Landmarks in Literacy

The Selected Works of Frank Smith

Frank Smith

• Publisher: Routledge
• Year: 2014
• ISBN:
  9781317592280
• Edition: 1
• Topic: Filología
• Subtopic: Alfabetización

Frank Smith is internationally acclaimed as an essential contributor to research on the nature of reading and as an originator of the modern psycholinguistic approach to reading instruction. In his publications his aim has always been to support teachers, to encourage them to make teaching decisions based on knowledge and understanding, to analyze what their students are trying to do and why what the students are doing doesn't always correspond with what they are expected to do. Now the major topics addressed in his work are available in one volume, Landmarks in Literacy, a thoughtfully crafted selection of 16 of his key writings.

In the World Library of Educationalists, international scholars themselves compile career-long collections of what they judge to be their finest works so the world can read them in a single manageable volume. Readers thus are able to follow the themes and strands of their work and see their contribution to the development of a field, as well as the development of the field itself.

Part I
From Essays Into Literacy¹

I always argue that learning is a natural activity which usually takes place effortlessly and inconspicuously. Sometimes, however, I find it appropriate to explain why learning goes wrong, when learning becomes stressful and unproductive. In the following extract, Demonstrations, Engagement, and Sensitivity, I explain why the environmental conditions for learning, while normally present, are nonetheless complex. In the second extract, Twelve Easy Ways, I show how the difficult conditions can sometimes be deliberately contrived, even by people who have the best of intentions.

Note

¹ Smith, Frank (1983) Essays Into Literacy. Heinemann Educational Books, Exeter, New Hampshire.

1
Demonstrations, Engagement, and Sensitivity¹

How does anyone learn about language? In a recent examination of writing and a reconsideration of reading and of spoken language learning, I have been struck by the enormous detail and complexity of language that most individuals learn. Current psychological and educational views of learning, including the hypothesis-testing approach on which I have tended to rely, now seem to me to be inadequate to account for all the language learning that is achieved. Nor do they account for those occasional difficulties or failures that most people experience with one aspect of language or another – perhaps with spelling, or with certain points of grammar, or with second-language learning – although the individual may have little difficulty with other aspects of language that are intrinsically no less complicated. What we all learn or do not learn with ease varies from one individual to another, and theories of learning should be able to account for this.

First, I shall illustrate aspects of the complexity of the language that most individuals master without apparent effort, although research is lacking on how much language individuals actually learn. Second, I shall briefly outline a revised approach to learning more compatible with how I now feel language is actually learned, although extensive research will be required to substantiate it. Finally I shall consider implications of the revised approach for classroom practice.

The Conventions of Language

Languages have become what they are by chance and by use. No one ever sat down and invented any natural language (“mother tongue”). To learn and use language may be natural, and there is probably a purpose for every nuance in language, but the particular forms of a language have developed and persisted unintentionally; they are conventional and not predictable to someone who does not already know them.

This arbitrary nature of language becomes apparent if we consider either a specific language or the thousands of languages that exist in the world and the wide range of their differences. There is nothing especially natural or even logical about English, for example, in comparison with Chinese or Greek or any other language. There is no particular reason why the animal that is called “chien” in French and “perro” in Spanish should be called “dog” in English. There is no special logic behind English phonology or grammar, or punctuation; other languages manage quite well with different forms and structures or even in their absence. Without knowing a particular language or the history of its derivation, it is quite impossible to predict or work out what the conventions will be.

The Nature of Conventions

Conventions are arbitrary but mutually accepted and expected ways of doing or expressing things (Halliday and Hasan, 1976). They are essentially contracts for doing things in certain ways which could be done quite differently, provided the alternative becomes the expected. For example, every culture with a road system has a convention for driving on one side of the road rather than the other. There is no particular logic or advantage about which side the traffic drives on, provided it is all on the same side. The form of the convention is a matter of chance or arbitrary choice (or of consistency with an existing convention which itself became established by chance) and therefore quite unpredictable. Thus, to learn a convention you must discover what it is and how it works, a necessity that has enormous implications for how language itself is learned.

Conventions are the basis of shared understandings and communication. If you can understand what I have written, it is because I have anticipated your expectations about the conventions of language that I have employed. While conventions are not initially predictable, when learned they are used as the basis for prediction, and prediction is the basis of communication.

Cultures (or human brains) seem to abhor free or random variation and to be ready always to capitalize upon conventions. Predictability is preferred and exploited. Take, for example, the fact that every language in the world that has an alphabet also has a conventional alphabetical order. I can find no logical reason for the conventional ABC … order of the English alphabet except the tradition itself, but there are good reasons for some order to be established and maintained. All our dictionaries and other reference books, not to mention libraries and entire bureaucracies, depend upon the existence of this arbitrary convention. Imagine the disruption if a rebellion against the tyranny of alphabetical order succeeded.

Conventions pervade language, from the sounds and spellings of individual words to the structures of discourse and the “grammars” of entire stories. The greatest and most complex set of conventions in any language lies in its registers, in the appropriate way of talking (or writing) on particular occasions depending on who is talking, to whom, when, and about what. Conventions have little to do with grammar and much to do with the way people are expected to communicate with each other. We are all sensitive to them. Children entering school quickly realize that teachers do not (and should not) talk to them the way their parents or friends do, and everyone becomes embarrassed when visiting a different culture to find that they cannot use or respond to the registers the natives employ and expect. Languages are not learned solely from grammar books.

Nonverbal aspects of language are also conventional. There are conventions for how close we may stand to someone when we talk, how often and how long we may gaze into their eyes, and how much (and where) we may touch them, again depending on all kinds of personal and situational factors. Every culture has conventions such as these, but it is not predictable what the particular conventions will be. What is polite in one culture is rude in another. Nothing is neutral; there is no room for free or idiosyncratic variation. With language, every difference makes a difference. And it all has to be learned.

Some Advantages of Conventions

I am making some very strong statements about language. I am saying that there is only one way to say anything, both verbally and nonverbally, in writing as well as in speech, and that to say anything differently is, in effect, to say a different thing. Synonyms and paraphrases are not substitutable for each other. Choice of a particular synonym always says something about the speaker (or the speaker’s perception of the listener). Petrol and gasoline may refer to the same thing, but I reveal something of myself if I use one term rather than the other. Dogs chase cats cannot replace cats are chased by dogs in any meaningful context. The first is a statement about dogs and the second about cats. In context each will be interlocked with other sentences by conventions of cohesion so that one cannot be substituted arbitrarily for the other.

It may be objected that reducing all of language to conventions is to deprive it of its productivity and creativity. On the contrary, conventions are economical, maximizing the flexibility of language and its precision. Because the same thing cannot be said in two different ways, every difference permits us to say something different.

Conventions make creativity possible. Breaking conventions is permitted – for a good reason. Ignorance is often sufficient reason. Foreigners and children (except perhaps in schools) are expected not to know conventions and therefore not to observe them. Sometimes it is an advantage to be regarded as a foreigner for the latitude allowed to unconventional behavior.

Conventions may also be broken intentionally, to achieve something that cannot be achieved by conventional means. Creative writers often do this to say something they cannot otherwise say, like James Joyce with grammar and e.e. cummings with punctuation. Without conventions there could be no unconventionality. The constraints of language in effect offer boundless opportunities.

The Extent of Conventions

I cannot begin to catalog all the conventions of a language or how much we learn about them, as the amount and complexity are staggering. Almost universally the extent of this learning is under-estimated, as I shall try to demonstrate with the example of spelling. This is not because I regard spelling as a particularly important, substantial, or difficult part of language learning. But spelling conveniently illustrates that proficiency in particular aspects of language is far more complex than it is generally assumed to be and that current theories of learning not only fail to account for how many individuals achieve unexpectedly large amounts of learning but also why others fail.

I have found no research on how many spellings are known by those individuals who would be characterized as “knowing how to spell” – I shall call them “spellers.” My estimate is that the total could be as high as 50,000, since that is a low estimate for the number of words in an average spoken language vocabulary, and most spellers (it seems to me) have a spelling for just about every word they know. Of course, some of these spellings may be wrong, but they are probably a small proportion of the total number of known spellings, and the incorrect spellings have themselves probably been learned – there is a persistence about them.

How does anyone achieve the capacity to spell scores of thousands of words, most of them correctly? We do not spell by writing down the sounds of words; the werst spellers are the wuns hoo rite foneticaly. But we also do not spell by applying spelling rules. Even if you have learned all of the 300 sound-to-spelling correspondences that can be found in a relatively small sample of 20,000 common English words your chance of being correct in applying the rules to spell any of those same 20,000 words is slightly less than 50 percent. Even if the rules do happen to spell an unknown word correctly for you, you still must inquire whether the spelling is correct, as none of the rules comes with a guarantee. There is only one way for anyone to become a speller and that is to find out and remember correct (that is, conventional) spellings.

Remembering many thousands of spellings would not appear to be the problem, not for most people at least. The brain usually takes remembering in its stride. We have all remembered the sounds of 50,000 words or more (sounds far more complex and arbitrary than their spellings) and also meanings and rules for the use of all those words, a total that must run into hundreds of thousands since the majority of words have more than one meaning. Also, of course, we are all capable of remembering astronomical numbers of other things – all the faces, places, animals, objects, events, relationships, and attributes that we find familiar in the world, by sight, sound, touch, taste, and smell.

What must be explained, in spelling and all the other conventions of language, is how an individual finds out and learns what is appropriate in the first place. How could anyone learn the conventional spellings of scores of thousands of English words – one at a time? The answer cannot be formal instruction. The number of word lists that would have to be studied and the limited success that even adept spellers have in learning and retaining items from lists makes it improbable that rote learning explains very much. And I do not think spellers can attribute their success to 50,000 trips to the dictionary or to being told a spelling when they happened to need it.

I was tempted to think that we learn to spell by writing, but that is not an adequate explanation either. Even if children learn a spelling every time a mistake is indicated (and I have found no evidence of that), I doubt whether many children write 50,000 words during their school career. Certainly they would not write 50,000 different words, and some of the words that we can all spell occur in the language with a frequency of one in a million or less. There are words with familiar meanings and spellings that we probably encounter only once every five years, words we have never written and perhaps not even spoken in our lives.

I can think of only one way in which such an enormous repertoire of spellings might be learned and that is by reading. We learn to spell by reading. But now I have problems. Reading may be necessary for learning spellings, but it is obviously not sufficient. Many people read who cannot spell. And I myself have argued that readers need not pay attention to spelling, that indeed it can interfere. How, then, does one account for the enormous amounts of learning that spellers must do when they read, without effort or awareness, and also for the fact that other readers do not learn to spell, though they may learn other equally complex and numerous conventions? Learning theories based upon association, or stimulus-response connections, or on reinforcements seem to me completely inadequate to account for either the learning or the failure to learn, while hypothesis-testing seems equally unsatisfactory. Surely we gain only a small part of our spelling competence by having hypothesized spellings confirmed or corrected. And where do those hypotheses come from?

How do writers learn to punctuate, to capitalize, to produce grammatical construction? Again, not from formal instruction on grammatical constructions, or on rules. Linguists cannot agree upon anything like a complete set of the internal “rules” that speakers of a language must respect to generate or evaluate grammatical sentences. And the other type of “rule” that is the common currency of classrooms and textbooks tends to be hopelessly circular: “Always begin a sentence with a capital letter and end it with a period (or question or exclamation mark).” “What is a sentence?” “Something that begins with a capital letter and ends with a period.” Rules like “A sentence expresses a complete thought” only appear to make sense when you can recognize a sentence in the first place.

But grammar, punctuation, and capitalization also cannot be learned by trial and error, from writing and correction. Not even professional writers write enough to hypothesize all the possible constructions that they will eventually master, nor are they edited sufficiently to eliminate almost all possible sources of error. I am not arguing that writing does not play a role in learning to write, just as speaking is important for learning to talk. But the source of the information that makes us writers and speakers must lie in the language of other people, accessible to us only through reading and in listening to speech.

How and when do we learn it all? The language we know is enormously subtle and complex, and we learn almost all of it without even knowing that we are learning. We do not know what we know. How is all this learning achieved?

The Learning Brain

I can think of only one general explanation for how we manage to learn so much about language (and concurrently about so much else) and that is that the brain is learning all the time. Learning is not an occasional event, to be stimulated, provoked, or reinforced. Learning is what the brain does naturally, continually. It is only in artificially contrived experimental or instructional situations that the brain usually finds itself not learning – and tolerating not learning. (And even then, learning is probably taking place; that such experimental or instruction situations are artificial, unproductive, and boring.)

This is the time bomb in every classroom – the fact that children’s brains are learning all the time. They may not learn what we want them to learn. They may not learn what we think we are teaching them. But they learn, if only that what we try to teach them is boring or that they are unlikely to learn what we think we are teaching. Learning is the brain continually updating its understanding of the world; we cannot stop the brain from doing this. The hazard of so much instruction is not that children do not learn, but what they learn.

The important question for researchers is not how we learn, in the sense of the underlying brain processes, but why a brain which normally learns so effortlessly, so cont...

Table of contents

• Cover
• Title
• Copyright
• CONTENTS
• Introduction
• PART I From Essays Into Literacy
• PART II From Unspeakable Acts, Unnatural Practices
• PART III From Joining the Literacy Club
• PART IV From Reading Without Nonsense
• PART V From Between Hope and Havoc
• PART VI From Comprehension and Learning
• PART VII From Insult to Intelligence
• PART VIII From The Glass Wall
• Index