lost.

lost.

preferred to use overt expletive it-raising in their production data.

Fourthly, ungrammatical sentences with VS order only exhibited 8.6% for the LP group and 4.35% for the HP group. This finding, similar to that in GJ task, suggests EFL learners were aware of the obligatory subjects at the head of the sentences.

Above 80% of learners for each group produced grammatical sentences like (42b,c) and (43b,c).¹⁶

(42) a. 突然出現了一隻小狗 b. a puppy suddenly appeared c. there appeared a puppy suddenly (43) a. 來了一個警衛

b. a guard came c. there came a guard

Finally, when we compare null subjects with null expletives, we can find that null expletives (11.25%) occurred more than null subjects (6.67%) especially in the LP group. The high percentage of errors in sentences with expletives is probably because about 3.8% of the LP learners tend to omit both expletive and be V together.