Literature and language teaching

Anthony Jennings

It is usual, when giving a talk in a language not one's own, to begin with an apology for the limits of one's linguistic ability. I do not intend to do this, not because I have any confidence in my ability to perform adequately in Italian1 - far from it - but because I consider the responsibility not to be mine but yours. I find myself at International House for the first time in twenty years, in the place where I followed my first and only course of Italian as a foreign language. Whatever mistakes I make, of grammar, pronunciation, appropriateness - or of any other kind you please - I intend to attribute them exclusively to that month of lessons, rather than to anything that happened before or since. In other words, it is all your fault!

I will apologise in advance, on the other hand, for the content of my talk. It is not at all clear to me, and I am sure that it cannot be very clear to any of you, why International House should have invited as a guest speaker on the teaching of Italian someone who not only is not Italian, but who has no experience at all (except on the receiving end, as I have mentioned) of teaching Italian. However, what's done is done; the invitation has been issued, and you will have to be patient with me, if only because it is much too early for lunch.

I will start with a simple quiz:

1. What is the life span of the elephant?

a) 150 years

b) 90 years

c) 25 years

d) 10 years

2. How many seconds are there in a century?

a) 3,1 billion

b) 31 billion

c) 310 billion

d) 3,100 billion

3. Imagine an empty room. Two strings are hanging from different points on the ceiling, but they are too short for you to reach them without standing on the chair. Even from the chair you cannot reach both of them at the same time. Your task is to tie them together. How do you go about this?

4. This question involves a physical exercise. Stretch out your arms with the palms of your hands facing outwards. Cross over your hands and link your fingers. Now turn your hands inwards and lift the little finger of your left hand.

5. And finally: which of the four previous questions requires abilities or capacities closest to those involved in the study of literature in a foreign language?

The first question simply requires knowledge. If you don't know the life span of an elephant, you can look it up in the encyclopaedia (it is 90 years). The second question is fundamentally a test of competence, or the application of knowledge. (I am giving my own definitions of these rather slippery terms) We all know the number of seconds in a minute, minutes in an hour, hours in a day, and so on, but it takes a while to work out the answer, even approximately. A competent mathematician will apply the required knowledge much more rapidly than the rest of us. (The answer is just over 3 billion). The third question requires problem-solving ability, and perhaps a bit of lateral thinking. One string must be swung towards the other, and to make it go on swinging for the time it takes to get off the chair, move it nearer the other string, and get up again, something heavy - maybe a shoe - must be tied to the string. The fourth question simply requires physical dexterity: a skill. So which of these is closest to the requirements of the student of literature in a foreign language?

Obviously each of the four is necessary to some degree. Physical dexterity (skill) is required for pronunciation, knowledge of grammar and vocabulary is indispensable, the rapid application of this knowledge (competence) is essential to oral communication. But most teachers say the third factor, the ability to imagine solutions to problems, is most specifically useful to the student of literature.

And yet this instinct is not reflected in most theoretical writings on the teaching of literature. In these writings, two assumptions are usually made, or even stated explicitly, which are both false, in my view, and which in combination produce a dangerous distortion of the aims and methods of literature teaching. The first is the idea that linguistic competence can be defined exclusively as the competence of the mother-tongue speaker. The second is that literary competence is the sum of linguistic competence and of specialised knowledge. Taken together, they exclude cognitive ability from the study of both language and literature.

The first idea - that linguistic competence can be defined exclusively as the competence of the mother-tongue speaker - seems intuitively right, at least as long as we really are talking about speakers. Most language students would be more than satisfied to speak like a native. What more could they ask? Probably, if they are anything like me, they never will speak like a native, but they may well do lots of things better than some natives: things like reading, for example. What does reading competence involve? Can reading competence in L2 never go beyond that of an L1 speaker?

Ask yourself who has the better understanding of a lecture in Italian on quantum physics: a Nobel-prize winning physicist with a knowledge of five different languages (but with limited Italian), or an Italian professor of literature with no interest in physics? The former, of course. An illustration of this principle was provided in a workshop yesterday. I was proud to be part of the winning team (against fierce mother-tongue Italian opposition) in a reading competence test in Italian. The merit for this goes in large part to my partner, but the result also demonstrates, I think, that the upper limit for reading skills in a foreign language is not fixed at the level of the average native.

So reading competence in L2 is not synonymous with language competence, as is often suggested. The graph of the threshold for reading competence is normally drawn as a straight line, as follows:

fig. 1: conventional model of reading competence

A certain threshold of linguistic competence is said to be required, and nothing else. But if we reflect for a moment, we can see that the graph should really be a curve, perhaps like this:

fig 2: alternative model of reading competence

An improvement on either axis, horizontal or vertical, specific or linguistic competence, will make comprehension easier, and there is no reason why specific competence should not compensate for lack of language to the extent of making the L2 performer superior. Imagine two students whose co-ordinates correspond respectively to the two dots over the "i"s of "comprehension" and "no comprehension". The purely linguistic level of the student represented by the dot on the right is much lower, but he or she is the only one who understands the text!

But even this example does not fully illustrate the situation of the literature student, because it suggests that specialised knowledge, not of quantum physics but in this case of literature, is what makes the difference. And here we come to the second limb of the dominant theory which I wish to criticise: the idea that literary competence is the sum of linguistic competence and of specialised knowledge. On this point, I will probably be preaching to the converted, for all the workshop activities I took part in yesterday show that we think in the same way. You are as convinced as I am that a knowledge of literary history and of critical metalanguage is neither essential nor fundamental in dealing with a literary or paraliterary product, like a film or a song. But we have the weight of academic opinion against us. Balboni, for example, one of the most authoritative names in the field of language and literature teaching in Italy, and a previous guest speaker at IH, has this to say:

"La letteratura può essere inserita nel programma [di L2] a tre condizioni:
che avvenga quando la padronanza della lingua è tale da consentire un reale accesso al testo letterario;
che il testo non sia presentato come modello linguistico (dato che la sua natura, come si vedrà, ne fa semmai un modello di deviazione dalla norma);
che venga presentato come uno dei tanti tipi testuali, in un programma che includa anche testi scientifici, tecnici, istruttivi, e così via [...]."

This because

"L'uso di testi letterari per insegnare italiano a stranieri rappresent[a] una perversione della natura del testo letterario, la cui finalità non è certo quello di fornire materiale glottodidattico."

And he concludes

"Insegnare letteratura significa anzitutto insegnare a leggere testi letterari, cioè a cogliere ciò che li rende letterari."2

Balboni could hardly be more wrong. And because he is wrong in a way that I think is not just confusing but seriously damaging both for students and teachers, it is worth spelling out the reasons.

We can skim quickly over Balboni's first point, that literature should only be studied "quando la padronanza della lingua è tale da consentire un reale accesso al testo letterario". As figure 2 shows, the question of when "real access" occurs is a tricky one. Does the two-year old child listening fascinated to a story have "real access" to this literary text? Can the bored university student mindlessly revising for an exam have real access? Or only the critic? Did Shakespeare's illiterate audiences have "real access"? Or is "real access" a term that we would do well to jettison, and substitute with "interest"? Clearly we should not ask students to read material that bores them, literary or not.

Balboni also claims that the use of literary texts for language teaching involves "a perversion" of the nature of the text, because it was not written for teaching purposes. This is an obvious absurdity. By the same logic, all authentic material should be banned from the classroom - and not just reading material, but songs, films, radio - anything that was not manufactured specifically for didactic purposes. Is talking to students (genuine conversation, not artificial "teacher talk") a "perversion", because the primary purpose of conversation "non è certo quello di fornire materiale glottodidattico"? By this logic, language learners should be rigorously segregated from real life, from anything authentic, anything that does not have the primary purpose of supplying didactic material!3 And the most successful students, according to Balboni's bizarre logic, will then be those who are least contaminated by the "perverse" introduction of authentic materials. If you want to learn Italian, stay away from Italy!

Actually, Balboni's philosophy is even more insidiously false than this caricature suggests. What he really means is not that authentic material should be kept out of the classroom, but that what happens in the classroom should be concerned only with the exemplification of the forms of language, and not with communication or meaning. Literature in the classroom is a perversion, not because it is authentic, but because it will be used there as a model of linguistic form, though it was written as a message with a meaning. But whoever suggested that meaning - in literature or any other text - should be kept out of the classroom?

It is worth making a distinction here between two types of authentic material. There are texts that lose their meaning when they are removed from their original context. An advertisement in a classroom loses its primary function: it no longer advertises. An information leaflet about trains from Glasgow no longer informs, unless the students happen to be in Glasgow. But a story still entertains, and a joke still makes us laugh. Their classroom use only becomes a "perversion" if they are not allowed to perform this primary function - if they are treated exclusively as examples of form and usage. Does Balboni think this should happen? This would obviously be paradoxical, and he himself would be the first to deny that he believes anything of the kind, yet this unspoken and unconscious assumption underlies his objection to literature.

But there is another and still more insidious belief that lies at the heart of his argument, at the basis of at least three of the five points quoted above, if not all of them.

Literature is a specialised register, Balboni claims, a "tipo testuale" like scientific or technical or didactic language. It is essentially different, a "modello di deviazione dalla norma" This is its fundamental characteristic, he believes, in the sense that what makes it literary is what makes it different, so that to teach literature means teaching what makes a text literary. This theory that literary language is deviant, and is characterised by deviance, has been around for a long time, propounded most forcefully perhaps by Widdowson, who is probably the source of Balboni's ideas. It needs to be looked at in more detail.

Most of Widdowson's examples of deviant grammar and lexis come, unsurprisingly, from poets like Dylan Thomas and e.e. cummings (deviant even in writing his own name). It would be harder to document the theory from the equally literary works of Defoe and Robert Frost. Few people seriously believe that literature must be deviant in this sense, however, and they rely instead on Widdowson's reserve line of defence, that literature is deviant not as text but "as discourse". What he means is that in literature the normal rules of discourse are suspended, so that, for example, senders and addressers are separated, and animate and inanimate categories confused. Thus Defoe sends a message in his first novel to his readers, but the addresser (the fictional narrator) of this novel is Robinson Crusoe. Keats speaks to urns, urns speak to Keats. Dead soldiers talk in poems to Wilfred Owen. And so on.

Objects do speak in literature, but not only in literature. If you go to England, you will see plenty of cars that say

When I grow up I want to be a Rolls Royce.

In Italy some of them tell you


Or, more drastically,

Demolitemi che faccio pena.

If they have a sense of humour, they might say

Non sono sporca, mi sto solo facendo i fanghi.

Cars (and other objects) speak, in literature and in the real world, and in the same way addressers and senders are often confused. If you have children, you will be used to junk mail advertising anything related to babies. Several years ago, I was offered a subscription to a magazine called "La mia sessualità da 0-9 anni." Who is the sender here, and who the addresser? The sender of the message is obviously the publisher of the magazine, but equally obviously he or she is not the addresser. We are not going to be told about a publisher's youthful sexuality. "La mia" actually means "la sua", or "your child's". First person replaces third; the apparent addresser, like Robinson Crusoe, is the object of the narrative, not its author.

Literature is not always deviant; "normal" language sometimes is. Literature and deviance are like cars and leopardskin seat covers: some have them, some don't. Some things that are not cars have leopardskin seat covers (hotel lobbies, perhaps). But then again, some don't. In fact the majority of cars, and also of non-cars (the term is no less meaningless than "non-literary language") don't have leopardskin seat covers. In the same way, it may well be true that the majority of literary texts are not deviant. The categories of literature and deviant language are related only by chance: their correlation is 0. Or maybe it is negative, in the sense that the more deviant the message, the less likely it is to be literature, rather than advertising, or shorthand, or baby talk, or a secret code, or lovers' sweet nothings, or an abbreviation, or the language of open-heart surgery, or the jargon of bilingual rat fanciers, or Eskimos hooked on Californian TV, or Internet surfers belonging to a chat group on applied linguistics, or defrocked Jesuit priests on the run from Interpol and hiding out in Hongkong, any one of a million other deviant uses of language.

Because all this is intuitively fairly obvious, defenders of the deviant theory of literary language tend to fall back on the weak form of the same argument, the idea that literature, if not deviant, is at least specialised, a micro-language or a separate register, like the language of the law or of science. If literature is a specialised register, it must have its own language. What can this language be like? Clearly it is not lexically different: what words are specifically "literary"? Not words like "forsooth", or "polyphiloprogenitive", which are merely obsolete or rare. If literature can be written (and it obviously can) without any "literary" words, in what sense can it be said to have a language of its own?4

The usual answer makes some reference to imagery, as if imagery did not exist outside literature. But this shows a fundamental misunderstanding of the nature of all language, which is metaphorical from start to finish, root and branch (to use a metaphor). Where do words come from, if not from metaphor? ("Grass long face": face grass, is the Vanuatu pidgin term for a beard.) A cartoon shows a group of men running after a single alien-like "thing", ignoring another similar "thing", while one woman says to another: "Men! They're only after one thing!" This is funny because it calls attention to the metaphorical nature of everyday language, which normally goes unnoticed. The densely metaphorical nature of "ordinary language" can be seen in this extract from a conversation in the jargon of young Romans (the kind of people we try to teach about the nature of metaphor and its supposedly special status in literature):

M: O' cì, come và, tuttapposto?
A: Bene e te?
M: Ma 'nsomma, stò 'n po' com' e banane5.
A: Che me dici? 'n comitiva? Che fine 'a fatto er Caciara? Pacciani6?
M: Bò no' 'o so', l'urtima vorta che l'ò beccato stava 'mpicciato fracico.
A: Aò bella Girolamo!
D: Bella, regà, 'ndò se buttamo stasera?
M: Bò?
D: Der tipo me sembra ch'era libera 'a casa der Pizza.

Literature is not a single and separate register. It is different from the jargon of a specialised area of human activity, like law or medicine, because it does not deal only with a specialised area of activity. There are books, like Tristram Shandy, which are full of legal and medical jargon, but they have no special jargon of their own. On the contrary, they imitate the language of a wide variety of fields. What is often thought of as literary jargon is in fact the metalanguage of literary criticism, the jargon for talking about literature, which is useful for would-be critics, but is not needed to read literature.

These, then, are the arguments against the two propositions that I began with: the widespread ideas that linguistic competence can be defined exclusively as the competence of the mother-tongue speaker, and that literary competence is the sum of linguistic competence and of specialised knowledge. Taken together, I said, they exclude cognitive ability from the study of both language and literature. What conclusions can be derived from the refutation of these propositions?

If linguistic competence involves cognitive abilities, the task of language teachers is much more stimulating than is often thought. They are not only what Italians call "esercitatori", instilling habits (like a tennis coach) by practice, but teachers, responsible for forming minds. And to do this, they would be foolish not to take advantage of one of the most stimulating and accessible tools for forming minds, which is literature. And this is true whether their task is formally described as the teaching of language or literature: the use of literature (well chosen, of course, not random or indiscriminate), as of other authentic materials, will inevitably develop language abilities.

Secondly, if literature cannot be isolated from ordinary language, then the rigid distinction many teachers make between the language and the literature lesson is redundant. Stories of all kinds - plays, novels, films, songs, cartoons and jokes - can be analysed with the same tools and used for the same purpose as the dullest purpose built non-authentic dialogue: to reflect on the way language creates meaning, and to stimulate the production of meaning by the students. This, it seems to me, is what is already happening at Dilit International House. But maybe your teacherly instincts collide with the principles you read about in Balboni and others (and Balboni's is only a more lucid expression of the dominant philosophy). Maybe, if you have done a bit of reading in the theory of literature teaching, you suspect that you are doing the right thing for the wrong reasons. My conclusion is a very simple and pleasant one, which goes as follows.

You are doing the right thing. You are also doing it for the right reasons.

Nota 1. This talk was originally given in Italian. Back to text
Nota 2. Balboni, 1984. Back to text
Nota 3. It is true that Balboni does not actually suggest banning literature from the L2 classroom. He merely refers to its presence as a "perversion". But the word suggests a certain fastidious distaste for something that is essentially a misuse, which a purist would probably wish to avoid. Back to text
Nota 4. It is not enough to say that a Shakespearean sonnet, for example, has its own characteristic language. This is the language of the Shakespearean sonnet, not of literature in general. Does Samuel Beckett use the same style? Back to text
Nota 5. casco da tutte le parti. Back to text
Nota 6. by analogy with the famous Florentine sex killer of young tourists, to indicate someone who goes out with girls two or three years younger than him. Back to text