how children learn to write words oxford university press (2014)

How Children Learn to WriteWords

1
How Children Learn to
WriteWords
 

 
3
Oxford University Press is a department of the University of Oxford.
It furthers the University’s objective of excellence in research, scholarship,
and education by publishing worldwide.
Oxford NewYork
Auckland Cape Town Dar es Salaam Hong Kong Karachi
Kuala Lumpur Madrid Melbourne Mexico City Nairobi
New Delhi Shanghai Taipei Toronto
With ocesin
Argentina Austria Brazil Chile Czech Republic France Greece
Guatemala Hungary Italy Japan Poland Portugal Singapore
South Korea Switzerland Thailand Turkey Ukraine Vietnam
Oxford is a registered trademark of Oxford UniversityPress
in the UK and certain other countries.
Published in the United States of Americaby
Oxford UniversityPress
198 Madison Avenue, NewYork, NY10016
© Oxford University Press 2014
All rights reserved. No part of this publication may be reproduced, storedina
retrieval system, or transmitted, in any form or by any means, without theprior
permission in writing of Oxford University Press, or as expressly permitted bylaw,
by license, or under terms agreed with the appropriate reproduction rights organization.

Inquiries concerning reproduction outside the scope of the above should be sent to the
Rights Department, Oxford University Press, at the addressabove.
You must not circulate this work in any otherform
and you must impose this same condition on any acquirer.
A copy of this book’s Catalog-in-Publication Data is on file with the Library of Congress
ISBN 978–0–19–990797–7
9 8 7 6 5 4 3 2 1
Printed in the United States of America
on acid-freepaper
CONTENTS
Preface xi
Symbols and Abbreviations xiii
Abbreviations xiii
General Symbols
xiii
Phonetic Symbols
xiii
CHAPTER 1 Introduction 1
1.1 Writing asaTool 2
1.2 Orthographic Knowledge as a Part of Writing 6
1.2.1 Cognitive Resources and TechnicalTools 7
1.2.2 Social Implications of Nonstandard Spelling 9
1.3 How Can Orthographic Knowledge Be Achieved? 11
1.3.1 Discovery Learning and the Whole-Language
Approach 11
1.3.2
Direct Instruction and the Phonics Approach 13
1.4
Spelling and Reading 16
1.5

OrthographicReform 17
1.6
Past Work on Writing Systems and How They Are Learned 18
1.7
Our Approach 19
CHAPTER 2 Writing Systems 23
2.1 Outer Form of Scripts 24
2.2 What Writing Systems Represent 25
2.2.1 Representing Ideas:Semasiography 25
2.2.2 Representing Speech:Glottographic Writing 27
2.2.2.1 Representing Words and
Morphemes:Logography 28
2.2.2.2 Representing Syllables:Syllabary 30
vi
|
Contents
2.2.2.3 Representing Phonemes:Alphabetic
Writing 31
2.2.2.4 Representing Subphonemic
Features:Featural Writing 34
2.2.2.5 Mixed Writing Systems 35
2.3 How Writing Represents Meaning 41
2.3.1 How Semasiographies Represent Meaning 41
2.3.2
How Glottographies Represent Meaning 42
2.4
Composition of Characters 44
2.4.1
Composing Semasiograms 44
2.4.2

Composing Logograms 45
2.4.3
Composing Letters 46
2.5
Underrepresentation 47
2.6
Arranging Multiple Characters 51
2.6.1
Lines and Pages ofText 51
2.6.2
Grouping of Characters 52
2.7
Conservatism in Writing 55
2.8
Sound Change and its Eects on Writing 56
2.8.1
Types of SoundChange 57
2.8.2
Eects of Sound Change on Writing 58
2.9
Which Language Do WeWrite? 61
2.10
Conclusions 63
CHAPTER 3 Learning and Teaching 65
3.1 Statistical Learning 66
3.2 Learning Through Language 68
3.3 Implicit and Explicit Knowledge 69
3.4 Learning of Language and Learning
About Language 71
3.5 Formal and Informal Teaching 76

3.6 Conclusions 83
CHAPTER 4 Theories 84
4.1 Rote Memorization 85
4.2 Dual-RouteTheory 86
4.3 Constructivism 91
4.4 PhaseTheory 93
4.5 Connectionist Theories 95
4.6 IMP 97
4.7 Methods of Testing the Theories 101
4.8
Conclusions 102

|
vii
CHAPTER 5 Graphic Form 104
5.1 Surface Properties of Writing 104
5.2 Learning About the Surface Properties of Writing 108
5.2.1 Artificiality and Two-Dimensionality 108
5.2.2 Iconicity 108
5.2.3 Sequentiality and Directionality 109
5.2.4 Knowledge AboutUnits 111
5.2.5 Dierences Among Types of Writing 113
5.2.6
Dierentiating Writing From Pictures and
Numbers
114
5.2.7
Summary 117
5.3
Theories 117

5.4
Teaching 120
5.5
Conclusions 123
CHAPTER 6 Symbolic Function 124
6.1 Learning That Writing Stands for Something
OutsideItself 124
6.2 Learning What Writing Stands for andHow 125
6.3 Theories 131
6.4 Conclusions 132
CHAPTER 7 The Order of the Alphabet 133
7.1 Principles in Ordering 134
7.1.1 Arbitrary Ordering 134
7.1.2 Principled Ordering 136
7.1.2.1 Deletions 136
7.1.2.2 Insertions 138
7.1.2.3 Reordering 139
7.1.2.4 Other Scripts 140
7.2
When and How Children Learn About AlphabetOrder 143
7.2.1
Oral Methods 143
7.2.2
AlphabetBooks 145
7.2.3
Learning About Alphabet Order atSchool 148
7.3
How Does Knowledge of Alphabet Order Influence
Children?
148

7.4
Conclusions 151
CHAPTER 8 Symbol Shapes 152
8.1 Principles That Underlie Systems of SymbolShapes 153
8.1.1 Economy 153
8.1.2 Conservatism 159
viii
|
Contents
8.1.3 Beauty 159
8.1.4 Expressiveness 161
8.1.5 Similarity 162
8.1.6 Contrast 165
8.1.7 Redundancy 165
8.1.8 Summary of the Principles That Underlie Systems of
SymbolShapes 166
8.2
Learning and Use of Shapes as Graphic Objects 167
8.2.1
Learning About the Similarities Among the Shapes
of Writing
168
8.2.2
Learning About Contrasts Among the Shapes
ofWriting
170
8.2.3
Production 174
8.2.4
Learning Variant Forms ofShapes 179

8.3
Nonarbitrary Links Between Symbol Shapes and
Functions
182
8.4
Formal and Informal Teaching 188
8.5
Theories 192
8.6
Conclusions 193
CHAPTER 9 Letter Names 194
9.1 Principles That Underlie Systems of LetterNames 194
9.1.1 Phonetic Iconicity 195
9.1.2 Legality 200
9.1.3 Similarity 201
9.1.4 Contrast 203
9.1.5 Economy 203
9.1.6 Conservatism 204
9.1.7
Other Principles 204
9.1.8
Summary of Principles That Underlie Systems of
LetterNames
205
9.2
Learning the Phonological Forms of LetterNames 205
9.3
Do Children Benefit From the Phonetic Iconicity of
LetterNames?
207

9.4
Should Children Learn LetterNames? 210
9.5
Names of AuxiliaryMarks 212
9.6
Theories 214
9.7
Conclusions 215
CHAPTER 10 Early Spelling in Phonographic Writing Systems 216
10.1 Do Beginners Spell Using One Symbol for Each Syllable? 217
10.2 Letter Names and Early Spelling 218

|
ix
10.2.1 Spellings With Whole LetterNames 219
10.2.2 Partial and Inexact Matches to LetterNames 225
10.2.3 Conclusions About Letter-Name Spellings 229
10.3 OtherLabels 229
10.4 Phonological Analysis and Classification 230
10.4.1 Consonant ClusterOnsets 231
10.4.2 One Versus TwoSounds 234
10.4.3
Final Consonant Clusters 235
10.4.4
Other Ambiguities Involving Phonemes 237
10.4.5
Suprasegmental Features 238
10.5
Beyond Phonology 239
10.6

Teaching 243
10.7
Conclusions 247
CHAPTER 11 Complex Spellings 249
11.1 Beyond the Dichotomy of Regular Versus
ExceptionWords 250
11.2 Conditioning by Neighboring Segments 253
11.2.1 Coda-to-Vowel Conditioning 254
11.2.2 Onset-to-Vowel Conditioning 256
11.2.3 Vowel-to-Onset Conditioning 257
11.2.4 Vowel-to-Coda Conditioning 258
11.2.5
Do Rimes Have a Special Status? 259
11.2.6
Extended Spellings of Intervocalic
Consonants
260
11.2.7
Summary of Results on Conditioning by
Neighboring Segments
262
11.3
Conditioning by Position 263
11.4
Conditioning byStress 265
11.5
Conditioning by Morphology 266
11.5.1
Influences of Morphology on Spelling 266
11.5.2

Summary of Results on Morphological
Conditioning
271
11.6
Other Types of Conditioning 272
11.7
Unconditioned Inconsistencies 273
11.8
Other Complexities 277
11.8.1
Homographs 277
11.8.2
Words With More Letters Than
Phonemes
278
11.8.3
Additional Complexities 279
11.9
Summary of Findings on Learning of Complex Patterns 280
x
|
Contents
11.10 Teaching 281
11.11 Conclusions 282
CHAPTER 12 Punctuation and Capitalization 284
12.1 Punctuation 285
12.1.1 PunctuationMarks 286
12.1.2 Word Separation 292
12.2 Capitalization 294
12.3 Teaching 296

12.4 Conclusions 297
CHAPTER 13 Conclusions and Extensions 298
13.1 Evaluation of Theories of the Learning of Orthography 298
13.2 Broader Influences of Knowledge About Writing 303
13.2.1 Influences on Reading 303
13.2.2 Influences on Language 303
13.2.3 Influences Outside of Language 307
13.2.4 Summary of Writing’s Influences 309
13.3 Instruction About Orthography 309
13.3.1
Teach Patterns 310
13.3.2
Include Activities That Focus Attention on
Writing Itself
311
13.3.3
Provide Feedback AfterErrors 312
13.3.4
Don’t Assume TooMuch 314
13.3.5
Teach Teachers as Well as Children 314
13.3.6
It’s Just Orthography 315
13.4
Assessing Children’s Spelling 316
13.5
Dierences Between Children 317
13.6
FinalWords 317
References

319
Author Index
363
Subject Index
379
PREFACE
T
he goal of this book is to examine how children learn to use writing systems.
To do so, we must understand the nature of writing systems themselves and
the nature of human learning. Thus, after an introductory chapter, we include
a chapter on each of these topics. In the chapters that follow, we discuss how
children learn about various aspects of writing. We include chapters on the many
things that are needed for correct use of a writing system, ranging from learning
about the shapes of the symbols to learning about punctuation. In each chapter,
we review and integrate research that has been done with learners of a variety of
dierent writing systems.
This book draws from a number of research traditions, including linguistics,
psychology, and education. Each field has its own terminology. We avoid special-
ized terms when possible. However, it is necessary or convenient to use some
terms that will be unfamiliar to some readers. Aterm is defined the first time it
is used, so looking up the term in the index will help you to find its definition.
We have kept the number of abbreviations and special symbols to a minimum.
But the few abbreviatory devices and typographic conventions that we do use are
employed repeatedly, so we invite you to familiarize yourself with the Symbols
and Abbreviations section hereafter. The largest list there is that of symbols of
the International Phonetic Alphabet, which are used to unambiguously indicate
pronunciation in various languages. If you are familiar with the IPA, you may
still find it useful to look over the list, because the examples illustrate how we
apply the IPA to North American English. If you don’t wish to memorize the IPA,
the most important thing to know is that anything written between slashes // or

square brackets [] is an IPA symbol, which you can look up as the need arises.
Many people helped with thisbook.
We thank Lindsey Clahsen, Rochelle Evans, Nicole Rosales, and Suzanne
Schechtman for help with references, and Kristina Decker for help with figures.
For reading drafts of various sections, we thank Rachel Adler, Ifani Aguanunu,
Justin Bender, Stephanie Berk, Siti Binte Faizal, John Caselli, Jen Coane, Jeremy
xii
|
Preface
Cohen, Kristina Decker, Todd Dragity, Heather Hayes, Benjamin Karaus, Diana
Leader-Cramer, Annukka Lehtonen, Michelle Lindblom, Michael Mishkin,
Amanda Mount, Sarah Moynan, Jessie Munger, Shoko Otake, Sarah Robins,
Nicole Rosales, John Schmidt, Sara Shapiro, Emily Stein, Erin Suron, James
Wherley, Xiao Wen, and LanZhang.
For helpful discussions of specific issues, we thank Selma Babyiğit, Markéta
Caravolas, Alain Desrochers, Dimitra Ioannou, Karin Landerl, Kevin Miller,
Tatiana Cury Pollo, Dorit Ravid, Sofia Vernon, and Heinz Wimmer.
Support for the preparation of this book was provided in part by a Leverhulme
Fellowship for work at the University of York, an Honorary Fellowship for work
at the University of Tasmania, and a grant from the National Institutes of Health
(HD051610).
SYMBOLS AND ABBREVIATIONS
1 Abbreviations
C consonant; for example, CC is a sequence of two consonants
IMP integration of multiple patterns
IPA International Phonetic Alphabet (International Phonetic
Association, 1999)
V vowel; for example, CV is a consonant followed by avowel
2 General Symbols
italics

Italicized content is being mentioned, cited, defined, or referred to in
running text:“It depends on what the meaning of the word isis.”
sans serif Sans serif font is used to show that the content isn’t from printed text.
It usually implies that the content is handwriting.
‹ › Angled quotes enclose content that is being cited as characters of
writing: “The word ax is sometimes spelled ‹axe›, but the child
wrote ‹oxe›.”
‘ ’ Single quotes enclose meanings: “the Portuguese word avô
‘grandfather’ ”.
§2.3 The section symbol refers to sections in the book where a topic is dis-
cussed in more detail.
(12) Numbers in parentheses refer to displayed examples. Numbering of
examples starts with (1)in each chapter. References refer to examples
in the same chapter unless otherwise stated.
3 Phonetic Symbols
When we cite words written in a script other than the Latin alphabet used for
English, we add a pronunciation using an ocial or otherwise well-known
romanization scheme for the language. When there is no well-known scheme,
or when it is important to give precise descriptions of speech sounds, we use the
IPA. IPA transcriptions are readily identified by the fact that they are enclosed in
xiv
|
Symbols and Abbreviations
slashes or square brackets. The following is an alphabetical list of the symbols
we use. We give their technical definition and some example words that contain
the sounds represented by the symbols. For most purposes, you should find the
examples sucient. Unless otherwise specified, example words are English and
assume a General American pronunciation. Because phonemic transcriptions
are somewhat simplified, the same symbol won’t always represent exactly the
same sound across all languages, but the sounds should be recognizably similar.

The technical definitions may also help you pronounce an unfamiliar sound
and better follow some of our discussions of phonetics. Further information
can be found in International Phonetic Association (1999).
The IPA symbols are defined in terms of the articulatory features of the
sound they represent:how they are produced. Consonants are defined mainly
by the place of articulation, manner of articulation, and voicing.
The place of articulation is where the tongue or lower lip moves in order to
shape thesound:
labial the upper lip, as for /b/, /m/, /p/
dental the upper teeth, as for /ð/ and /θ/
alveolar the gums behind the upper teeth, as for /d/, /l/, /n/, /s/, /t/, and /z/
postalveolar the roof of the mouth just behind the gums, as for /d͡ʒ/, /ʃ/, /t͡ʃ/, and /ʒ/
palatal the hard palate, the bony part of the roof of the mouth, as for /j/
velar the soft palate, as for /ɡ/, /k/, and /ŋ/
pharyngeal the throat
glottal the vocal folds (also called vocal cords) in the larynx, as for /h/
Place of articulation can be described in more detail by also describing what
the lower lip or tongue isdoing:
bilabial lower lip approaches upper lip, as for /b/, /m/, /p/
labiodental lower lip approaches upper teeth, as for /f/, /v/
retroex the tongue is curled back
Manner of articulation describes how the tongue or lip interferes with the
flow ofair.
plosive a complete blockage, resulting in a little burst of noise when released, as
for /b/, /d/, /ɡ/, /k/, /p/, and /t/
nasal air is completely blocked in the mouth, but escapes through the nose, as
for /m/, /n/, and /ŋ/
fricative air is highly restricted in the mouth, causing a turbulent noise, as for
/ð/, /f/, /s/, /ʃ/, /v/, /z/, /ʒ/, and /θ/
aricate a plosive that is released gradually, resulting in a fricative-like noise

rather than a burst, as for /d͡ʒ/ and /t͡ʃ/ (see also §2.2.2.3).
approximant the airflow isn’t blocked enough to make any noise, as for /j/, /l/, /ɹ/,
and /w/
  
|
xv
When producing lateral sounds such as /l/, the air flows around the sides of the
tongue.
Voicing describes whether the vocal folds are vibrating, resulting in the buzzing
sound called voice. Most sounds are voiced, but some sounds are voiceless, as /f/, /k/,
/p/, /s/, /ʃ/, /t/, /t͡ʃ/, and /θ/.
Vowels don’t block the airflow at all. They are defined in terms of how far the body of
the tongue moves in two dimensions. Open vowels (also called low vowels), such as /ɑ/
and /æ/, are pronounced with the tongue low in the mouth; close vowels (also called high
vowels), such as /i/, /ɪ/, /u/, and /ʊ/, are pronounced with the tongue high in the mouth;
mid vowels, such as /ə/, are pronounced with the tongue elevated halfway. Intermediate
degrees of openness can be expressed by compound terms such as open-mid, for /ɛ/
and /ɔ/, or close-mid, for /e/ and /o/. The second dimension is how far the body of the
tongue moves forward or backward. Front vowels place the tongue body near the hard
palate (/æ/, /e/, /ɛ/, /i/, and /ɪ/); back vowels place it closer to the soft palate or the
throat (/ɑ/, /o/, /ɔ/, /u/, and /ʊ/); a central vowel like /ə/ is halfway between. Alax
vowel is pronounced with the tongue relaxed more toward the mid-central position than
comparable vowels. The vowel /ɪ/ is a lax version of /i/, and /ʊ/ is a lax version of /u/.
Certain vowels are further characterized by being pronounced with rounded lips (/o/,
/ɔ/, /u/, and /ʊ/) or by being nasalized, allowing air to flow through the nose.
Vowels are usually the most prominent part, or nucleus, of a syllable. Consonants are
relegated to the onset or coda of a syllable—the positions before or after the nucleus. It is
also possible for a consonant to form the nucleus, which is indicated in IPA by a vertical
line under the consonant letter:apple /æpl̩/.
/ / Slashes enclose pronunciations in IPA, in a broad transcription: “English cat

/kæt/”. The elements in such a transcription represent phonemes, sounds that
are represented only at the level of detail required to distinguish between words
in a particular language (§2.2.2.3).
[ ] Square brackets enclose pronunciations in IPA:“English cat starts with a [kʰ].”
The elements in such a transcription represent phones, that is, individual seg-
ments of speech (§2.2.2.3). We use this notation in contexts where we aren’t
referring to phonemes in a particular language or when the level of phonetic
detail is more specific than that required for phonemic transcription.
a open unrounded vowel:out /aʊt/, aisle /aɪl/, Spanish paso /ˈpaso/ ‘step’
æ raised front open unrounded vowel, with the tongue raised a little higher than
for /a/:apple /ˈæpl/
ɑ back open unrounded vowel:father /ˈfɑðɚ/
ʌ back open-mid unrounded vowel: An old-fashioned pronunciation in butter
/ˈbʌtɚ/
b voiced labial plosive:big /bɪɡ/
c voiceless palatal plosive:Irish céad /ceːd/ ‘first’
d voiced alveolar plosive:dog /dɑɡ/. Often a tap in American English:kitty /ˈkɪdi/
[ˈkʰɪɾi]
d͡ʒ voiced postalveolar aricate:jump /d͡ʒəmp/
ɖ voiced retroflex plosive:Swedish bord [buːɖ] ‘table’
xvi
|
Symbols and Abbreviations
ð voiced dental fricative:this /ðɪs/
e front close-mid unrounded vowel:ape /ep/
ə central mid vowel:above /əˈbəv/
ɚ another representation of /ɹ ̩/:/ɹ/ used as a syllable nucleus:bird /bɚd/
ɛ front open-mid unrounded vowel:edit /ˈɛdɪt/
f voiceless labiodental fricative:fine /faɪn/
ɡ voiced velar plosive:go /ɡo/

h glottal fricative:happy /ˈhæpi/
ʰ aspirated:followed by a pu of air that continues the voicelessness of the plosive
into the next phone:cat [kʰæt]
i front close unrounded vowel:marine /məˈrin/
ɪ lax front close unrounded vowel:itch /ɪt͡ʃ/
j palatal approximant:yes /jɛs/
k voiceless velar plosive:kid /kɪd/
l alveolar lateral approximant:look /lʊk/
m labial nasal:moon /mun/
n alveolar nasal:now /naʊ/
ŋ velar nasal:sing /sɪŋ/
o back close-mid rounded vowel:over /ˈovɚ/
ɔ back open-mid rounded vowel:all /ɔl/; boy /bɔɪ/
p voiceless labial plosive:pay /pe/
r alveolar trill:Spanish río /ˈrio/ ‘river’. We also use this symbol when a language
has an [r] -like sound but its exact articulation isn’t important to the discussion.
ɹ postalveolar approximant:read /ɹid/
ɾ alveolar tap:butter [ˈbəɾɚ], Spanish cara /ˈkaɾa/ ‘face’
ʁ uvular approximant:French Paris /paʁi/
s voiceless alveolar fricative:sound /saʊnd/
ʂ voiceless retroflex fricative:Swedish färsk [fæʂk] ‘fresh’
ʃ voiceless postalveolar fricative:sheep /ʃip/
t voiceless alveolar plosive:toy /tɔɪ/
t͡ʃ voiceless postalveolar aricate:teach /tit͡ʃ/
ʈ voiceless retroflex plosive:Swedish karta [kʰɑːʈa] ‘map’
u back close rounded vowel:rude /ɹud/
ʊ lax back close rounded vowel:pull /pʊl/
v voiced labiodental fricative:vat /væt/
w /u/ used in a syllable onset:weed /wid/
  

|
xvii
x voiceless velar fricative:German Buch /buːx/ ‘book’
z voiced alveolar fricative:zoo /zu/
ʒ voiced postalveolar fricative:measure /ˈmɛʒɚ/
β voiced bilabial fricative:Ewe Èʋe /βe/ ‘Ewe’
θ voiceless dental fricative:think /θɪŋk/
ʔ glottal stop:Arabic
أ as in /ʔislaːm/ ‘Islām’
ʕ voiced pharyngeal fricative:Arabic ع as in /ʕiːd/ ‘feast’
͡ The tie connects two characters that together represent a single phone; e.g.,
/t͡ʃ/, /d͡ʒ/
̃ A tilde over a letter means the sound is nasalized:French bon /bõ/ ‘good’
́ An acute accent over a letter means the sound is pronounced with a high tone.
That is, the pitch of the voice is elevated (§2.2.2.2).
̀ A grave accent over a letter means the sound is pronounced with a low tone. That
is, the pitch of the voice is lowered.
̩ A vertical bar under a letter means that the sound is a syllabic consonant, that is,
one that appears as the nucleus of a syllable, in place of a vowel:apple [ˈæpl̩].
̞ A downward tack under a fricative letter means the sound is pronounced
weaker, as an approximant:Spanish pueblo [ˈpweβ̞lo] ‘people’
ˈ The following syllable is stressed:believe /bəˈliv/
ː The preceding phone is longer than a comparable basic phone:Italian nonna
/ˈnonːa/ ‘grandmother’
˨˩, ˦ The preceding syllable is pronounced with a low, falling tone, a fairly high tone,
etc. The line to the left of the vertical sta iconically shows the relative height
and contour of the pitch of the voice.

HUMANS HAVE A REMARKABLY powerful means of communication: language.
Human language allows people to communicate virtually any type of thought:a

current perception, a memory, a new idea. Language has permitted people to
cooperate in creative ventures to an unprecedented degree, and it is one of the
most important foundation stones of human culture. Language, however, has
a serious shortcoming:It is naturally evanescent. Asentence lives on only in
the (imperfect) memory of those who hear it spoken or, in the case of a signed
language, those who see it signed. Without special technology, one can use
language to communicate with someone who is nearby but not with someone
who is in the next town or the next century. To overcome these limitations
of space and time, humans have worked out other methods of communica-
tion, ones that are designed to last. These permanent communication systems,
which are often visual, allow people to convey information and ideas to those
who are remote in time and space. They vastly increase the range over which
people can cooperate.
The system of permanent communication that we discuss in this book,
writing, is much younger than spoken language. While spoken human lan-
guage must have been around for at least 50,000years, the earliest writing is
only about 5,000years old. Despite its relative youth, writing has been vitally
important because it allows people to learn about things that they haven’t
directly experienced and because it allows information to be preserved for long
periods of time. In modern societies, people gain much of their knowledge
about the world through writing rather than through direct personal experi-
ence. Indeed, it has been said that “humankind is defined by language; but
civilization is defined by writing” (Daniels, 1996,p.1).
Our goal in this book is to examine what types of writing systems exist
in modern societies and how children learn to use these systems for pur-
poses of production. We discuss how children learn about such things
as the shapes and names of letters, the spelling of individual words, and
Introduction
CHAPTER 

2
|
How Children Learn to Write Words
capitalization and punctuation. These are all part of orthography, the cor-
rect way of using a writing system to write a language. To set the stage, we
begin this introductory chapter by discussing writing as one of many tools
that people have developed. We discuss the need to master such things as
spelling and punctuation in order to make eective use of the tool, and we
introduce some of the controversies about how such mastery is achieved.
We consider how the learning and use of orthography have been studied in
the past, and we preview the approach that we will take in the remainder of
thisbook.
1.1 Writing asaTool
Writing is one of many tools that people have developed to circumvent their
natural limitations. To understand the nature of writing, and to understand
how children learn to produce it, it is helpful to view writing within the frame-
work of tools more generally.
Some of the limitations that motivate the development of tools reflect
people’s physical characteristics and those of the environment. For example,
feet can take us a certain distance in a day, but no further. Our metabolism
can keep us comfortable at moderate temperatures, but not at extremes.
Fortunately, evolution has equipped people with the ability to invent and use
tools. People have developed cars and warm fabrics, among other things, in an
attempt to overcome their physical limitations.
People have cognitive limitations as well as physical ones. Despite a
common belief, it is not true that human memory is like a video recorder.
According to that view, people would take in detailed information about each
event they experience. They would record one event as easily as another, stor-
ing it in memory like a video clip. They would later replay the stored informa-
tion “as it echoes in the mind’s ear or flashes in the mind’s eye” (Brainerd,

Wright, Reyna, & Payne, 2002, p.121). However, psychologists now under-
stand that the human mind doesn’t usually work this way. Limitations in
perception and attention mean that people don’t take in everything that
occurs around them. Limitations in memory mean that some information
isn’t stored and that what is stored fades over time. People can hold just a
small number of items at one time in short-term memory, the part of memory
that retains information that is currently active. Information in this memory
store decays quickly, in around 30 seconds or so, unless special steps are
taken to prolong it. These limitations make it dicult to do such things as
add a long series of numbers in one’s head. People can store a great deal
of material in long-term memory, the part of memory that retains informa-
tion over long periods of time, even indefinitely. However, it takes eort to
get things into and out of this vast store. Some things tend to be remem-
bered well, such as information that is personally relevant or the meaning of
what someone said. Other things tend to be remembered poorly, such as the

|
3
exact words that were used to do so. Someone may be temporarily unable to
retrieve a piece of information from long-term memory, such as a word that
she doesn’t use veryoften.
People have developed various tools in an attempt to circumvent their cog-
nitive limitations. Some cognitive tools, such as calculators and abaci, are out-
side the body but may become, in some ways, extensions of it. Other cognitive
tools, such as remembering a list of things by the mnemonic method of asso-
ciating each one with a place on one’s daily walk to work, are internal. When a
person writes down a list of numbers that he wants to add, his friend’s address,
or the words that someone said, he is using an external tool in an attempt to
circumvent his internal cognitive limitations. When he imagines the spelling
of a word, he is internalizing atool.

Tools allow people to do certain things more economically than they other-
wise could, and they often improve the quality of the result. Indeed, some tools
allow people to do things that would otherwise be impossible. For these rea-
sons, individuals and societies are willing to put in the time and eort needed
to develop, learn about, and use tools. It’s easy to learn to use a television
remote control device, and it saves one from getting o the couch to change
the channel. Writing is harder to learn, but the eort returns much more.
When a tool is dicult to learn and use, additional tools may be developed as
aids, such as dictionaries in the case of writing.
The design, learning, and use of tools are influenced by people’s tendency
to satisfice (from a blend of satisfy and suce; Simon, 1957):to do the least they
can in order to attain what they consider to be a satisfactory result. Achieving
perfection often requires a good deal of time and eort, and people are often
willing to accept a less than perfect result if it is easier or quicker. Thus, eco-
nomy is a strong influence on human behavior. Economy may be fostered in a
various ways. For example, production often becomes easier and quicker when
movements overlap orblend.
The tools that people invent are influenced by their abilities and limitations,
even as they are designed to overcome the latter. For example, the button on
a remote control device for a television is designed so that a human thumb
or finger can easily depress it. Writing is the way it is, in part, because people
invented it and must learn to use it. Acomputer could handle a writing sys-
tem in which each word was represented with a unique and detailed shape,
the shape for snow being quite dierent from the shapes for snows and snow-
man. However, children would have great diculty learning and using such a
system.
The nature of a tool depends on what results people consider satisfactory.
If it is satisfactory to have a general idea of the time, a sundial will suce. If a
high degree of precision is required, an atomic clock may be needed. Scriptio
continua, a method of writing that uses only letters but no spaces or punc-

tuation (1), is satisfactory if people expect to have to study and annotate a text
before reading it aloud. If people expect to be able to read unfamiliar texts
quickly and easily, as they do nowadays, scriptio continua no longer suces.
4
|
How Children Learn to Write Words
Punctuation and word spacing demand eort from writers and aren’t always
strictly necessary; (1)has only one reasonable interpretation. But the spaces
that appear between words in many writing systems provide additional cues
that benefit readers.
(1) BEFOREADITWASCOMMONTOLEAVENOSPACES
BETWEENWORDS
In judging whether an outcome is satisfactory, people tend to consider the
present rather than the future. For example, one can sometimes achieve a sat-
isfactory result with a computer program by using a series of familiar com-
mands rather than a more ecient but unfamiliar command, or by asking
one’s spouse or colleague for help. Acomputer user may therefore be reluctant
to learn another method that, although dicult at first, would save time in the
long run. This means that even experienced users of certain computer tools
may routinely use inecient procedures (Bhavnani & John, 2000). Such phe-
nomena, together with the common observation that people sometimes post-
pone medical tests and dental appointments, show that people find it dicult
to discount local rewards in favor of greater global ones. As we will see, these
things aect the learning of writing.
What is satisfactory for the tool maker and what is satisfactory for the tool
user don’t always agree. Ease of use trumps ease of manufacture when many
people use a tool but few must make it. This is the case for an abacus or a tele-
vision remote control device. With other tools, including writing in the case
of modern societies, people expect to be able to produce the tool themselves.
The needs of writers and readers aren’t identical, and writing systems must

balance thetwo.
Those who are learning about a tool and those who are experts in its use
may also have dierent needs. For example, a beginner may find it easy to use
a computer program that depends on selecting from functions displayed on
the screen, but having to call up menus may impede an expert. With writing
systems, too, there are trade-os between the needs of learners and experts.
Learners’ needs are especially important in many societies that value mass
literacy. Korean used to be written with Chinese characters, a system that
could take many years to master. But an emerging belief in the 1400s that
literacy should not just be the domain of the educated male elite led to the
development of an alphabetic writing system that could be learned quickly.
This system, hangeul, opened up literacy to women and to all classes, but it
encountered opposition from the educated elite who were already expert in the
use of Chinese characters.
Tools are invented to serve particular functions:facilitating movement in
the case of wheels or freezing language in the case of writing. Even though
a tool is a means to an end, people want it to be attractive as well as useful.
Thus, they may make the eort to decorate a tool even when decoration doesn’t
enhance its function. Apot with designs carved on it cooks no better than a pot

|
5
without, but people prefer it because it is attractive. The designs may identify
the owner as a member of a particular social group or may express his indi-
vidual tastes. Writing, like other tools, has been shaped by people’s desires for
beauty and expression.
The availability of a cognitive tool means that people don’t have to do certain
things on their own. For example, calculators mean that people don’t have to
add and multiply in their heads, and paper or computer calendars mean that
they don’t have to store the time of a meeting in their own memories. When

a new cognitive tool is introduced, debates often arise about whether people
will get dumber if they use it. What will they do if a tool is temporarily unavail-
able, as when a cash register breaks and a cashier can no longer make change?
Similar debates arise in the case of writing. Before writing was invented, cer-
tain important texts existed only in the minds of learned people who preserved
that knowledge in their excellent memories. Some Greek thinkers, including
Socrates, feared that people would become less intelligent when they wrote,
because their memories would atrophy. That doesn’t seem to have happened,
but it is true that people sometimes use writing as a form of external stor-
age rather than storing information in their own memories (Eskritt, Lee, &
Donald,2001).
Learning to use the tools of a society in the agreed-upon way is an impor-
tant part of becoming a full member of that society, and children are highly
motivated to do so. When a tool is widely used, children see many examples
of it from an early age. They see the actions that experienced tool users per-
form, and they see the results that ensue. Transmission of tool use from one
generation to the next is facilitated by children’s tendency to observe and imi-
tate the purposeful actions of experienced others, especially others from their
own social group. So strong is children’s drive to do this that they may even
reproduce actions that yield no obvious or interesting results, whose functions
they don’t understand, or that have been demonstrated to other people rather
than to the children themselves (Lyons, Young, & Keil, 2007; Nielsen, Moore,
& Mohamedally, 2012; Yang, Sidman, & Bushnell, 2010). Thus, young children
who see adults or older children write may try to reproduce the movements
and the results. They start learning about some of the formal properties of
writing—what it looks like—before they know much about how writing func-
tions to symbolize language.
Transmission of tool use from one generation to the next is facilitated not
only by children’s drive to fit in but also by adults’ tendency to demonstrate and
explain (Csibra & Gergely, 2009). Adults may consciously or unconsciously

modify their tool use in an attempt to help children learn about the tool. They
may signal that an action is intended for children to imitate, they may take
steps to ensure that children attend to the action, or they may perform an
action in a repetitive or exaggerated way (Brand, Baldwin, & Ashburn, 2002).
And adults may explain to children what they are doing and why. Their lan-
guage may highlight distinctions that are dicult to glean from observation
or convey generalizations that go beyond particular instances. Adults expect
6
|
How Children Learn to Write Words
children to learn to use tools in the culturally prescribed manner, and they may
verbally or nonverbally correct children who don’t conform.
Transmission of tool use from adults to children is also facilitated by the
fact that young children, who have so much to learn, tend to think that adults
know everything. Understanding why the sky is blue is dicult enough that
an advanced student of physics may be asked to explain this on an exam.
Six-year-olds, however, generally think that their parents understand such
things.
Tools are shared by members of particular groups. A child who knows
that the members of a group have the same conventions about tool use can
learn from any experienced tool user in the group. The child need not learn
new facts for each person. For example, a child who hears his mother refer to
the shape V as /vi/ can assume that his father and his aunt call it /vi/ too.
(See Introduction, “Symbols and Abbreviations,” 3, for a key to the phonetic
symbols.)
Learning about a tool allows a child to use it, but it may have broader eects
as well. This happens because tools aren’t neutral; they have their aordances
and their constraints. For example, two-dimensional maps distort curved sur-
faces. This can lead to distortions in thinking, such as thinking that Greenland
is larger than Africa because of how it appears on most maps. But maps help

people, too; for example, by helping them to understand the spatial relations
among parts of a city that they may not have noticed before. Similarly, learning
a writing system helps children to understand some aspects of language; for
example, that a continuous string of speech can be conceptualized a series of
smaller units. At the same time, the tool promotes certain fictions, such as the
idea that the latter portion of spit sounds exactly like pit. It doesn’t. If the /s/ of
spit is removed from an audio clip, the remainder sounds likebit.
1.2 Orthographic Knowledge as a Part of Writing
People write to record their ideas, to share them with others, and to search
for information, as in a catalog or on the Web. Everyone agrees that children
should learn to do these things in order to participate fully in a modern soci-
ety. But how important is it to learn the details of spelling, punctuation, and
handwriting? For example, if a computer search engine will find information
on ptarmigans when a child types ‹tarmigan›, must the child learn to pro-
duce the conventional spelling? Some people believe that, just as the invention
of writing freed people from some of the constraints of spoken language, so
the invention of spelling checkers and grammar checkers frees them from
the need to learn about the details of orthography. As one teacher said, “My
kids don’t need to worry about correct spelling, because they can get the right
spelling by using the spell-checkers on the computer” (Graham, Harris, &
Fink Chorzempa, 2002, p.683). Consistent with this view, researchers and
educators sometimes call spelling a low-level skill. This is often understood as