R is for Representation

14 04 2013

ELT Journal debate: IATEFL Conference 2013 (photo courtesy Jessica Mackay)

In the debate sponsored by the ELT Journal at this week’s IATEFL Conference in Liverpool, I proposed the motion that published course materials do not reflect the lives nor the needs of the learners.

To challenge some of the assumptions inherent in much published teaching material, I used this mock-up of a coursebook page (see below). It wouldn’t have been ethical, or even legal, to have shown, and then criticized, pages from current coursebooks, but I think you’ll agree that the replica is a plausible one.  I used this to argue that the choices, both of images (more often than not taken from the same kinds of photo archives as the images used in advertising) and of the text accompanying the images, serve to ‘position’ the user to assume a certain kind of identity with respect to the language that they are learning.page1

The physically-attractive, ethnically-mixed, well-dressed and youthful characters are surrounded by iconic consumer items that reflect their upwardly mobile, middle-class aspirations: they exemplify the observation made in a recent survey of general English courses by Tomlinson and Masuhara (2013: 248) that ‘there seems to be an assumption that all learners are aspirational, urban, middle-class, well-educated, westernized computer uses.’

Moreover, the questions they are asking – and the language choices available to answer them with – make certain assumptions about their (and, by extension, the student’s) economic status, sexuality, and agency. As I pointed out, somewhat facetiously, the questions Are you married? What’s your job? and Where do you live? do not invite, and may even preclude, a response along the lines of: Actually I’m gay and unemployed, and I’ve been sleeping on the sofa at my folks’ place ever since the bank re-possessed my apartment.  And you?

As it happens, a search through a number of intermediate-level coursebooks widely used in Spain finds little or no reference to an economic situation in which 50% of under-30s are out of work. The words unemployed, on the dole, out of work simply do not appear. Struggle, inequality, deprivation, etc have been air-brushed out of the picture.  As the authors of a survey of general education textbooks in the US note: ‘The vision of social relations that the textbooks we analysed for the most part project is one of harmony and equal opportunity — anyone can do or become whatever he or she wants; problems among people are mainly individual in nature and in the end are resolved’ (Sleeter & Grant, 2011: 205).

The cheery, sanitized, even anodyne, world of the coursebook has, of course, been endlessly targeted for criticism. In fairness, it is not the fault of the coursebook writers themselves (who generally would love to include more ‘edgy’ content), but more an effect of the constraints that they have to work within. These include the authors’ guidelines that many publishers impose, including the famous ‘PARSNIP’ proscriptions (no Pork, Alcohol, Religion… etc). As Diane Ravitch (2004: 46) points out (with regard to textbook production in the US) , ‘the world may not be depicted as it is and as it was, but only as the guideline writers would like it to be’.

The ‘erasure’ of particular, potentially problematic representations – such as those of minority ethnic groups in Russian language textbooks (as reported in Azimova & Johnston 2012) or of Canadians in US-published French language textbooks (Chappelle 2009) – is seen as a deliberate, ideologically-motivated attempt to ‘rewrite’ history and demography. Hand in hand with this erasure of difficulty is found what Gray (2010: 727) describes as ‘the new salience of celebrity in textbooks’, indexing a neoliberal agenda that associates the use of English with success, individualism, glamour, and wealth.page2

However we view it, the way that the learner is represented (or not represented) in the materials they use, has strong ideological ramifications. As Asimova and Johnston (2012: 338-9) put it:

Representation is a highly political business. By this statement we mean that, consciously or unconsciously, those who create and distribute representations play a central part in power relations, challenging or, more usually, reinforcing existing hegemonic relations. Another way of looking at this issue is that representations are never neutral. Though they often seem “normal” and, in the case of visual images especially, can be hard to challenge (Postman, 1993), representations are highly ideological and are a crucial component in forming, maintaining, and changing our view of the world, of groups of individuals, and of the relationships between them. Relations of class, gender, race and sexual orientation are among the most important relations that are centrally mediated by representation.

This, then, was the gist of the case I made: arguing that the way that learners are represented in published materials is both ideologically motivated and out of synch with reality .

In defense of these representations, what arguments were offered by my opponent and in the open discussion during the debate?  Here are four:

1. Students don’t want to be reminded of their less than perfect lives: the view through rose-tinted spectacles offers some respite from the general grimness in which they live;

2. The aspirational culture instantiated in coursebook images and texts has a strong motivational charge, and represents the sort of ‘ideal self’ that some scholars (e.g. Dörnyei 2009) argue is the prerequisite for success in second language learning;

3. All learning involves first identifying (proto-)typical examples of a behaviour, and only later accommodating more marginal phenomena. Hence the need to start with exemplars of the ‘norm’: e.g. white, middle-class, heterosexual family structures, before engaging with the ‘exceptions’;

4. It is not for the textbooks to reflect the reality of the learners’ lives (an impossible task anyway), but for the teacher to mediate – and exploit – the ‘reality gap’, by, for example, having the learners interrogate the texts and even subvert them.

Which way would you have voted?


Azimova, N., & Johnston, B. (2012) ‘Invisibility and ownership of language: problems of representation in Russian language textbooks,’ Modern Language Journal, 96/3.

Chappelle, C. (2009) ‘A hidden curriculum in language textbooks: Are beginning learners of French at U.S. universities taught about Canada?’ Modern Language Journal, 93/2.

Dörnyei, Z. (2009) ‘The L2 motivational self system’, in Dörnyei, Z., & Ushioda, E. (eds) Motivation, Language Identity and the L2 Self, Bristol: Multilingual Matters.

Gray, J. (2010) ‘The branding of English and the culture of the New Capitalism: Representations of the world of work in English language textbooks,’ Applied Linguistics, 31/5.

Ravitch, D.  (2004) The Language Police: How pressure groups restrict what students learn, New York: Vintage Books.

Sleeter, C. E., & Grant, G. A. (2011) ‘Race, class, gender, and disability in current textbooks,’ in Provenzo, E., Shaver, A. & Bello, M. (eds.) The textbook as discourse: sociocultural dimensions of American schoolbooks, New York: Routledge.

Tomlinson, B., & Masuhara, H. (2013) ‘Survey Review: Adult Coursebooks’, ELT Journal, 67/2.

Thanks to Piet Luthi for the mock-up.

You can watch a  recording of the debate here:


S is for Sexist language

18 03 2012

You know the one about the judge who, on being shown the name of the culprit to be tried, said, ‘I can’t try him – he’s my son!’  Another judge was summoned, who – no sooner given the name of the culprit – said, ‘I can’t try him – he’s my son!’

How can this be?  Well, of course, one judge was the culprit’s father, the other his mother. (I’ve adapted this from an original which was about two surgeons, but you could create a similar puzzle involving two nurses, or two plumbers, or two manicurists, and so on).  The puzzle plays on the fact that, while in English most professions are not marked for gender, we are disposed to assume (foolishly, of course) that the representatives of certain professions are either male or female, by default.

The puzzle wouldn’t work in Spanish, where – apart from anything else – the determiner clearly flags the gender of el juez and la juez. (Not to mention the feminine ending, in the case of the relatively recent form: la jueza).

On the other hand, because the masculine form is the unmarked (i.e. the default) form, when you have a group of judges or architects or surgeons, they are automatically ‘masculinized’, whether or not the group includes any women members. So the puzzle becomes: Dos jueces caminaban por la calle, y los dos dijieron… (Two judges were walking down the street, and both said ‘That’s my son!’  etc).

That the masculine is the default form in Spanish accounts for all sorts of oddities, such as the fact that a parents’ association in Spain is una asociación de padres, even though the only people who attend are las madres. Or that, when you walk up to a crowded stall in the market, you ask ¿Quién es el último? (Who’s the [masculine] last?), even if the bulk of those in line are women.

The invisibility of women that is instantiated in such usages is, of course, way out of step both with reality and with feminist aspirations.  Consequently, this linguistic bias has prompted the publication, in Spanish as in English, of a number of ‘style guides’ that promote the use of non-sexist and inclusive language. The Spanish style guides, however, have irritated the self-appointed guardians of ‘la lengua’, i.e. the Real Academia Española (RAE), to the point that, a couple of weeks ago, a leading Spanish grammarian and lexicographer, Ignacio Bosque, was moved to write a long piece in the national press, signed by 24 of his fellow ‘academicians’, taking these style guides to task. Unsurprisingly, this put el gato (or la gata?) among las palomas (or los palomos?) and, as they say here, the debate was served.

The principle arguments levelled by the (mainly) male academicians at the (mainly) female style guide writers are that their recommendations are

  • inelegant
  • wrong
  • impractical
  • ineffectual
  • misconceived
  • ideological

Inelegance, it is true, often results from the doubling up of gender-marked nouns in the interests of inclusiveness (los hermanos y las hermanas de mi padre y mi madre son mis tíos y mis tías), and an extract from the Constitution of Venezuela that takes this trope to an extreme is quoted with derision.

But worse than inelegant, many of the recommendations are (according to Bosque) simply wrong, and ‘contravene the norms laid down by the RAE’.  In 2008 the then Minister for Equality was roundly mocked for addressing her parliamentary colleagues as ‘miembros y miembras’ (members). The RAE declared this attempt at greater visibility ‘incorrect’.  That coinages like miembra or jueza are wrong simply because they are coinages overlooks the fact that new words are being coined at an astonishing rate (8,500 English words a year, according to a recent study). Language changes, resist it as we may.

Of course, it is not always that easy to unpick the deeply embedded cognitive structures of a language, especially those encoded in its grammar. Attempts to invent gender-neutral pronouns in English have failed miserably. But change does occur, relentlessly, even at the level of grammar. Apart from a few residual suffixes, English lost its gender distinctions long ago, as did Persian, while some languages (e.g. French and, hey!, Spanish) collapsed a three-way distinction into two. Will there still be grammatical gender in Spanish in 500 years? Place your bets.

More serious is the charge that militating for non-sexist language is both ineffectual and misconceived. It’s ineffectual, according to the RAE, because you can’t legislate for language change, which is ironic, since this is exactly what the RAE (through its grammars and dictionaries) has always done. As Deborah Cameron (1990: 162) has argued, ‘defenders of “the language” regard language as their property’, and the tone of the Bosque article is very much ‘back off!’

And the movement for greater inclusiveness is misconceived, so the argument goes, because linguistic change does not effect social change: it merely reflects it. Again, Cameron (1990: 90) is eloquent on this subject:

Anti-feminists are fond of observing that eliminating generic masculine pronouns does not secure equal pay.  Indeed it does not – whoever said it would?  Eliminating generic masculine pronouns precisely eliminates generic masculine pronouns. And in so doing it changes the repertoire of social meanings and choices available to social actors.

Moreover, as she writes elsewhere (Cameron 1995: 143), ‘changing what counts as acceptable public behaviour is one of the ways you go about changing prevailing attitudes – ask anyone who still smokes cigarettes’.

Finally, a great deal of the debate about the RAE communique hinges on the claim that language is ideologically neutral. ‘La gramática no tiene ideología’[i] ran a headline in El País (March 13, 2012). But this is debatable. Languages reflect the cultures out of which they have emerged and bear the traces – the scars, even – of their origins. A language which divides all of creation into two genders, and then selects one of those genders as the preferred form, is ideological to the core.


Cameron, D. (1990) ‘Demythologizing sociolinguistics: Why language does not reflect society’, in Joseph, J.E., and Taylor, T.J. (eds.) Ideologies of Language, London: Routledge.

Cameron, D. (1995) Verbal Hygiene, London: Routledge.

[i] ‘Grammar has no ideology’

Q is for Quote marks

21 07 2010

Enjoying the relative cool of Central Park last week, I discovered that the bench that I was sitting on bore a message:

Wishing no disrespect to the deceased, I was intrigued, nevertheless, by the quote marks around abruptly. How was I meant to interpret them? What extra nuance of meaning did the author of this tribute intend? What, in short, is the pragmatic force of “abruptly” as opposed to abruptly?

Quote marks, after all, are not innocent bystanders in the processes of text creation and interpretation. Traditionally, of course, they separated quoted matter from the writer’s own words. Hence, they’re called (variously) quote marks, quotation marks, speech marks and so on. But they’ve come to fulful a number of other functions too. Look at these examples of headlines, all taken from the BBC News website on a single day, along with the first line of the accompanying report:

Pakistan’s ISI ‘supports’ Taliban

Pakistan’s intelligence service has direct links with the Taliban in Afghanistan, a report claims, but Pakistan denies it.

Slovak opposition ‘wins’ election

A centre-right coalition wins a majority in Slovakia, but the prime minister says he will still try to form a new government.

‘Israeli spy’ arrested in Poland

Polish authorities have arrested a suspected Israeli agent in connection with the murder of a Hamas operative in Dubai in January.

Japan PM warns of debt ‘collapse’

Japan is at “risk of collapse” under its huge debt mountain, the country’s new prime minister has said.

‘Threat’ to porn site visitors

Visitors to porn sites are at serious risk of being exploited by cyber criminals, a study has suggested.

Adobe fixes ‘critical’ Flash flaw

Adobe has fixed a “critical” security flaw that had the potential to allow hackers to take control of affected computer systems.

‘Bullying’ link to child suicides

As many as 44% of suicides among 10-14 year olds between 2000-2008 may be bullying-related, a charity suggests.

Only one of these seems to flag  direct speech quotes – the one about the Japanese debt collapse. It seems that the others serve to distance the writer (and website sponsor) from some kind of assertion: they act as a form of hedging or mitigation. Note how often they collocate with verbs like claim, deny, suggest, suspect, as well as with modals like may. The doubt implied by quote marks is why they are popularly known as  ‘scare quotes’. In some cases they even suggest that the claim is so disputed as to be false, as in the example: Slovak opposition ‘wins’ election. Note also how many of the hedged items have negative connoations: spy, collapse, threat, bullying, etc. It’s as if the BBC doesn’t want to be too closely associated with reporting bad news.

But just as interesting as the cases where quote marks are used, are the cases where they are not – and I wonder if there isn’t a covert political agenda in operation here. For example (from the same webpage on the same day):

Burma denies nuclear programme

The Burmese government has denied recent reports that it is developing a nuclear weapons programme.

Cuba frees paraplegic dissident

The Cuban government has freed a jailed dissident and moved six others to jails closer to their homes.

Clash reports on Iran anniversary

Sporadic demonstrations in Tehran and other Iranian cities have been reported on the anniversary of the disputed presidential election.

What would the effect have been, had nuclear programme, dissident, and clash been in quotes?  Given their absence, can we  infer that the BBC feels less need to hedge when reporting on some countries than on others? Is not using quote marks a veiled form of criticism? To me, all this suggests that the way quote marks are used (or not used) might be a useful focus for raising awareness about the ideological sub-text of texts – does anyone know of material that does this?

But none of this solves the mystery of the ‘true New York lady’ – neither her life nor her manner of leaving it!