Transfer and markedness
The evidence reviewed in Reflective task 8 partially goes against what is predicted by the Contrastive Analysis Hypothesis. Interference from the L1 seems to be occurring in one direction, but not in another. In order to explain the German data, Eckman (1977, cited in Gass and Selinker, 1994: 96-97) appealed to a notion of markedness. A linguistic feature is said to be 'marked' in relation to another linguistic feature, when it is in some way 'more unusual' or 'less basic' in a given language or languages. He claimed that the voicing contrast in final consonants is a marked feature in that it is relatively rare in the world's languages. He proposed that learners avoid transferring marked elements from their L1, particularly when the target language has an unmarked form.
Thus, English is 'odd' in having a final voicing distinction, so English learners experience little difficulty in giving up this oddity in favour of a more basic language feature, ie final devoiced consonants. The poor German learners however find it very difficult to abandon their unmarked feature for our odd (marked) one! The same argument goes for the preverbal placement of French object pronouns. The unmarked word order of French is subject-verb-object, as it is in English. Thus, French learners of English should be happy to give up their 'oddity' to adopt the more 'basic' English form, while English learners have to overcome both the influence of their L1 and standard French word order to correctly place object pronouns.
Markedness remains a rather fuzzy concept in SLA. However, it seems to help explain why transfer works in some directions but not in others. In Reflective task 9, we consider further data which shows that learners sometimes ignore their L1 knowledge in developing their interlanguage.
|
Look at the following data concerning conditional sentences in English, French and Dutch. All three languages are broadly similar in that they construct a conditional sentence by 'backshifting' to a simple past tense in the 'If' clause, and then using a conditional marker in the main clause.
There is of course a difference in the way in which English and Dutch, on the one hand, mark the conditional by using a modal auxiliary verb (would or zou), while French marks it with a different ending on the lexical verb (achète+rais). There are also word order differences between Dutch and the other two languages. Based on the Contrastive Analysis Hypothesis, and just focusing on the sequence of verb forms, would you predict that this would be a difficult or an easy structure for English learners of French, French learners of English, Dutch learners of French, Dutch learners of English, etc? Irrespective of the Contrastive Analysis Hypothesis, what would your prediction be? Easy or difficult? Why? Click here for Commentary. |
