Elizabeth Johnston, Investigating Minds, 12/5/97

Lecture 22: Learning to Read

Phonics vs Whole Language

How to best teach reading is a hotly debated, politicized topic. People refer to the 'reading wars', with the two sides of this particular dichotomy represented by "phonics" and "whole language". One side emphasizes the skills involved in recognizing written words, it is analytic focusing on the phonic elements of words. The other side emphasizes comprehension of words in a meaningful story context. At first it is difficult to comprehend the rationale for this polarization; in a comprehensive recent text on the subject of 'Beginning to Read' Marilyn Jager Adams argues that both sides of the debate are deeply rooted in the nature of our writing system. In order to understand the difficulty and challenges of learning to read we must first examine the nature of the writing system.

Writing Systems

Logograms

The earliest writing systems consisted of pictures. In terms of representational transparency this is an ideal system; anyone can look at a picture of a dog and understand the symbol. The major disadvantages as a writing system are the need to interpret complex pictures and the difficulty of reproducing the symbols. To solve these problems, logographic scripts evolved with each word represented by a schematized and standardized picture. The Sumerians of ancient Mesopotamia are credited with the establishment of the first logographic writing system. It spread widely to other cultures, and is still the prevailing symbology for a number of writing systems, such as Chinese. In order to represent unpicturable words new symbols were introduced moving the system away from a direct iconic representation. This version of the symbol system is less readily interpretable, consequently memorization of these symbols became necessary. The difficulty such a writing system presents is the large number of logograms needed: the Chinese writing system contains as many as 80,000 basic logograms, but even Chinese adults are said to have working knowledge of only 4,000 to 5,000.

The Rebus Principle

The notion is that two words that sound the same can be written by the same sign. For example the sound 'I' can be written by the pictogram and 'deer' by ; these can be combined using the rebus principle to represent the word 'idea' thus:
+ .
Two main types of sound writing developed from the logographic systems by development and extension of the rebus principle: syllabary and alphabet. The above example is a syllabic representation.

Syllabaries

Syllabaries are hypothesized to have developed from the need to represent foreign words and proper names. For these words it was necessary to be able to be able to retrieve verbatim forms from their written representations. The custom developed of representing the pronunciations of such terms with "sound-alikes" from the logographic system, that is, with strings of monosyllabic logograms. A syllable is a single vowel or a diphthong which may be preceded and/or followed by one or more consonants. Syllables are relatively salient units of spoken language so the rationale was intuitively clear. Languages which only allow a few syllable types are ideally suited for such a writing system. For example, Japanese allows only vowel alone or consonant+vowel syllable types, and it can be represented adequately with a vocabulary of about 50 syllabic signs. In English as many as four consonants can cluster in syllable-final position and consonants are allowed in both initial and final positions. By one estimate, English contains about 5,000 syllables. A syllabic script for English would present a formidable initial learning task.

Alphabetic Scripts

Eventually people began representing whole syllables by only their initial consonants. The Greeks were the first to introduce the alphabetic principle of one symbol for each elementary speech sound, or phoneme, in the language. This type of system has the advantage of conveying the full range of expressive meaning captured by logographs and syllabaries using far fewer symbols. The loss is in terms of transparency: pictures are symbolically transparent, logograms less so, and syllabograms and letters not at all. With the passage from each of these symbol systems to the next, the significance of the symbols became more and more abstract. The symbolic significance of the letters of the alphabet is so abstract as to not even exist in any physically robust sense. Letters map onto phonemes, but the sounds of individual phonemes have no constant or isolable correspondent in the speech stream, and many cannot even be produced in isolation (we can say /buh/ or /bah/ or /bih/, but not /b/). The following example from Rozin (1976) shows that the that there is no constant or isolable sound corresponding to /d/.
These sound patterns are suffficent to produce the syllables di and du. The dark bands on the graph represent the presence of energy at the indicated frequency and time. It is tempting to assume that the initial 50ms represent /d/ However, these segments differ for the two sounds (lack of constancy) and if they are played alone result only in the perception of a chirping sound that does not resemble /d/ at all. The sound spectrograms shown here are idealizations, they do sound like the appropriate syllables to the human ear, but the sound spectrograms produced by a human saying 'di' and 'du' are much more complex. The compactness and simplicity of the alphabet as a representational system has a cognitive toll: the referent of a letter is perceptually abstract and conceptually sophisticated. This is the first reason that cracking the alphabetic code is challenging: making sound-letter or grapheme-phoneme correspondences requires an identification of the units of each code and forming a mapping between them.

The difficulty of this enterprise is compounded when we take into consideration that written English is not perfectly alphabetic: the letters of English do not necessarily map one-to-one onto phonemes. Sometimes the phonemic significance of a letter is modified by the letter or even letters next to it (ran versus rain, sit versus sight), sometimes by one or more adjacent letters (bit versus bite, nation versus nationality), and sometimes only by the identity of the word (father/fathead).

Hints on Pronunciation for Foreigners
I take it you already know
Of tough and bough and cough and dough?
Others may stumble but not you,
On hiccough, thorough, laugh and through.
Well done! And now you wish,, perhaps,
To learn of less familiar traps?

Beware of hear, a dreadful word
That looks like beard and sounds like bird,
And dead: it's said like bed, not bead--
For goodness' sake don't call it "deed"!
Watch out for meat and great and threat
(They rhyme with suite and straight and debt.)

A moth is not a moth in mother
Nor both in bother, broth in brother,
And here is not a match for there
Nor dear and fear for bear and pear,
And then there's dose and rose and lose --
Just look them up -- and goose and choose,
And cork and work and card and ward,
And sony and front and word and sword,
And do and go and thwart and cart --
Come, come, I've hardly made a start!
A dreadful language? Man alive.
I'd mastered it when I was five.
Given the complex nature of the alphabetic code it is not surprizing that it takes children some time to crack it and that it can cause serious difficulties in learning to read. Alphabetic insight, realizing what type of system you are dealing with, is an early and necessary step in the reading process. Without alphabetic insight and phonemic awareness novel words cannot be sounded out, so the acquisition of new reading vocabulary is difficult. Children often begin with 'sight reading' of words: recognizing the whole pattern for common sequences such as 'The End' or 'Stop', but the memory demands of learning to read in this way are far too taxing to make it workable. There are many stages in learning to read that are determined by both the nature of the writing system and the cognitive processes readers bring to bear on the task.

The following table from Spear-Swerling and Sternberg (1997) outlines the necessary and sufficient stages in acquiring literacy.
Phase Defining Feature(s) Approximate Grade Approximate Age
Visual-cue word recognition Child uses visual cues (e.g., word shape, color, a logo), rather than phonetic cues, in word recognition Preschoolers and some kindergartners Two to Five
Phonetic-cue word recognition Child uses partial phonetic cues in word recognition Kindergarten and first grade Five to Six
Controlled word recognition Child makes full use of phonetic and orthographic cues in word recognition, but is not automatic in recognizing words Second grade Six to Seven
Automatic word recognition Child recognizes common words accurately and automatically Second to third grade Seven to Eight Years
Strategic Reading Child routinely uses strategies to aid comprehension Beginning at third to fourth grade Beginning at eight to nine years
Proficient adult reading Individual has higher-order comprehension skills Beginning at later high school or college level Beginning in later adolescence
An appreciation of the components involved in each stage provides a sound understanding of the ways in which reading acquisition can go wrong, and that brings us to the topic of the next lecture.

Sources