The one about the horse who walked into a bar: taking a step back

by James Pengelley

A horse walks into a small, smoky bar and takes a seat.  What happens next?  The answer to that question depends entirely on where the bar happens to be.

What can I get you?

We don’t serve your type in here.

I’m sorry, can I see your ID please?

Why the long face?

The response you expect in this situation is based upon your ordering a drink at a barscript.  The notion of scripts was introduced by Abelson and Schank (1977) as predictable and stereotyped patterns of interaction that allow speakers to form expectations about both the sequence and manner of an event, thus reducing workload of the real-time processing of speech.  The concept of scripts comes from the cognitive psychological concept of schema.  Schema allow us to describe the way human cognition is reflected in the associations we attribute to our surroundings and the way we integrate incoming information with existing knowledge.  Schema explain why, for example pre-listening activities that involve discussions, lexis and prediction tasks make it easier for learners to deal with a listening text – because the process of pre-listening tasks activate schematic (thematically related) knowledge and allow learners to make predictions and assumptions about what they are about to hear.

In the same way, scripts allow speakers to make assumptions about the content and the manner of spoken exchanges, and ensure speakers do not have to process spoken interactions word for word, in real-time.  When you walk into a bar, you expect to be greeted and have someone take your order.  And most of the time, it works, you get your drink, you pay for it, everyone kicks a goal and if you’re really lucky, you don’t go home alone!

That is, of course, assuming you’ve got all those assumptions you’ve been making correct.  Enter the English language learner.  Learners are, by their nature, victims of restricted opportunities to observe how people interact in English and many of the issues we, as teachers, strive to address in the classroom arise from what we call L1 interference.  If I say x in Spanish, I should say y in English: A symptomatic response from an overly product-driven approach to the reality of real-world language use.

I recently presented a group of upper-intermediate Colombian general English learners with a scenario (Holmes & Riddiford, 2011):

Staying late tonight

An unexpected and urgent request from the CEO means that you would like your secretary, Mrs Jenny Smith, to stay late tonight to help you prepare a report. You have worked with your secretary for three years. Discuss with your secretary if she can stay on at work for two extra hours.

First, I asked my learners to answer a series of questions encouraging them to predict where the conversation would happen, the nature of the speakers’ relationship, and how the speakers feel at the beginning of the dialogue.  The questions were designed to focus learners’ attention on what is referred to as the sociolinguistic context: the who, what, where and why that determines appropriate register and style (and ultimately the appropriate forms) for the dialogue to continue – but notice here that choice of form is the final decision learners make.  It is a result of the preceding process that grounds subsequent conversation firmly in a social context.

Interestingly, the vast majority of the group’s responses included things like:

–          “It’s happening in an office”

–          “It is a formal conversation, the secretary has no choice”

–          “He feels maybe a bit embarrassed”

When we compared the group’s performance and choices of language we discovered that, in fact, their choice of interaction patterns, sequence of moves and selection of forms was almost an identical English equivalent of 2 Colombian speakers performing the same task in Spanish.

What the group did not realize was that this was a result of their prediction being based on assumptions that work in Colombian culture: direct requests, a subservient receptionist and an unequal distribution of status. The script my learners were relying on was founded in assumptions about how a request should be worded (i.e. what should be said), how the speakers should fulfill their roles as “boss” and “secretary” and the manner in which these characters would normally interact in Colombian society.

When we compared learners’ examples to two English speakers (colleagues of mine who have management experience in the USA and Australia) performing the task, we identified a number of interesting features: the use of direct requests embedded in hedges (I hate to do this to you but… / I hope I can count on you, this is a big deal for me) and a degree of compromise and negotiation between the speakers.

I would argue that applying a purely product-focused approach (i.e a focus on what the speakers said) to this scenario, would, in practice only be useful to my learners when they were in situationally identical scenarios: speaking to their bosses or secretaries, it was a Friday afternoon, and they felt embarrassed – how often can any one person say they have been in this exact situation often enough to warrant it forming the linguistic objectives of an entire group of general English learners?

Ironically, The Communicavist’s drive to empower his/her learners to do things with the language taught in class may actually be doing more harm than good.  The communicative approach is arguably reductionist in its view of language being a tool to fill information gaps, and of communicative competence being simply a matter of matching form x to situation y (Byram, 1990 in McConachy, 2009) which risks producing a structuralist classroom reality – a far cry from what many communicavists believe they are actuallydoing.  The reality of language is that it is a social tool and to be used correctly, it must reflect the assumptions, the rules and the patterns that govern the way people interact within a society and that linguistic choices will vary between cultures.  This essentially means “taking a step back from a narrow focus on linguistic forms and ask[ing] what happens to people’s language when they interact socially?” (Tarone & Yule, 1989, p. 93), and for teachers, going beyond what should be said in a given situation, and exploring whyit might be said by a native speaker.

In a world where learners are encouraged to see comprehension of a spoken exchange as simply finding information, and appropriate contributions to that conversation being dependent on the production of memorized phrases for requests, buying or giving suggestions, learners are left with a limited view of how native speakers actually perform.

When Canale and Swain (1980) identify four components of a speaker’s ability to communicate – grammatical, discourse, strategic and sociolinguistic competence they are pointing out just this: speaking is viewed as a balance of what, when, who, where and why.  So, speaking ability suddenly becomes a balance of manipulating forms in a given context, the process of choosing the grammar that I know, the way I manage the situation to say what I want to say, the tools I use to fill the gaps and the way I do all that appropriately to the situation I’m in and the people I’m speaking to.

Contextual factors are just as important if not more, if we are to promote language as a phenomenon that evolves from relationships between people rather than a mechanical tool used to trade information (McConachy, 2009).

For my learners, the problem wasn’t so much not knowing what to say, but not being able to explain the differences between their model and the English model: the why.  For them, the English model was very strange and reflective of a social relationship governed by parameters that they would never have predicted, and struggled to comprehend.

In fact many of them wanted desperately to draw conclusions about one conversation being slightly more polite than the other.  Politeness is not to be confused with style: it is the contextually appropriate application of style and genre, and thus politeness is not something you do, it is “encoded in a language” (Thornbury, 2005, p. 19) and something that happens when speakers make the right linguistic choices.

We come from a school of thought that champions authentic language use in real-world situations. But as language teaching evolves, and the spread of English takes it into classrooms and cultures in which both the native-speaking teacher and native English cultures are an anomaly, we may very easily be led to make assumptions about the way our students ought to interact in English.  A product-driven approach to speaking is likely to assess communicative competence in terms of what the learners say, rather than the key contextual factors that determine why certain structures and interaction patterns are acceptable. Which makes me wonder, really, not about what the horse in the bar said to the barman, but why he was there in the first place.

Author’s Bio: James has been working as Senior Teacher for IH Bogota since 2012, having previously taught in Thailand, London and Australia. He has also run several sessions at recently IHWO online conferences and seminars. He is particularly interested in psycholinguistics, teaching phonology, homebrewing and baking and in between finishing DELTA studies he maintains a regal authorship of his blog Wider Horizons of The Hairy Chef.