3.2.2 Items and fillers

Now that all of the experimental stimuli have been considered individually, these results will be compared to those pertaining to the grammatical and ungrammatical fillers. Therefore, the individual results achieved by the fillers will be illustrated in different tables so that they can be judged in relation to the experimental stimuli.

Grammatical Fillers

Table 11

Table 12

Table 13

Table 14

Looking at the percentages given to the grammatical fillers, the reader will notice that the AE speakers show much more variation in their acceptance of the different items which ranges from 63.3% to 96,7%. In comparison, BE speakers' judgements range from 88,0% to 100,0% and those of the ALE speakers from 80,6% to 94,3%. The reason for this very low rating of the sentence He himself said it was right (6) by the AE group could be explained by the fact that the reflexive pronoun is used emphatically in this context. Consequently, it might be more natural to hear such a sentence in spoken language with the emphatic use of the reflexive pronoun being supported by stress, mimic, and gesture. Another rather low rating is given to the sentence They asked themselves whether to go or not (9) (66,7%). Again, a possible explanation could pertain to the unnaturalness of this sentence to American ears. Using the verb to wonder in this context might be an easier and more usual structure. A corpus analysis concerning the frequency of to wonder and to ask oneself would help to investigate this hypothesis.

Ungrammatical Fillers

Table 15

Table 16

Table 17

Table 18

Table 19

As far as the judgements of the ungrammatical fillers are concerned, there again is a noticeable exception in the AE group's data. Sentence 2, Robin washed itself, achieved 43,3% although it contains an obvious mistake of concord between the subject and the reflexive pronoun. ALE speakers gave similar ratings for this sentence (40,5%). However, they seem to have a general tendency of giving relatively good judgements to the ungrammatical fillers (between 36,9% and 54,7%). This could be due to a general uncertainty of non-native speakers as far as grammatical judgements in a foreign language are concerned. Moreover, it could well be that ALE speakers are simply not as precise in their detection of errors because the violations do not strike them as they would strike a native speaker. Sentence number 1, She troubles myself with grieves, which, astonishingly, was only rated half as bad as the reference sentence (54,7%), constitutes an exception even among the generally high ratings of ungrammatical fillers by the ALE speakers. Interference with the native language German might help to explain this result. In German, one can say Sie geht mir mit....auf die Nerven. Hence, some ALE speakers probably thought that the reflexive pronoun of sentence 1 had the same function in English as mir in German and thus misjudged the meaning of the sentence.

Experimental Stimuli in comparison with the Filler Sentences

Considering Rohdenburg's thesis that the “trend towards the zero variant has affected AmE much faster and more extensively than BrE” (Rohdenburg 2009:180), one could have expected the AE speakers' judgements of the non-reflexive uses of verbs to be as good as those of the grammatical fillers whereas the judgements of the reflexive uses of verbs might be more similar to those of the ungrammatical fillers. However, when comparing the results of the experimental stimuli with those of the filler sentences, one notices that the situation is not as simple as that. Not every verb used non-reflexively is judged as being as good as the grammatical fillers. AE and ALE speakers' judgements of to trouble and to overwork, for example, overlap their judgements of ungrammatical fillers. Only the BE group gives consistently better ratings to the sentences containing the zero variant than to the ungrammatical sentences. Mostly, the percentages calculated for the experimental stimuli lie between those of the grammatical and ungrammatical fillers, thus showing that the experimental stimuli are neither as acceptable as the grammatical fillers, nor as unacceptable as the ungrammatical fillers. Leaving out the generally very positive judgements of the non-reflexive use of to qualify (AE: 96,7%, ALE: 92,8%, BE: 99,6%), this tendency is even more obvious.

Regarding the verbs used in their reflexive form, opposite tendencies can be noticed. With the exception of to keep and to overwork in the AE group, all verbs used in combination with the reflexive pronoun received lower percentages than the grammatical fillers. However, only one word per group (AE: to empty, BE: to keep, ALE: to empty) was judged worse than an ungrammatical filler, meaning again that the rest of the sentences using reflexive verbs, is neither as good as the grammatical fillers nor as bad as the ungrammatical fillers.

But why could this be the case? The way we see it, these 'in-between' judgements of the experimental stimuli might be caused by the fact that neither the non-reflexive nor the reflexive use of the verbs mentioned in this study is grammatically correct or wrong as such. It is rather a question of individual preferences. Thus, it could well be that both solutions are not seen as completely acceptable nor unacceptable, especially when compared to the relatively clear-cut examples of grammatical and ungrammatical sentences.

Created with the Personal Edition of HelpNDoc: Full-featured Documentation generator