Background to L1 interference
The influence of the mother tongue: a summary of the research
Paul Shoebottom teaches English at Frankfurt International School
A German student in your English Language B class who claims: "I must not go to the assembly because I have already the informations" is clearly making a statement containing various language errors. If you know German, you will suspect that the student wanted to convey the following thought:
Ich muss nicht zur Versammlung gehen, da ich die Informationen schon habe.
What is interesting here is that only some aspects of the original German have been negatively transferred into English. For example, muss nicht has become must not instead of don't have to; informations reflects the fact that German uses the plural form. On the other hand, the order of words in the independent and dependent clauses has not transferred into English, otherwise the student would have said something like:
I must not to the meeting go because I the informations already have.
The above examples of selective negative transfer encapsulate the history of research into second language learner errors. Essentially, this began in the 1950's in a field of study called Contrastive Analysis. Linguists at the time based their investigations on the hypothesis that second language errors are the result of interference, or negative transfer, from the first language. Lado, for example, claimed that: "Those elements which are similar to [the learner's] native language will be simple for him, and those elements that are different will be difficult" (as cited in Odlin, 1989).
In the 1970's, however, Contrastive Analysis was largely abandoned as a useful predictive instrument of learner error. Further research had clearly demonstrated that in fact some learner errors have nothing to do with the influence of the mother tongue, but are based on the learner's own developing system of rules (usually called interlanguage). Some of these rules develop in a predictable sequence that appears to be followed by all learners irrespective of their mother tongue, and is similar to the sequence followed by native-English speakers learning their own language. Furthermore, it was determined that many differences between the first language and English do not typically manifest themselves negatively in actual English production. This is seen in the example above where German word order was not transferred into English. Jarvis and Pavlenko (2007) summarize the consensus of this second wave of researchers in error analysis:
Contrary to the strong version of the Contrastive Analysis Hypothesis, differences between the source and recipient languages do not necessarily lead to learning difficulties or to CLI (Cross-linguistic Influence).
The last twenty years or so has seen a proliferation of CLI research into the large number of language learning variables (for example, learner characteristics such as age and native language, and the learning context). This research has proved conclusively that not all learner errors are interference errors, while at the same time re-confirming that some errors are indeed based on negative transfer, and offering more reliable predictions on when they are most likely to occur. In summary, a CLI error is more probable when the particular feature (grammatical, semantic or phonological) shows significant differences between the native and the target language. Conversely, it seems that negative transfer errors are also likely if the two languages share similar but not identical features, leading learners to overgeneralizing what is correct or possible in the target language.
In summary of the current consensus it is helpful to cite Swan (2001):
There is less disagreement than there used to be about how far interlanguages are influenced by learners' native languages, and most linguists would probably now agree that the mother tongue can affect learners' English in several ways.
While not all of a learner's problems, by any means, are attributable to direct mother-tongue 'interference', the overall patterns of error do [...] tend to be language-specific.
This consensus reflects my daily experience of the common errors I see in student work in over 30 years of teaching English as a second language. German learners, for example, are more likely than Japanese learners to make mistakes of sentence structure, and less likely to make article mistakes. It is to be hoped that further research will provide teachers with the means to predict with greater accuracy and detail the language transfer errors made by the various learners in their learning contexts.
See below for relevant references and links to more information on the history of research in this field. The seminal works are those by Odlin and Jarvis & Pavlenko.
For more on typical errors made by German learners of English, see the reference page German .
- Odlin, Terence. Language Transfer: Cross-Linguistic Influence in Language Learning. Cambridge Applied Linguistics, 1989. Print.
- Jarvis, Scott, and Pavlenko, Aneta. Crosslinguistic Influence in Language and Cognition. Routledge, 2007. Print.
- Lightbown, Patsy, and Spada, Nina. How Languages are Learned. 4th ed. Oxford, 2013. Print.
- Mitchell, Rosamund, Myles, Florence, and Marsden, Emma. Second Language Learning Theories. 3rd ed. Routledge, 2013. Print.
- Swan, Michael, and Bernhard Smith, eds. Learner English: A Teacher's Guide to Interference and Other Problems. 2nd ed. Cambridge University Press, 2001. Print.