Chapter 1  Unlocking texts
Discovery activity 1.1   Unlocking texts
Put yourself in the position of a beginner, a complete beginner. Here is a text - it comes from the tomb of an Egyptian noble who lived several thousand years ago.4 I'm assuming you can neither read hieroglyphs nor speak Ancient Egyptian. But study it closely. Can you make any sense of it? Can you at least spot some patterns or regularities?
ii ii iu n ii n ii n
Commentary ■ ■ ■
The urge to make sense of text - even of the text of a language that we neither speak nor read - is such that we are quite capable of inventing meanings on the basis of the flimsiest of evidence. It's possible, for example, that you took a guess that the text is about birds, since figures of birds occur frequently. This, in fact, was the strategy adopted by those scholars who first attempted to decipher hieroglyphs: 'They all looked for a symbolic meaning for each hieroglyphic sign. They expected a picture of three wavy lines to mean water and only water; a picture of a head to mean a head, that of an owl to mean an owl and so on. They made no allowance for the fact that such pictures may, in fact, be phonograms (sound signs), or, indeed, letters of an alphabet rather than pictographs.'5 It took some time before this very literal and 'bottom-up' approach to decipherment was abandoned. Nevertheless, some of the symbols in the text are pictographs, as we shall see, so the strategy is not entirely unproductive.
On the other hand, knowing that the text was found in a tomb, you may have guessed that it had a religious significance, that it was a sacred text, or a biography of the deceased, for example. This 'top-down' strategy, using contextual clues, would have put you on the right track. You may also have noticed some repetition of elements - the four owls and their associated plant-like symbols, for example. Or the four sets of three vertical bars. Putting two-and-two together and using your background knowledge of text types and funerary culture, you might have guessed that these reiterations indicate a ritualistic discourse style, as befits a funerary text, such as a prayer or incantation.
8
Chapter 1   Unlocking texts
If you turn to page 185, you'll find a transcription, transliteration and translation of the text and you'll see that the interpretation we have reached is not far from the mark. Some of the signs do in fact mean what they resemble, such as bread, oxen and geese. The repeated plant-like object and owl combination seem to form the meaning a thousand of, which does form a kind of refrain and the whole text does have a ritual function: the offering of funerary gifts.
So, our 'reading' of the text, while in no way profound, is amazingly accurate, given our zero knowledge of the 'code'. What we have done is 'read between the lines' - or between the glyphs - exploiting different clues in order to access different types of knowledge. There are at least three different types of clue we used: the signs themselves, the patterns of signs, and the context. These clues in turn triggered inferences at the level of word meaning, text type, and the overall purpose of the text.
At the same time, by studying the translation, we are already in the position of making some tentative deductions about Ancient Egyptian - its writing system (eg that the three vertical bars are possibly some kind of plural marker) and possibly its grammar and even the culture which the language gave voice to. This one text is starting to reveal the secrets of a whole language and society. ■
Learners of English are faced with similar - if not quite such daunting -challenges when confronted with English-language texts. They too must mobilize a variety of'text attack' strategies in order to glean some kind of sense from the text. And, through texts, they have access to 'insider knowledge' - about the language and the culture, of which the text is a realization.
Discovery activity 1.2   What makes texts difficult?
Take, for example, the following text6. What would you expect learners to find difficult about it?
1.2
Police dog sacked after biting innocent man
A police dog in Basel, Switzerland, has got the sack for biting an innocent bystander at the scene of a burglary.
Shep, a six-year-old German Shepherd, was taken off duty after the incident. The man had to be taken to hospital in an ambulance for treatment to a leg injury.
Shep's handler, who had been called to a burglary at a city boutique, was told the
suspect was still in the building. But as officers carried out a search, the dog wandered outside to where a group of people were watching events. He then allegedly bit a 20-year-old man.
Shep was one of eight dogs serving with Basel police. A spokesman said this was the first such incident to have occurred in the city.
9
Chapter 1   Unlocking texts
Commentary ■ ■ ■
Here, in fact, is what some learners - all of roughly the same level of proficiency -had to say, on being asked to rate the text for difficulty:
1 'Difficult. Because I don't understand some vocabulary and for this I need a dictionary.'
2 'The vocabulary is difficult but the text is easy to understand.'
3 'A little easy. Because it's a little history.'
4 'Difficult, not because vocabulary but for the content.'
5 'Not difficult. It's easy to understand some word by its context.'
6 'Difficult because it uses more complex constructions and more unknown vocabulary.'
Note how a variety of factors seem to interact in terms of the learners' perceptions of the difficulty of the text. There are the more obvious ground-level language factors, such as vocabulary and grammar ('complex constructions'), but also the higher-level features, such as the context (which helps the deduction of unfamiliar vocabulary) and the text type itself: 'a little history', ie a story or anecdote. For this learner, the fact that it is a recognizable text type seems to off-set the ground-level difficulties of vocabulary and grammar, perhaps. And then there is the 'content' difficulty, which, to the fourth respondent, is something quite apart from the vocabulary. What does he mean by this? Perhaps the text failed to activate a coherent mental picture (or schema) of the events. This may in part be due to the fact that the events are recounted not in their chronological order, but in an order favoured by news reporting, where the outcome of the situation is summarized before the events themselves are detailed.
All this suggests that learners approach texts from different directions and with different expectations. This, of course, has implications for the way teachers deal with texts in the classroom. At the very least, we need to bear in mind that a text on the page may 'generate' very different texts in the minds of the learners. ■
So far we have been discussing the way learners respond to texts. But, of course, they also have problems creating texts. Here, for example, is the reconstruction from memory by one learner of the text we have just read:
1.3
A police dog has been sacked for biiing an innocent 20-years-old-man at ike scene of a burglary. The dog is from Switzerland and it has bitten (he man in the leg. The fact is thai the policeman that hand the dog - there are 8 police dogs in Basel ■ had been told that a man, accused of burglary was still in the shop. When the police were in the building where the facts had taken place, the dog wandered in the street and bit the man, who had been taken to the hospital in an ambulance. Police said it is Ike first time thai occurred.
The learner has succeeded in including all the main facts and events into the account and has done so using language that is generally well-formed (apart from the verb hand instead of handle). To a reader unfamiliar with the story, though, it may not be that easy to reconstruct the events and participants from this account. Some of the information, such as the fact that the dog is from Switzerland, is incorporated rather randomly into the text and might be distracting. Nor is it
10
Chapter 1   Unlocking texts
entirely clear which man - the burglar or someone else? - was bitten by the dog (the dog wandered in the street and bit the man). The use of narrative tenses helps give a generally accurate idea of the order of events, but only our knowledge of the order in which such things happen in real life helps us make sense of the use of the past perfect in the phrase the man... had been taken to the hospital. And the use of the phrase the fact is in the third sentence sends a slightly misleading signal as to how this sentence connects to what has preceded it: it suggests that what we have been told so far is in fact a fiction! Finally, the word that in the last sentence clearly refers to something further back in the text - but what exactly? And does the writer, by choosing that rather than this, really intend to put some distance between herself and the events? (Compare this with the last line of the original text, for example.)
All this suggests that the ability to write connected and intelligible text is - like the ability to interpret text - a complex interaction of a variety of skills. It is clearly not simply a matter of stringing sentences together.
Why texts?
The above discussion reinforces the view that learning a language is more than the learning of its grammar, vocabulary and pronunciation. And that the ability to handle texts does not necessarily result from the ability simply to read and produce sentences. There is nothing new, of course, about this. Over a hundred years ago, a leading writer on second language learning, Henry Sweet, had this to say: 'When the sounds of a language have once been mastered, the main foundation of its study will be connected texts.'7 Not words, nor sentences, note, but connected texts. Sweet added that 'it is only in connected texts that the language itself can be given with each word in a natural and adequate context'.
Nor was the idea that you can learn a language through detailed analysis of texts new even in Henry Sweet's time. In 1850 a certain T. Robertson published an English language textbook for French speakers (subsequendy re-edited for Spanish speakers) that was based entirely on the study of a single text, spread over 20 units. The first unit of the first course starts with the first sentence of the text (apparently a story from the Arabian Nights):
1.4
We are told that the Sultan Mahmoud, by his perpetual wars abroad and his tyranny at home, had filled the dominions of his forefathers with ruin and desolation and had unpeopled the Persian empire.
The text is first translated word by word and phrase by phrase and this forms the basis of exercises that involve translating the text back and forth. At this point, children and women and anyone else who simply wants to get to grips with the language as soon as possible, are advised to skip the next section, which 'is written specifically for those laborious persons who wish to know exactly what it is they are committing to memory'.
For those laborious persons, the one-sentence text is then sub) ect to rigorous analysis at the level of pronunciation, vocabulary (every word is translated into
11
Chapter 1   Unlocking texts
French) and grammar. For example, the word unpeopled is analysed into its component parts and each part is commented on, including the prefix un- and the suffix -ed, with other examples of similar compounds being supplied. In the grammar section, the first phrase, We are told, is analysed as an instance of the passive; his perpetual wars and the Persian empire are broken down into their components to demonstrate the formation of noun phrases in English. Further exercises of deconstruction and reconstruction follow. And the pattern is repeated for the next 19 units, that is, until the end of the story.
The problem with this approach is that, as Robertson himself acknowledged, it can become very pedantic and hence very boring. Also, unless the texts that are chosen for study bear some resemblance to the language needs of the learners, all this analysis is somewhat academic. It is hard to imagine there being much need, even in 1850, for the verb to unpeople, however good an example of English word formation it might be.
Nevertheless, the idea that texts - even very short ones - can 'deliver* a great deal of information about the language, is suggestive. Updating Robertson's method somewhat, let's have a look at a relatively modern text. Here is the 20th century translation of a Japanese haiku-like poem by the hermit-monk Ryokan (born c. 1758)8. As you read it, consider what features of language it embodies that might be of use to a learner:
1.5
Have you forgotten the way to my hut?
Every evening 1 waitfor the sound ojyourfootsteps, But you do not appear.
Short as it is, the poem is a complete text. It consists of 21 words (of which two are repeated twice, to make 23 in all) and these are organized into two sentences, the second of which consists of two clauses, making it a compound sentence. All but seven of the 26 letters of the written alphabet appear in the text. Of the 24 consonant sounds in English, 16 are represented, along with 15 of the 20 (British English) vowel sounds.
The text shows a typical distribution between grammar words (or function words) and content words. Function words include have, to, o/and not, and content words are the ones that carry the burden of meaning, such as forgotten, -way, hut and footsteps. Of the 21 words in the text, 15 are in the top 200 words in English according to frequency counts, and seven (ie a third of all the words) are in the top 20. Again, this is a fairly typical distribution.
Turning our attention to the grammar, what does the text tell us about the way English sentences are organized? First of all, all the eight different parts of speech are represented, apart from adjectives (unless you classify my and your as adjectives rather than determiners). So, we have nouns (eg hut), pronouns (egyou), verbs (egforgotten), determiners (eg the, every), prepositions (eg to), a conjunction (but) and an adverb (not). Of the nouns, one has the plural suffix -5. Of the verbs, one is in the form of the past participle (forgotten); there are two auxiliary verbs (have, do) and two finite verbs (wait and appear).
12
Ciiapter 1   Unlocking texts
These word types are in turn organized into three of the most common phrasal combinations in English. So, we have noun phrases (my hut, every evening), verb phrases (haveforgotten) and prepositional phrases {to my hut). The noun phrases show instances of pre-modification with determiners (as in my. hut, the sound) and post-modification using either a prepositional phrase {the way to my hut) or an of construction {the sound of your footsteps). These are all very common ways of grouping words in English.
These words and groups of words in turn realize the main functions typically found in combination in sentences, such as subjects (7), objects {the way to my hut), verbs {appear) and adverbials {every evening). The three clauses each demonstrate three common verb patterns, respectively: verb + object (ie a transitive verb pattern); verb + preposition (for) + object; and a verb with no object (ie an intransitive verb). As well as the basic statement form of subject + verb (I wait...) there is an example each of (1) the inversion of subject and verb to make questions (Have you forgotten..?) and (2) the use of a 'dummy operator' + not to form negative statements (you do not appear). The way English verbs are marked for tense and aspect is also exemplified, with two verbs in the present tense unmarked for aspect (wait, appear) and one example of a present perfect construction {have. ..forgotten). Finally, at the level of connected discourse, the conjunction but connects the two parts of the second sentence, signalling that what follows contrasts in some way with what went before. And the repetition of you across the two sentences helps connect them.
This rather detailed analysis of one very short text is simply intended to demonstrate how much 'language' there is in a text and, therefore, how much potential texts have for the purposes of exemplifying features of language - of phonology, orthography (ie the writing system), vocabulary, grammar and discourse - for teaching purposes. Stuck on a desert island with only a book of haikus at hand, a resourceful teacher has the means to teach a great many facts about the language. (This is not meant to imply, of course, that teaching facts about the language is all that is involved in language teaching. For a start, it ignores the possibility of using texts to trigger language production on the part of the learners.)
Discovery activity 1.3   Unpacking a text
Here is another poem by Ryokan9. Can you 'unpack' its grammar? That is to say, what further features of English grammar does it display that might usefully be highlighted for learners?
1.6
I expected to see only pink blossoms
but a gentle spring snow has fallen and the cherry trees are wearing a white coat.
13
Chapter I   Unlocking texts
Commentary ■ ■ ■
Like the first Ryokan poem we looked at, this one has a representative range of high frequency features of English, including all the parts of speech (except for prepositions, if you discount the infinitive marker to). Unlike the first poem, however, this one is particularly rich in adjectives (pink, gentle, white). It also includes another form of noun pre-modification which is very common in English: the use of nouns to modify other nouns, as in spring snow and cherry trees. Also of interest is the verb + lo-infinitive pattern, extremely common in English, represented by expected to see. Here, too, is an example of an infinitive (to see), as well as a regular past tense verb (expected) and examples of each of the two aspects in English: the perfect (has fallen) and the continuous (are wearing). These, incidentally, demonstrate agreement (or concord) in English, ie the way singular subjects take singular verb forms (has) and plural take plural (are). Another feature that is well represented in this text is the article system, including both the definite and indefinite articles (the, a) and the 'zero article', ie the absence of any determiner in front of a noun phrase, as in pink blossoms. ■
Incidentally, between the two texts (1.5 and 1.6), only three letters of the alphabet (j, q, z), plus three each of the consonant and vowel sounds in English, remain unrepresented. Eleven of the 25 most frequent words in English occur, some of them more than once. Grammar coverage across the two texts includes:
• all the parts of speech
• the basic article system
• common ways of forming noun phrases and preposition phrases
• first and second p erson subj ect pronouns and possessive adj ectives
• transitive and intransitive verb constructions
• the infinitive
• affirmative and negative statements and question forms
• present and past simple tenses
• continuous and perfect aspect
• sentence-initial and sentence-medial adverbials, and
• additive and contrastive connectors.
In short, a good deal of the traditional elementary syllabus is locked up in these two poems - 'a world in a grain of sand', as William Blake put it.
I should stress, at this point, that I didn't have to look hard to find texts that have such rich seams of grammar running through them. All texts have grammar and -especially if they are authentic texts, but not too specialized - the grammar that is embedded in them is bound to be fairly representative of English grammar as a whole. In this sense, language shares a feature of other complex systems: its smallest self-standing components (ie texts) are miniature representations of the system as a whole (ie lexico-grammar).
I said that a good deal of grammar is locked up' in the two poems and I chose the (phrasal) verb 'locked up' deliberately. The fact that the texts contain examples of X, Y and Z features of English grammar is not of much use to language learners a) if they don't notice these features, and b) if they don't know how representative, typical, frequent, generative, etc, these features are. That is where the teacher comes in. It is by means of the teacher's expertise that these features are 'unlocked'. The process is not dissimilar to the way the secret of the hieroglyphs was unlocked by Champollion using the Rosetta Stone.
14
Chapter 1   Unlocking texts
Classroom applications
At strategic points in each chapter we will be looking in more detail at ways of exploiting texts for language work. But here are a few basic text-unlocking techniques to get you going.
1 Dictate a short text - such as one of the haikus by Ryokan - and allow learners to compare and correct their texts, before asking them to:
• count how many sentences there are
• count how many words there are and how many words are repeated
• identify the word classes (noun, adjective, etc)
• say how many countable nouns there are
• say how many uncountable nouns there are
• say how many adjectives, determiners, adverbs, etc, there are
• underline all the verbs
• identify the tense, aspect and voice of each verb phrase
• find any collocations, ie words that you think might co-occur frequently (learners can check their intuitions against a good learners' dictionary)
• find any figurative or idiomatic use of language, including phrasal verbs
• identify any cohesive devices
• find any pronouns and identify their referents (ie the words they refer to).
2 After this detailed analysis of the text, ask the learners, working together, to try and re-construct it from memory. It may help to provide some word prompts on the board, eg:
,,. pink blossoms ... gentle spring snow cherry trees ... white coat.
3 Prepare a 'gap-fill' version of the text for a subsequent lesson. Here is how a colleague of mine used this gap-fill idea with a group of ESL students in Oregon, who are studying the English of Natural Resource Technology (NRT), in order to work in that field once they return to Central America:
There was a lull in the conversation. I waited, then piped up with how interesting one of the required books for NRT was. The book's all about how to find one's way in the wilderness, read topographical maps and things like that. I said I'd just learned how to determine direction using a non-digital watch. Interest! Well, I said, let me read the short paragraph to you...
The paragraph became a point of discussion among pairs. [...] Finally, students read the paragraph to themselves, discussed it, then listened to me read it once more as they read along.
We took a break, during which I wrote up the paragraph on the board. Next, I created gaps in place of the 'grammar' or function words, which I transferred to the other half of the board in random order. If a word occurred more than once, I marked it accordingly, e.g. 'the (x7)'.
After the break, I asked the students to fill in the blanks then compare with a partner and finally with the original text. [...] Fifteen minutes left... I ask for the difference between function/grammar words and content words by pointing out that we've been finding function words. I talk quickly about how learning grammar might just be learning how to fill in the spaces between the content words or mix them successfully with the grammar words.
15
Chapter 1   Unlocking texts
How could you do this on your own? Right, find a text, copy it, get rid of the function words and put them back like we did. Problems? How to best get rid of the words. We need to leave the text alone for a while so we're not just working from memory as we did today to some extent (still provided a meaningful challenge to students, I think).
Rob Haines
Rob's lesson is a good example of how even short texts can be used productively to unlock language features. After all, every wordin a text yields a potential language lesson. In any text, therefore, there resides a whole syllabus waiting to be uncovered.
16