S is for SLA

1 03 2015

The criteria for evaluating the worth of any aid to language learning (whether print or digital, and, in the case of the latter, whether app, program, game, or the software that supports these) must include some assessment of its fitness for purpose. That is to say, does it facilitate learning?

But how do you measure this? Short of testing the item on a representative cross-section of learners, we need a rubric according to which its learning potential might be predicted. And this rubric should, ideally, be informed by our current understandings of how second languages are best learned, understandings which are in turn derived — in part at least — from the findings of researchers of second language acquisition (SLA).

This is easier said than done, of course, as there is (still) little real consensus on how the burgeoning research into SLA should be interpreted. This is partly because of the invisibility of most cognitive processes, but also because of the huge range of variables that SLA embraces: different languages, different aspects of language, different learners, different learning contexts, different learning needs, different learning outcomes, different instructional materials, and so on. Generalizing from research context A to learning context B is fraught with risks. It is for this reason that, in a recent article, Nina Spada (2015) urges caution in extrapolating classroom applications from the findings of SLA researchers.

Cautiously, then, and following VanPatten and Williams’ (2007) example, I’ve compiled a list of ‘observations’ about SLA that have been culled from the literature (albeit inflected by my own particular preoccupations). On the basis of these, and inspired by Long (2011), I will then attempt to frame some questions that can be asked of any teaching aid (tool, device, program, or whatever) in order to calculate its potential for facilitating learning.

Exposure to input is necessary

Here, then, are 12 observations:

  1. The acquisition of an L2 grammar follows a ‘natural order’ that is roughly the same for all learners, independent of age, L1, instructional approach, etc., although there is considerable variability in terms of the rate of acquisition and of ultimate achievement (Ellis 2008), and, moreover, ‘a good deal of SLA happens incidentally’ (VanPatten and Williams 2007).
  2. ‘The learner’s task is enormous because language is enormously complex’ (Lightbown 2000).
  3. ‘Exposure to input is necessary’ (VanPatten and Williams 2007).
  4. ‘Language learners can benefit from noticing salient features of the input’ (Tomlinson 2011).
  5. Learners benefit when their linguistic resources are stretched to meet their communicative needs (Swain 1995).
  6. Learning is a mediated, jointly-constructed process, enhanced when interventions are sensitive to, and aligned with, the learner’s current stage of development (Lantolf and Thorne 2006).
  7. ‘There is clear evidence that corrective feedback contributes to learning’ (Ellis 2008).
  8. Learners can learn from each other during communicative interaction (Swain et al. 2003).
  9. Automaticity in language processing is a function of ‘massive repetition experiences and consistent practice’ in ‘real operating conditions’ (Segalowitz 2003; Johnson 1996).
  10. A precondition of fluency is having rapid access to a large store of memorized sequences or chunks (Nattinger & DeCarrico 1992; Segalowitz 2010)
  11. Learning, particularly of words, is aided when the learner makes strong associations with the new material (Sökmen 1997).
  12. The more time (and the more intensive the time) spent on learning tasks, the better (Muñoz 2012). Moreover, ‘learners will invest effort in any task if they perceive benefit from it’ (Breen 1987); and task motivation is optimal when challenge and skill are harmonized (Csikszentmihalyi 1990).

On the basis of these observations, and confronted by a novel language learning tool (app, game, device, blah blah), the following questions might be asked:

  1. ADAPTIVITY: Does the tool accommodate the non-linear, often recursive, stochastic, incidental, and idiosyncratic nature of learning, e.g. by allowing the users to negotiate their own learning paths and goals?
  2. COMPLEXITY: Does the tool address the complexity of language, including its multiple interrelated sub-systems (e.g. grammar, lexis, phonology, discourse, pragmatics)?
  3. INPUT: Does it provide access to rich, comprehensible, and engaging reading and/or listening input? Are there means by which the input can be made more comprehensible? And is there a lot of input (so as to optimize the chances of repeated encounters with language items, and of incidental learning)?
  4. NOTICING: Are there mechanisms whereby the user’s attention is directed to features of the input and/or mechanisms that the user can enlist to make features of the input salient?
  5. OUTPUT: Are there opportunities for language production? Are there means whereby the user is pushed to produce language at or even beyond his/her current level of competence?
  6. SCAFFOLDING: Are learning tasks modelled and mediated? Are interventions timely and supportive, and calibrated to take account of the learner’s emerging capacities?
  7. FEEDBACK: Do users get focused and informative feedback on their comprehension and production, including feedback on error?
  8. INTERACTION: Is there provision for the user to collaborate and interact with other users (whether other learners or proficient speakers) in the target language?
  9. AUTOMATICITY: Does the tool provide opportunities for massed practice, and in conditions that replicate conditions of use? Are practice opportunities optimally spaced?
  10. CHUNKS: Does the tool encourage/facilitate the acquisition and use of formulaic language?
  11. PERSONALIZATION: Does the tool encourage the user to form strong personal associations with the material?
  12. FLOW: Is the tool sufficiently engaging and challenging to increase the likelihood of sustained and repeated use? Are its benefits obvious to the user?

Is it better than a teacher?

This list is very provisional: consider it work in progress. But it does replicate a number of the criteria that have been used to evaluate educational materials generally (e.g. Tomlinson 2011) and educational technologies specifically (e.g. Kervin and Derewianka 2011). At the same time, the questions might also provide a framework for comparing and contrasting the learning power of self-access technology with that of more traditional, teacher-mediated classroom instruction. Of course, the bottom line is: does the tool (app, program, learning platform etc) do the job any better than a trained teacher on their own might do?

Any suggestions for amendments and improvements would be very welcome!


Breen, M. P. 1987. ‘Learner contributions to task design’, republished in van den Branden, K., Bygate, M. & Norris, J. (eds) 2009. Task-based Language Teaching: A reader. Amsterdam: John Benjamins.

Csikszentmihalyi, M. 1990. Flow: The psychology of optimal experience. New York: Harper & Low.

Ellis, R. 2008. The Study of Second Language Acquisition (2nd edn). Oxford: Oxford University Press.

Kervin, L. & Derewianka, B. (2011) ‘New technologies to support language learning’, in Tomlinson, B. (ed.) Materials Development in Language Teaching (2nd edn). Cambridge: Cambridge University Press.

Lightbown, P.M. (2000) ‘Classroom SLA research and second language teaching’. Applied Linguistics, 21/4, 431-462.

Long, M.H. (2011) ‘Methodological principles for language teaching’. In Long, M.H. & Doughty, C. (eds) The Handbook of Language Teaching, Oxford: Blackwell.

Muñoz, C. (ed.) (2012). Intensive Exposure Experiences in Second Language Learning. Bristol: Multilingual Matters.

Nattinger, J.R. & DeCarrico, J.S. (1992). Lexical Phrases and Language Teaching. Oxford: Oxford University Press.

Segalowitz, N. (2003) ‘Automaticity and second languages.’ In Doughty, C.J. & Long, M.H, (eds) The Handbook of Second Language Acquisition. Oxford: Blackwell.

Segalowitz, N. (2010) Cognitive Bases of Second Language Fluency. London: Routledge.

Sökmen, A.J. (1997) ‘Current trends in teaching second language vocabulary,’ in Schmitt, N. and McCarthy, M. (Eds.) Vocabulary: Description, Acquisition and Pedagogy. Cambridge: Cambridge University Press.

Spada, N. (2015) ‘SLA research and L2 pedagogy: misapplications and questions of relevance.’ Language Teaching, 48/1.

Swain, M. (1995) ‘Three functions of output in second language learning’, in Cook, G., & Seidlhofer, B. (eds) Principle and Practice in Applied Linguistics: Studies in Honour of H.G.W. Widdowson. Oxford: Oxford University Press.

Swain, M., Brooks, L. & Tocalli-Beller, A. (2003) ‘Peer-peer dialogue as a means of second language learning’. Annual Review of Applied Linguistics, 23: 171-185.

Tomlinson, B. (2011) ‘Introduction: principles and procedures of materials development,’ in Tomlinson, B. (ed.) Materials Development in Language Teaching (2nd edn). Cambridge: Cambridge University Press.

VanPatten, B. & Williams, J. (eds) 2007. Theories in Second Language Acquisition: An Introduction. Mahwah, NJ: Lawrence Erlbaum.

This is a revised version of a post that first appeared on the eltjam site:  http://eltjam.com/how-could-sla-research-inform-edtech/


A is for Automaticity

26 02 2012

Class photo: thanks to Lucy Bodeman

I have a debt to repay.

Sometime in the late eighties I attended a talk given by Stephen Gaies (the then editor of TESOL Quarterly) at the North American Institute in Barcelona. The topic was fluency. Apart from being an excellent speaker with a good line in personal anecdote, Gaies made an indelible impression by outlining, and demonstrating, a set of criteria for the design of activities that target ‘creative automatization’.

Automaticity (and I’m using automaticity in preference to automatization only because it’s marginally easier to pronounce) is defined in An A-Z of ELT as the ability to perform a task ‘without conscious or deliberate effort’:

In language speaking terms, this automatization process means being able to draw on a set of memorised procedures in order to take part in real-time interaction. Without these procedures (or routines) you would have to assemble each utterance from scratch, word by word, at the obvious expense of fluency.

Notice that I talk about ‘memorised procedures’ rather than ‘memorised chunks’. Because if automaticity is simply the ability to retrieve memorised chunks, this may result in a repertoire that is fast and accurate, but functional only in situations of the utmost predictability. Fine, if you’re a tourist – just memorise a phrase-book. But for a more sophisticated command of language  – one that is adaptable to a whole range of situations – you need to be able to customise your chunks. In short, you need to be creative. Hence, creative automaticity.

In his workshop, Stephen Gaies put a strong case for activities that were communicative – in the sense that there was a genuine desire to communicate – but that were also narrowly focused and formulaic – in the way that old-fashioned pattern-practice drills used to be.

As an example, he described the ‘Class photo’ activity, in which the students take turns to ‘pose’ the group, requiring the use of such ‘partly-filled constructions’ as V, stand next to W; X, stand behind Y;  Z, kneel in front of V, and so on. Once posed, the class photo is taken.

Sonia Omulepu and her class

The language that the task generates is communicative, in the sense that it is purposeful and reciprocal,  but also formulaic, while allowing a degree of creativity within relatively tight constraints. Moreover, there is lots of built-in repetition.  Gaies added that, by timing the class photo just ten minutes before the end of the lesson, an extra element of urgency is added, which is also conducive to the development of automaticity.

I was so taken by this idea, and the principles on which it was based, that I failed to register who first thought of it, assuming it was Gaies himself. The five criteria for creative automaticity became a staple of my teacher training sessions, and worked their way, re-phrased and unattributed, into the section on fluency in How To Teach Grammar. These criteria are:

Activities [that promote creative automatization] should be …

1. genuinely communicative  i.e. require students to make use of utterances as a result of a task-related need, rather than simply for the purpose of saying something.

2. psychologically authentic i.e. require students to allocate attentional resources to both the encoding and decoding of language, and to the effect of that language on events.

3. focused i.e. organised around one or a few functions and notions so as to establish particular utterances as characteristic exponents of particular functions/notions.

4. formulaic i.e. utterances must be short, memorizable, and multi-situational.

5. inherently repetitive

Ever since, I have been ‘collecting’ activity types that match these criteria. The classic Find someone who… is an obvious candidate, as are many guessing games, such as What’s my line? or What kind of animal am I? (“Do you have four legs? Can you fly? Do you lay eggs?” etc).

Going native, Cairo 1976

It was only last week, to my shame, that I accidentally discovered who originated these principles, including the ‘class photo’ idea. It appears in an article by Elizabeth Gatbonton and Norman Segalowitz, published in the TESOL Quarterly in 1988. As editor of that journal, Gaies would surely have mentioned this fact, but I was too dim to notice. Hence the debt I need to repay.

Over 20 years later the article still stands the test of time. The challenge of devising tasks that develop automaticity through the rehearsal and real-time deployment of memorised procedures is still as topical as ever – maybe even more so, as increasing credence is given to the view that fluency involves the seamless interweaving of both the second-hand and the new, of the formulaic and the creative, of phrase and grammar.


Gatbonton, E. and Segalowitz, N. (1988) ‘Creative automatization: Principles for promoting fluency within a communicative framework’, TESOL Quarterly, 22, 3.