3.2.1 Individual experimental items

Table 6

It is important to notice that the two sentences with the verb to empty do not show any semantic difference. Therefore, there is not much of a difference between choosing the reflexive pronoun or the zero variant. Interestingly enough, all three groups of informants prefer the zero variant to the reflexive pronoun. This tendency seems least significant for BE speakers, while AE and ALE speakers show a clear preference for the zero variant. The reason why ALE speakers follow the AE model might be interference with their mother tongue, i.e. Germans would use münden in for English to empty (itself) into in this context.

Comparing these results to Rohdenburg's findings, the reader will find an interesting similarity between Rohdenburg's corpus-based research and the results pertaining to acceptability which are gained by using Magnitude Estimation. The corpora contain 0% of reflexive use of the verb to empty for AE and 9,8% for BE (cf. Rohdenburg 2009: 170). Accordingly, the AE speakers participating in our study rated the sentence containing the reflexive pronoun with a much lower percentage than the BE speakers, namely 38,3% for AE speakers and 65,6% for BE speakers (cf. Rohdenburg 2009: 170). Turning, however, to the results concerning non-reflexive use, our findings are contradictory to those of Rohdenburg. While he claims that “the centuries-old trend towards the zero variant has affected AmE much faster and more extensively than BrE” (Rohdenburg 2009: 180), our AE informants revealed a lower acceptance of the non-reflexive sentence than the BE speakers. Perhaps this difference is due to the smaller number of informants in the AE group or the fact that the BE speakers are very young and their language might therefore show more progressive tendencies (cf. Eckert 1997: 163).

What has to be kept in mind is that our informants were not asked to judge whether they would find a sentence acceptable in written or spoken usage, but simply how generally acceptable they found a sentence. By contrast, Rohdenburg used British and American newspapers for his research concerning to empty (cf. Rohdenburg 2009: 170) so that we cannot clearly say whether the trends found in our study concern spoken or written language or both of them.

Table 7

According to Rohdenburg’s thesis, AE speakers tend to drop the reflexive pronoun more often than BE speakers (cf. Rohdenburg 2009: 167). The results of the mini-study, however, contradict Rohdenburg inasmuch as the AE speakers in the case of sentences 10 and 12 chose sentence 10 to be more acceptable. But when interpreting the data, it has to be kept in mind that the use of the verb to trouble shows a semantic difference. In sentence 10, to trouble is used in the sense of 'to torture oneself', whereas in sentence 20 it is used in the meaning of 'to care about'. Thus, the slight preference for sentence 10 could simply be a result of the different semantic use of the verb in combination with or without the reflexive pronoun.

Table 8

It is striking that informants of all three groups judged sentences 12 and 17 similarly. As expected, AE speakers clearly prefer the non-reflexive structure. In opposition to Rohdenburg, however, the acceptance of the zero variant by BE informants is even higher than that by the AE group of informants. Further, it is noticeable that the BE informants also prefer the zero variant to the reflexive structure. These results could be explicable by the age of the BE informants, which is between 15 and 16. Again, it could be assumed that the younger BE informants tend to use more innovative structures than the older AE speakers. Moreover, all informants of the BE group share the same classroom and education, probably assimilating their language use to each other. Thus, their judgements could be influenced by the peer group and explain the overall high acceptability given to sentence number 17.

As noticed earlier, the ALE informants rated sentence 17 to be one of the top three of the most acceptable sentences, which illustrates that they, too, chose the zero variant as the more acceptable solution. Generally, this might be due to the fact that speakers tend to minimise the linguistic structure when speaking in order to facilitate the communication, which is a pragmatic phenomenon (cf. Glück 2000: 665f.).

Table 9

Similarly to the judgements of the sentences 12 and 17, the BE informants preferred the zero variant to the reflexive structure while it is the other way round for speakers of AE. Thus, our results stand in opposition to Rohdenburg’s findings. Again, we assume the results of the mini-study represent a very particular and homogeneous group (in age and social background) of BE native speakers that might tend to use language in a less formal way. As far as the ALE speakers are concerned, the results show a clear preference of the reflexive to the zero variant, which might be a German interference (cf. Sie hat sich überarbeitet). The AE informants’ ratings were equally surprising as they, too, preferred the reflexive structure to the zero variant.

Table 10

Interestingly, the BE informants judged the acceptability of these two sentences in complete opposition to the two other groups of informants. The AE and the ALE group show the same trend of accepting the reflexive structure and judging it higher than the zero variant. The BE informants ranked sentence 16 the third lowest of all, although it is a grammatically correct sentence. These results prove the fact that in spoken language the speaker’s acceptability judgements of a linguistic item do not necessarily depend on its grammaticality.

Created with the Personal Edition of HelpNDoc: News and information about help authoring tools and software