Background to the L2 User Perspective

'An L2 user is any person who uses another language than their first language (L1), that is to say, the one they learnt first as a child.'

• an English school-child staying with a family in Germany on an exchange
• Luc Vandevelde the Belgian head of Marks and Spencer in England
• the tennis-player Martina Hingis with Czech L1 being interviewed in English
• a London newsagent using Bengali and English to his customers
• a Canadian trucker with L1 English driving through French-speaking Montreal
• a street trader in Singapore switching between English and two Chinese dialects
• Kirsten Flagstad the Norwegian opera singer singing Wagner in German in New York
• a Greek student using Italian to study in Perugia
• Billy Wilder code-switching from English to German to explain how he directed Some Like it Hot
• a child in Vancouver speaking Chinese at home and English at school
• an Arabic businessman switching to English for e-mails.

Home languages of London schoolchildren

rank: language: percentage of children
1. English 67.86%
2. Bengali+Sylheti 4.51%
3. Panjabi 3.32%
4. Gujarati 3.19%
5. Hindi/Urdu 2.91%
6. Turkish 1.74%
7. Arabic 1.23%
8. English-based
Creoles 1.20%
9. Yoruba 1.16%
10. Somali 0.93%
11. Cantonese 0.77%
12 Greek 0.71%
13. Akan (Twi
(Ashanti) + Fante) 0.67%
14. Portuguese 0.67%
15. French 0.63%
16. Spanish 0.61%
17. Tamil 0.41%
18. Farsi (=Persian) 0.37%
19. Italian 0.28%
20. Vietnamese 0.27%
21. Igbo 0.22%
22. French-based
Creoles 0.20%
23. Tagalog 0.18%
24. Kurdish 0.16%
25. Polish 0.17%
26. Swahili 0.12%
27. Lingala 0.11%
28. Albanian 0.10%
29. Luganda 0.09%
30. Gã 0.09%
31. Tigrinya 0.09%
32. German 0.09%
33. Japanese 0.09%
34. Serbian/Croatian0.08%
35. Russian 0.08%
36. Hebrew 0.07%
37. Korean 0.06%
38. Pashto 0.05%
39. Amharic 0.05%
40. Sinhala 0.05%

Nemser (1971) 'approximative system': 'Learner speech at a given time is the patterned product of a linguistic system, La [approximative language], distinct from Ls [source language] and Lt [target language] and internally structured'.

Selinker (1972) interlanguage: 'the existence of a separate linguistic system based on the observable output which results from a learner's attempted production of a TL norm. This linguistic system we will call interlanguage' (IL). (Selinker, 1972)

Interlanguage is established by 5 central processes: language transfer, overgeneralisation of L2 rules, transfer of training, strategies of L2 learning, communication strategies,

Cook (1991) multi-competence 'the compound state of a mind with two grammars'

• English primary school children who are taught Italian for one hour a week learn to read better in English
• people who speak a second language are more creative and flexible at problem-solving than monolinguals, e.g. Einstein, Nabokov …
• ten days after a road accident, a bilingual Moroccan could speak French but not Arabic; the next day Arabic but not French; the next day she went back to fluent French and poor Arabic; three months later she could speak both
• the Voice Onset Time (VOT) for /t/ /d/ sounds of French people who speak English is different in French from those who don't
• L2 learners rapidly learn the appropriate pronunciations for their own gender, for instance that men tend to pronounce the "-ing" ending of the English continuous form going as "-in’ " but women tend to use "-ing".
• after seeing an American flag, Chinese/English bilinguals are more likely to interpret behaviour of fish as driven by internal forces; after a Chinese dragon as driven by external forces

The concept of multi-competence, defined as 'the compound state of a mind with two grammars', started as part of an argument that UG-oriented SLA research had ignored the problem of two co-existing grammars in the same mind (Cook, 1991). It became clear that this concept could be used in many aspects of language knowledge such as the relationships between the two or more phonological systems, lexicons (Laufer, to appear), pragmatic systems (Pavlenko, 2002), syntactic processes (Cook et al, 2003) sets of concepts (Cook et al, 2002), leading to the idea of an integration continuum between the two or more languages in multi-competence (Cook, 2002). Particular developments from multi-competence were: 
- the re-evaluation of the use of native speakers as the norm in favour of L2 users in their own right
- seeing transfer as a two-way process in which the L1 in the L2 user's mind is affected by the L2 as well as the reverse (Jarvis, to appear)
- looking at the benefits of L2 acquisition on other aspects of the user's linguistic competence and cognition (Kesckes & Papp, 200). 
This has had repercussions for language teaching, in particular seeing its goal as being successful L2 use, not imitation L1 use, and re-instating the valued role of the first language in the classroom. A distinctive research methodology has evolved of comparing L2 users with native monolinguals in both L1 and L2 with a view to establishing the uniqueness of L2 users rather than their deficiencies. Going back to its origins, the power of the concept is, however, that it essentially describes the potential state of any human mind: approaches in linguistics or psychology that restrict language acquisition and use to monolinguals fail to account for what any human mind can do (Satterfield, 1999) and what statistically probably the majority of minds in the world have done.

Satterfield, T. (1999), Bilingual Selection of Syntactic Knowledge: Extending the Principles and Parameters Approach, Kluwer