What have been the most important findings in SLA research on instructed/classroom second language learning in recent years? What have been their main influences on ELT?
With the growth of our global village and with our seemingly permanent state of interconnectedness the need for a worldwide common tongue is paramount for many people; more often than not that common tongue is English. People have sought to learn such lingua franca for millennia, yet the study of how people acquire this 'second language' is relatively recent [Ellis, 1997/Gass and Selinker, 2008]. Additionally, the term 'second language acquisition' is something of a misnomer, not simply referring to the study of the second language learned additionally to a mother tongue but to any language learned which is not the mother tongue [Ellis, 1997]. In this essay I will discuss the growth of the study of SLA from Applied Linguistics, the importance of its first dominant paradigm, namely Stephen Krashen's Monitor method, various other approaches such as behaviourism and other aspects of cognitive psychology before reflecting on the influence of each and their current relevance to the world of ELT.
No discussion of SLA would be complete, however, without a mention of Chomsky's Universal Grammar. This theory was strongly influential on second language acquisition research [Mitchell and Myles p52]. Chomsky considered “.... each language [to be] the result of two factors: the initial state and the course of experience. We can think of the initial state as a 'language acquisition device' that takes experience as 'input' and gives the language as an 'output' – an 'output' that is internally represented in the mind/brain.” [Chomsky, 2000 p4 quoted in Mitchell and Myles, p52]. He sees the acquisition of language, both mother tongues and second languages to be subject to the principles of an inherent structure of predominantly functional categories and that the role of the learner the traits of the language the hope to acquire is to overlay that language's examples into the pre-existing mental compartments, dubbed as 'lexical parameterization hypothesis' [Mitchell and Myles p67]. It must be noted however that Universal Grammar is a linguistic, not a learning, theory. While it is still highly relevant and influential today it does not consider language as a social tool and the learning of it from without, unlike many more current (learning and language) theories. It is therefore more of an interesting side abstraction than of direct pertinence to our question.
Perhaps the most influential thinker directly relevant to our question then is Stephen Krashen. His ideas came to prominence during the early 1980s and seemed more useful and more applicable to the classroom than the audio-lingual (based on an antiquated behaviourist approach) and grammar-translation approaches that had held sway until that time [Krashen, 1982]. His main thrust was that language acquisition was a result of comprehensible input and that formal learning did not result in language production. This became the dominant paradigm of the time in SLA. Krashen's most important and influential ideas were:
i) the acquisition-learning hypothesis
ii) the natural order hypothesis
iii) the monitor hypothesis
iv) the input hypothesis
v) the affective filter hypothesis.
The acquisition-learning hypothesis is the idea that we need separate and distinct terms for gaining expertise and ability in a language (acquisition) and understanding and developing the more formal mechanisms of the language itself (learning). The natural order hypothesis posited that the order in which we acquire a language is predictable in terms of its grammar. The monitor hypothesis is the idea that rules which are 'learned', that is to say formally attained, can only be utilised effectively as an 'editor' [Romeo], my inference of which being that production of acquired language should be given more free rein. The input hypothesis is the idea that beyond the field of a learner's language there exists a zone of meaning that is attainable, through 'acquisition', through context and deduction rather than through teaching. The affective filter hypothesis is the idea that the motivation (or otherwise) to learn another language is key, and that the less motivation the greater the filter. While being and extremely useful approach for language teachers that sat better with many teachers' innate feelings regarding their role it it was criticised, not least for not accepting the merit of formal instruction or other methodology, and that also Krashen had gone too far in some of assertions. That said, Krashen's contribution was, and remains, of enormous significance to the field.
One very common teaching method in EFL these days is Task-Based Learning. This is when a lesson is more focused around a central task and the language studied is that which flows naturally in order to aid the task's completion rather than a more traditional approach where the student is given the language to use and told to practice it in a particular situation or by rote. Students enjoy TBL and it is motivating because the need for the language used is very clear and results are instantaneous as the task either successfully completed or not, depending upon the language used. This approach has its roots in, but has grown beyond, the Behaviorist learning theory, the dominant psychological field of the 1950s and 1960s [Ellis, p31]. The behaviourist learning theory while important and interesting is not adequate alone to account for language learning, as often the output of language learners does not echo the input. My two year old daughter has never heard “That be wrong” (one of her favourite phrases) unless she is secretly watching movies about pirates, and my students often will, for example, add -ness to a verb to make a noun even when it is not the correct form. They cannot be echoing input so what are they doing? My daughter and my students are building a framework of rules, what some call a grammar and that is not accounted for adequately in behaviourist learning theory.
The 1980s saw some interesting findings regarding language acquisition. The growth and later ubiquity of decent computers saw new development of an old approach called connectionism. This idea likens the processes and functionings of the brain to a computer with neural networks being built and creating clusters between informational nodes [Mitchell and Myles, p121]. A learner might notice patterns over time such as in English we say we 'do the housework', rather than 'make' it but that we 'make' a cup of tea, not 'do' it. Rules are not inferred, instead different instances of use are noted and grow according to frequency of use. There is some debate about how true this is, with examples taken from the French use of gender assignment and the English past tense, both of which have regular and irregular components allowing observation of the development of the correct form [Mitchell and Myles, p125]. Cognitive psychology has had a strong influence over the study of SLA, not surprising given that language and cognition are so closely intertwined. A significant model of learning borrowed from cognitive psychology is Anderson's ACT model [see Appendix A]. Learning takes place in the model in three stages [Mitchell and Myles, p103]:
1) The cognitive stage.The learner starts to understand the problem in the abstract. 2) The associative stage.The learner develops a method to perform the skill. 3) The autonomous stage.The skill becomes second nature requiring less and less thought.
How could this apply to language teaching? Imagine a student struggling with a rule such as the English language’s third person singular 's'. In stage 1 the student would learn the rule, perhaps from a teacher in class or a grammar book. They would understand this quirk arbitrarily exists and go to stage 2, seek to apply it when needed. It might come more easily in the written form, maybe through exercises in a textbook designed to check this concept. However in spoken English our student might forget the form, or over apply the form (perhaps incorrectly to the third person plural form) or make other feasible but incorrect combinations. Learning and correct usage of the third stage of this form in my experience comes very late, at level B2 or even later, but students make fewer and fewer mistakes until the verb is conjugated correctly without thought. Anderson's model has been applied to language learning strategies by O'Malley and Chamot [Mitchell and Myles, p107] who say that by “focussing on selected aspects of new information, analysing and monitoring information during acquisition, organising or elaborating on new information during the encoding process, evaluating the learning when it is completed, or assuring oneself that the learning will be successful as a way to allay anxiety.” [O'Malley and Chamot, 1990, p43 as seen in Mitchell and Myles, p105]
Sociocultural theoryis another significant field of SLA. This has become increasingly important in linguistics fairly recently despite being based upon the work of the Soviet psychologist from the 1930s. The argument is broadly that cognitive development (which includes the learning and development of languages) is a result of social interaction through his “zone of proximal development ” [Vygotsky, 1986]. It has been a fairly popular idea in EFL because it accords nicely with the communicative language approach, currently the dominant English Foreign Language teaching paradigm.
What English teachers have actually gained from the study of SLA is a difficult question to answer simply. There are some fairly uncontroversial statements that can be made such as variation in rates of progress and language attainment are quite closely related to the age at which a student starts to learn their L2, what their attitude to that language is and how motivated they are to learn it [Long 1990, p657] but these seem almost so well known at to be common knowledge. Additionally we have the concept of 'interlanguages' [Ellis, p33], an internal bridging mechanism for the language learner between their L1 and L2, a mental construction or scaffolding that serves as a functional but orthographically incorrect version of the language allowing communication to happen and to be the springboard for later L2 growth. We also know that those learning acquire new aspects of said language when they alone are ready to learn them because they follow 'developmental sequences' [Pienemann 1985]. Lightbown [1985, pp 176 -180] has brought together some additional commonalities from her studies of the subject.
I) Adolescents and adults do have the capacity to acquire a second language. II) Learners build their own 'interlanguages'.
III) The progression of learning is (broadly) predictable.
IV) Knowing the language is often not enough and sometimes an ingredient is missing to prevent successful communication (similar to Anderson's later second and third stages). V) Mistakes 'fossilise' in adults fairly early in their language learning, which is to say adults cannot get beyond a mistake that nevertheless does not impede communication greatly (perhaps such as the third person singular 's' or apostrophe s in written English). VI) Learning a foreign language is very difficult as their task is vast; language is extraordinarily complex.
In conclusion, it important for teachers to have an awareness and understanding of the research and findings into SLA. These findings are fairly broad in nature and show that we are still struggling to understand both the very core natures of learning and language, as well as second language learning, which has shown to be different from L1 acquisition. An advantage of have empirical research into any subject is that it provides us with yardsticks and measures by which to judge our own prejudices and tendencies, our likes and biases and maybe give us a tool to draw ourselves away from unhealthy ones. The research shows that there is no easy fix, that SLA is a slow process and seemingly cannot be rushed, and the research shows me that perhaps I should be more tolerant of those silly mistakes that I hear repeated again and again; perhaps my students are unable to go beyond that point of development now, perhaps they have fossilised the error. I shall leave you with Lightbown: “Language acquisition research can offer no formulas, no recipes, but is an essential component of teacher education, because it can give teachers appropriate expectations for themselves and their students.” [Lightbown 1985, p183]
Word Count: 2003
Second Language Acquisition, Rob Ellis, Oxford University Press, 1997
Second Language Acquisition (Third Edition), Susan M. Gass and Larry Selinker, Routledge, 2008 Principles and Practice in Second Language Acquisition, Stephen D Krashen, Pergamon Press. Found at http://www.sdkrashen.com/Principles_and_Practice/ March 2013, p3, originally published 1982
SLA Research in the Classroom, Patsy M. Lightbown, The Language Learning Journal, 2003
How Languages are Learned (Fourth Edition), Patsy M. Lightbown and Nina Spada, Oxford University Press, 2006
Great Expectations. Second Language Acquisition. Research and Classroom Teaching, Patsy Lightbown, Applied Linguistics, 1985
The Least a Second Language Acquisition Theory Needs to Explain, M.H. Long, TESOL Quarterly, 1990
Second Language Learning Theories (2nd Edition), Rosamond Mitchell and Florence Myles, Hodder Arnold, 2004
Thought and Language. Lev Vygotsky, The MIT Press, 1986 as seen in the journal 'The Institute of Mind and Behaviour, 1987, Volume 8 Number 1 pgs 175-178
Romeo, Krashen and Terrell’s “Natural Approach”
Found at http://www.stanford.edu/~hakuta/www/LAU/ICLangLit/NaturalApproach.htm
Appendix A: http://www.instructionaldesign.org/theories/act.html
Here's an image I liked of Anderson's ACT model. There are many and none the same but I feel this one is closest to the spirit of his model as applied to language.