ChapterPDF Available

Learning to Read Hebrew

Authors:

Abstract and Figures

Around the world, children embark on learning to read in their home language or writing system. But does their specific language, and how it is written, make a difference to how they learn? How is learning to read English similar to or different from learning in other languages? Is reading alphabetic writing a different challenge from reading syllabic or logographic writing? Learning to Read across Languages and Writing Systems examines these questions across seventeen languages representing the world's different major writing systems. Each chapter highlights the key features of a specific language, exploring research on learning to read, spell, and comprehend it, and on implications for education. The editors' introduction describes the global spread of reading and provides a theoretical framework, including operating principles for learning to read. The editors' final chapter draws conclusions about cross-linguistic universal trends, and the challenges posed by specific languages and writing systems.
Content may be subject to copyright.
275
Chapter
Chapter Chapter
Chapter 7
77
7
Learning to
Learning to Learning to
Learning to r
rr
read Hebrew
ead Hebrewead Hebrew
ead Hebrew
David L. Share
David L. ShareDavid L. Share
David L. Share
Department of Learning Disabilities, Faculty of Education
and Edmond J. Safra Brain Research Center for the Study of Learning Disabilities
Haifa University
Introduction
IntroductionIntroduction
Introduction
1.1
1.11.1
1.1
Hebrew and its orthography
Hebrew and its orthographyHebrew and its orthography
Hebrew and its orthography
There has been growing concern among social scientists that conclusions from
studies conducted on highly educated populations from wealthy European cultures
(especially English-speaking) may have limited generalizability regarding human
behavior in general (Henrich, Heine, & Norenzayan, 2010). These reservations also
extend to the field of language (Beveridge & Bak, 2012; Evans & Levinson, 2009)
and literacy (Frost, 2012; Share, 2008). Reading research, in particular, has been
overwhelmingly dominated by work on English which, unfortunately, appears to be
an outlier among European alphabets (Share, 2008; 2012). Around the globe,
however, most individuals are neither native English speakers nor acquire literacy
in a European alphabet. The study of reading acquisition in Hebrew, therefore,
276
presents an intriguing case study of learning to read and write in a non-European
non-alphabetic orthography.
Since a detailed review of this literature was published in 1999 together with
my late colleague, Iris Levin (Share & Levin, 1999), this chapter will pick up where
the previous review left off by first summarizing prior findings and conclusions
(which I will shorthand as S&L99) before moving on to a sketch of current and
possible future research trends. Beyond the scope of the present chapter is
important work on second language Hebrew learning (Frost, Siegelman, Narkiss, &
Afek, 2013), and on special populations such as bilinguals (Geva, 2008; Leikin,
Schwartz & Tobin, 2011), the deaf (Miller, 2009), dyslexics (Share, 2003), and socio-
economically disadvantaged groups (Aram, 2005; Korat, 2005; Schiff & Lotem,
2011).
1.2
1.21.2
1.2
Synchronic and diachronic characterization
Synchronic and diachronic characterizationSynchronic and diachronic characterization
Synchronic and diachronic characterization
Hebrew, the language of the Jewish bible, the Christian Old Testament, has a long
and unbroken history. Used continuously from antiquity until today, its reformed
version became the official language of the modern state of Israel. Contrary to
popular belief, Hebrew was never a dead language. Even after the Biblical period
(circa 1300 BCE to 200 BCE), Hebrew continued to evolve, and although colloquial
usage ceased for some 1500 years, it remained in continual use not only in liturgy
and reading of sacred texts, but also in correspondence, and in scientific and
creative writing (Berman, 1997; Hetzron, 1987.)
277
Although Hebrew is not one of the world’s major languages (with only some
10 million speakers), Semitic languages are the native tongue of some hundreds of
millions, and the consonantal writing systems (abjads) first developed nearly four
thousand years ago by Semitic speakers are the progenitor of writing systems now
used daily by billions (Daniels & Bright, 1996; Diringer, 1948; Naveh, 1982). The
many offspring of the original Semitic abjad include today’s Semitic (e.g., Hebrew,
Arabic) and non-Semitic abjads, (e.g., Urdu, Malay), and all European alphabets
originating with the Greeks’ borrowing of the Semitic (Phoenician) abjad. This first
alphabet later gave birth to the Roman alphabet which was then disseminated first
throughout Europe, then across the globe by traders, empire-builders, and
missionaries. Eastward, the Semitic abjad evolved into the abugidas
(alphasyllabaries) of India, South-East Asia, and many parts of central Asia (Daniels
& Bright, 1996; Gnanadesikan, 2009).
1.3
1.31.3
1.3
Literacy and schooling
Literacy and schoolingLiteracy and schooling
Literacy and schooling
In Israel, formal reading instruction begins in Grade 1, around age 6, when children
are introduced to the pointed (fully-vocalized) Hebrew script. However, as in North
America, there is a great deal of informal literacy teaching in the pre-school years
emphasizing phonological awareness, letter knowledge, writing and invented
spelling, storybook reading, print conventions, language development and varieties
of discourse structure. On entry to school, children are normally taught via
systematic phonic-emphasis methods, but the unit of instruction is typically
not
the
individual letter but an integral CV syllable block (termed a /ʦɛruf/ ‘combination’)
278
consisting of a consonant letter with a small (diacritic-like) vowel sign underneath.
Owing to its regularity, children are expected to master pointed script by the end of
the first year or even earlier. Around Grade 3, the vowel points are dropped and
children are expected to be able to read unpointed (i.e., partly vocalized) text. (In
their writing, children are not expected to write the vowel points at any stage.)
Among the language-majority Hebrew-speaking (Jewish) population of
Israel, literacy levels are very high and, according to international comparisons such
as PIRLS, among the highest in the world.
Description of Hebrew and its written forms
Description of Hebrew and its written formsDescription of Hebrew and its written forms
Description of Hebrew and its written forms
2.1
2.12.1
2.1
Linguistic System
Linguistic SystemLinguistic System
Linguistic System
2.1.1
2.1.12.1.1
2.1.1
Phonology
PhonologyPhonology
Phonology
Modern Hebrew (Ivrit) has a simple vowel system consisting of five vowels; high
front /i/ and back /u/, midfront /ɛ/, back /o/ and low /a/—all pronounced close to
the cardinal vowels (Glinert, 1994). There are three dipthongs /aj/, /oj/, /uj/, each
with a glide. There is no longer any phonemic contrast between short and long
vowels or single versus geminate vowels (or consonants). In addition to consonants
familiar to English speakers (/p, b, t, d, k, g, f, v, s, z, ʃ, m, n, l/), and the semivowel
(/j/), Hebrew has an unvoiced uvular liquid (/ʁ/), a voiceless velar fricative /x/, the
affricate /ʦ/, and, of late, the loan /w/. Biblical Hebrew’s Semitic pharyngeals have
been neutralized to /x/ and the glottal stop to zero. Emphatics, too, have
disappeared in Modern Hebrew. Non-native affricates now exist in loan words (/ʧ/
279
for ‘chips’, /ʤ/ for ‘jeep’). The glottal stop and /h/ are now pronounced as zero in
most environments. Of the six original stop-spirant alternations, only three (/p/,
/b/, /k/) now have fricative counterparts, distinguished in pointed Hebrew script
by a dot (spirantization
dagesh)
inserted into the letter (see Table 1). The numerous
historical neutralizations create considerable challenges for spelling, with six of
Hebrew’s nineteen consonantal phonemes represented by at least two letters (see
Table 1). Few words are monosyllabic, most consisting of two and three syllables
and, owing to extensive clitics and inflections, possessives and pronominals, can
range up to five or more syllables.
Spoken Hebrew has a very extensive range of consonant clusters (e.g., /gz/,
/pʦ/, /ʦx/, etc.; Glinert, 1994; Schwarzwald, 2002) but because written forms
consist exclusively of CV sequences (consonant letter plus vowel sign), it is often
misrepresented as a language with a simple CV/CVC phonology—a classic case of
confusion between the written signary and the spoken language.
In summary, a once extensive inventory of phonemes (vowels and
consonants) has now been whittled down to a deceptively small set. Unfortunately
for the speller, the orthography adamantly preserves all these once-distinct
phonemes. Nonetheless, the orthography makes up for this spelling complexity
somewhat by simplifying the task of learning to read.
2.1.2
2.1.22.1.2
2.1.2
Inflectional morphology
Inflectional morphologyInflectional morphology
Inflectional morphology
Modern Hebrew has a rich inflectional morphology that permits considerable
versatility in word order. Nouns and adjectives are marked linearly for number,
280
near-arbitrary gender and definiteness, verbs are marked for number, gender and
person, take infinitive, imperative or indicative moods including past tense, future,
and present or participial forms. There is no grammatical marking of verb aspect.
2.1.3
2.1.32.1.3
2.1.3
Word formation pr
Word formation prWord formation pr
Word formation processes
ocessesocesses
ocesses
Hebrew is a synthetic (S)VO language with a highly productive morphology
(Berman, 1985; Ravid, 1990). Whereas English has many different lexemes for
semantically related words such as
play
and
game
, Hebrew uses derivational
devices that operate on a limited number of common tri-consonantal roots (leSaXeK
‘to play’, miSXaK ‘game’). Possibly because the lexical base of Old (Biblical) Hebrew
mainly consisted of just three consonants, (allowing only around 8000 (20
3
)
possible roots), the morphology assumed the burden of word-making. Many mono-
morphemic words (particularly Aramaic) entered the lexicon at a later time, but the
root-and-pattern word-formation apparatus remains dominant. This system, in
many ways, is the key to understanding the Hebrew writing system and the nature
of learning to read and write in Hebrew.
The heart of Hebrew word-formation is the uniquely Semitic root-and-
pattern system (McCarthy, 1981). Most Hebrew content words are poly-morphemic
combinations of a tri-consonantal skeleton interwoven with a vowel pattern (via
infixing, with or without affixed consonants). These morpho-phonemic vowel
patterns convert an unpronounceable and discontinuous consonantal root (e.g.,
דגב
,
B,G,D) into a pronounceable word BeGeD (
דֶגֶבּ
‘article of clothing’).
1
The consonantal
1
I adopt the common convention of representing the root letters with capital
letters and the vowels (and any non-root affixed consonants) with lower-case
281
root is the semantic core of most content words; hence, it is a fundamental unit of
processing in both the spoken and written language. To illustrate, verbs derived
from the tri-consonantal root
טלק
KLT include; KaLaT (‘he grasped’), niKLaT (‘was
grasped/absorbed’), hiKLiT (‘he recorded’) and many more. Most noun forms
operate on the same root-plus-pattern principle, e.g., KLiTah (‘absorption’), miKLaT
(‘shelter’), KoLTan (‘receptor’) and adjectives, KaLiT (‘absorbant/accessible’).
Notice, again, that where English has distinct lexemes, Hebrew has the common KLT
root morpheme. While some noun patterns represent semantic categories (for
example, the form CaCaC is characteristic of professions; KaTaV ‘journalist’, NaGaR
‘carpenter’), many are highly unpredictable.
In contrast to the English-speaking child’s early lexicon, the core lexicon
encountered by beginning readers in Hebrew is multi-morphemic (Ravid, 2001).
Even many common everyday nouns are morphologically complex and hence oblige
the Hebrew reader to attend to word-internal structure. Some additional Hebrew
word-formation devices include the increasingly common linear derivations
(Berman, 2003), e.g., SaFRan ‘librarian’, SaFRan-ut ‘librarianship’, (hyphenated)
prefixes bɛn- (‘inter-’), as in
ןיב
-
ימוחת
(bɛn-TXuMi, ‘inter-disciplinary’), blends –
בוחרדמ
(miDRaXoV), merging
הכרדמ
(miDRaXa ‘sidewalk’) and
בוחר
(ReXoV ‘street’) to
generate ‘pedestrian mall’. Straight non-native borrowings are very popular
nowadays (e.g.,
ייא
-
ןופ
/ajfon/ ‘i-phone’).
In new word formation, verbs can only be formed by combining a root with
one of the seven verb patterns (
binyanim
). These patterns express transitivity,
letters. This is somewhat misleading because Hebrew vowel diacritics (in pointed
script) mostly appear directly
beneath
(not beside) consonants.
282
reciprocity, reflexivity, causitivity, and inchoativeness. For example, GaDaL ‘he grew’
(simple intransitive), GiDeL ‘he grew’ (transitive), hiGDiL ‘he enlarged’ (causitive),
etcetera. Borrowed verbs must be adapted to an existing Hebrew pattern but nouns
can be borrowed intact (e.g., /intɛʁnɛt/ ‘internet’; Hoberman & Aronoff, 2003).
In contrast to content words, most Hebrew grammatical morphemes are
monosyllabic and mono-morphemic, including prepositions (e.g., /im/ ‘with’),
conjunctions (e.g., /ki/ ‘because’), pronouns (e.g., /at / ‘you’), quantifiers (e.g., /kol/
‘all’), and articles (e.g., /zɛ/ ‘this’).
As a highly synthetic language, many grammatical morphemes in Hebrew are
affixed to a stem, creating long poly-morphemic strings where a single letter string
can express an entire English sentence (cf. Turkish or Greek) thereby placing
considerable demands on morphemic analysis, and perhaps also slowing reading
rate (Shimron & Sivan, 1994).
Summarizing, Hebrew is heavily inflected, permitting flexible word order,
with word-formation primarily via non-linear (root-and-pattern) morphology,
although linear structures and direct borrowings are becoming increasingly
common. The poly-morphemic nature of even simple words calls for considerable
morphemic unpacking but, as we see below, the architecture of the writing system
excels in making the morphology transparent.
2.2
2.22.2
2.2
Writing System
Writing SystemWriting System
Writing System
Hebrew is an abjad or consonantal writing system (Daniels, 1990). This observation
has far-reaching implications for literacy learning, instruction and assessment.
283
Hebrew represents all consonants as fully-fledged letters linearly arrayed from right
to left. Vowels, however, are either omitted or have only subordinate status. And
although Hebrew writing has had many opportunities to become a “true” alphabet,
it has never abandoned its consonantal architecture. Thus vowels neither are, nor
ever were, on an equal footing with consonants. Yiddish (an Indo-European,
Germanic language), in contrast, is also written with Hebrew letters, but with full-
sized vowel and consonant letters arrayed linearly side by side as in all European
alphabets.
2.2.1
2.2.12.2.1
2.2.1
Script and punctuation
Script and punctuationScript and punctuation
Script and punctuation
Since Biblical times Hebrew was a purely consonantal script, practically identical
with the Phoenician abjad (see the leftmost column of Table 1). By the end of the
sixth century BCE, however, the Biblical script was gradually replaced by an
Aramaic script which evolved into the so-called Jewish
square
script—the standard
script used today (Goerwitz, 1996). As in ancient times, Hebrew has twenty-two
consonantal letters, five of which take a word-final form. Three letters have an
added apostrophe marking borrowed palatals and affricates (
ג
'
(/ʒ/,
צ
'
/ʧ/, and
ז
'
/ʤ/.
Unlike Arabic, Hebrew letters are never ligatured. Three letters represent
predictable stop/fricative alternations
b
/
v
,
k
/
x
,
p
/
f
which, in pointed Hebrew, are
marked by inserting a dot or point (
dagesh
) into the letter, (e.g.,
ב
).
2
2
These stop/fricative alternatives are not entirely predictable (e.g.,
רכמ
‘he
sold’,
רכמ
‘acquaintance’) hence the absence of this point in unpointed script adds
consonantal ambiguity.
284
With the post-biblical demise of spoken Hebrew, the introduction of vowel
signs became a necessity. Two separate systems of vocalization evolved. The first,
called
mothers of reading
, employs four consonantal letters (AHVY,
יוהא
) to serve the
dual function of signifying vowels as well as consonants (e.g.,
תכיבה
, /KTiVa/ ‘writing’
תכוב;
/KaTuV/ ‘written’).This system is the standard
unpointed
version of Hebrew
appearing in periodicals and books
but is inconsistent and incomplete (Ravid &
Schiff, 2006; Shimron, 1993) creating extensive phonological under-specification as
well as pervasive homography (Bar-On, in press).
A second system of vocalization employing diacritic-like dots or points
developed in the eighth century CE (Chomsky, 1941). So-called
pointed
Hebrew is
largely restricted to poetry, sacred texts, and materials for beginning readers. This
vocalization system provides a complete and virtually unambiguous representation
of the vowels by means of relatively inconspicuous dots and dashes appearing
mostly under, but sometimes also above and between the letters. For example:
ִד
/di/,
ד
/do/,
ֻד
/du/,
ָד
/da/,
ֶד
/dɛ/, etcetera. Note that each of these consonant-vowel
combinations forms a graphically integral CV syllable block. These same units are
the basic phonic building blocks of pointed Hebrew and also, as mentioned above,
the unit of initial instruction in most Israeli schools. However, duplication abounds
in this system with numerous signs preserving historically neutralized phonemic
distinctions such as vowel length. Children learn to read in pointed Hebrew, which
has almost perfect letter-to-phoneme correspondence, negligible homophony and
no homography. This phonological consistency allows beginning readers to rely
entirely on serial bottom-up letter-sound translation to identify novel printed words
285
with little need for supplementary contextual or higher-order lexical or morpho-
phonological information (see Shatil & Share, 2003).
Punctuation in standard Hebrew text adheres to the same conventions as in
English (as seen in Figure 1).
Figure 1.
The same text in standard unpointed and
pointed Hebrew.
Begin Table 1
Begin Table 1Begin Table 1
Begin Table 1
Table 1
Old (Paleo
Old (PaleoOld (Paleo
Old (Paleo-
--
-)Hebrew Letters, Modern Hebrew Letters, Their Names, and
)Hebrew Letters, Modern Hebrew Letters, Their Names, and )Hebrew Letters, Modern Hebrew Letters, Their Names, and
)Hebrew Letters, Modern Hebrew Letters, Their Names, and
Phonemic Values
Phonemic ValuesPhonemic Values
Phonemic Values
Biblical
sign Modern sign
(word-final
form) Letter
name Reading Phonemic
value Spelling
Alternatives
א
א
Alef
Ɂ/
Ø
ע
ה
א
ב
ב
Bet
/b/
ב
(/v/)
ו
ב
ג
ג
Gimel
/g/
ד
ד
Dalet
/d/
ה
ה
He
h/
Ɂ/
Ø
ו
ו
Vav
/v/
ז
ז
Zayin
/z/
ח
ח
Xet
/x/
כ
ח
ט
ט
Tet
/t/
ט
ת
י
י
Yod
/j/
ך
כ
(
)
ך
Kaf
/k/
כ
כ
ק
ל
ל
Lamed
/l/
מ
מ
)
ם
(
Mem
/m/
ן
נ
(
ן
)
Nun
/n/
ס
ס
Samex
/s/
ש
ס
ע
ע
Ayin
/
Ɂ/
Ø
פ
פ
)
ף
(
Pe
/p/ /f/
פ
צ
צ
)
ץ
(
Tsadi
/
ʦ /
ק
ק
Kof
/k/
ר
ר
Resh
/
ʁ /
ש
ש
Shin/Sin
/
ʃ / /s/
ת
ת
Tav
/t/
Note
. The final column labeled Spelling Alternatives lists the letters sharing the same phonemic
value.
286
End Table 1
End Table 1End Table 1
End Table 1
2.2.2
2.2.22.2.2
2.2.2
Orthography
OrthographyOrthography
Orthography
The strict adherence to the consonantal principle derives from the consonantal
nature of the roots that provide the core meaning of a Hebrew content word. The
orthography highlights the root consonants by transforming the phonologically
discontinuous spoken root into an orthographically
continuous
unit which is easily
unitized as an integral (linear) lexical-orthographic representation. Consider the
following derivations from the common root KTV denoting the concept of writing.
The listener hears /katav/ (‘he wrote’), /jixtov/ (‘he will write’), /ktuba/
(‘marriage contract’), and etcetera. These words often have little in common
(/jixtov/ and /ktuba/ share only a single phoneme) but appear in print as follows,
כתוב
הבתכ
יתכוב
; the three (bolded) root letters appearing as a continuous or near-
continuous string; only the relatively unobtrusive VAV
ו
and YOD
י
are interpolated
between the root letters. Thus, as in the case of English morphemic transparency
(e.g.,
soft/soften
,
two/twice
), Hebrew provides a fine example of morpheme
constancy (Rogers, 1995), consistently marking phonologically distinct allomorphs.
Owing to its synthetic nature, Hebrew words are highly dense
morphemically. Not only are tense, person, number, etc., usually indicated by
inflecting roots, but many function words (
to
,
from
,
the
, etc.) and possessives (
my
,
his
) are frequently affixed to both nouns and pronouns. For example, the first
(rightmost) six-syllable word on the second line of Figure 1 (
םהיתויוכזבו
) translates
into four English words ‘and by their rights’. As a consequence, Hebrew texts are
significantly shorter than the same text in English. This demands considerable
287
morphemic unpacking by the reader and creates an additional source of
homography.
Owing to its distinctive morphology, full representation of vowels in Hebrew,
at least for the skilled reader, may well be unnecessary to distinguish basic
morphemic contrasts. Compare words derived from the root SPR (root letters again
bolded); SiPeR (
סירפ
) ‘he told’, SiPuR
)
סיפור
(
‘story’, SuPaR
)
סורפ
(
‘narrated’, etcetera. In
English, words sharing the same consonants, but with different vowels are typically
unrelated morphemes (e.g.,
SPRee
,
SPooR
,
SuPeR
,
aSPiRe
, etc.). By adding fully-
fledged vowel letters, Greek, an Indo-European language like English, became the
first fully vocalized segmental script, that is, an alphabet. Not only is vowel
information less critical for the Hebrew speaker, but this information is far more
easily inferred in printed Hebrew owing to the small inventory of vowels and the
limited number of morpho-phonological patterns. Yiddish uses Hebrew script
alphabetically, with full and equal representation of vowels. Ancient Egyptian, a
Semitic language, experimented briefly with alphabetization at one point, before
abandoning the idea (Gnanadesikan, 2009)
Unlike letter-to-phoneme correspondence, however, phoneme-to-letter
relations are not one-to-one with six pairs of (once phonemically distinct)
graphemes each representing the same phoneme (see Table 1, column headed
Spelling)
). The vast majority of Hebrew words, therefore, contain consonants and
vowels which could be spelled with alternate letters.
288
2.3
2.32.3
2.3
Conclusion
ConclusionConclusion
Conclusion
The bi-scriptal (pointed/unpointed) Hebrew abjad seems ideally suited to
preserving and highlighting root morphemes (essential for the skilled reader) as
well as providing phonological transparency (via the addition of pointing) for the
beginning reader. From a writing systems perspective, the common appellation
alphabet
is simply incorrect because this typically implies representation of
all
phonemes—both consonants and vowels, thereby helping legitimize the mistaken
notion of unpointed Hebrew as a “defective” script on the misguided assumption
that full and equal (European-style) representation of vowels and consonants is
necessarily superior to a script (i.e., Semitic abjad) that “fails” to represent vowels
systematically. This is far from a mere debate over nomenclature, and has far-
reaching implications for reading and writing assessment, literacy instruction, and
even definitions of dyslexia.
Acquisition of reading and spelling in Hebrew
Acquisition of reading and spelling in HebrewAcquisition of reading and spelling in Hebrew
Acquisition of reading and spelling in Hebrew
3.1
3.13.1
3.1
Becoming linguistically aware
Becoming linguistically awareBecoming linguistically aware
Becoming linguistically aware
3.1.1
3.1.13.1.1
3.1.1
Phonological development and phonological awareness
Phonological development and phonological awarenessPhonological development and phonological awareness
Phonological development and phonological awareness
The small number of longitudinal/predictive and training studies of phonological
awareness in Hebrew reviewed by S&L99 largely converged with the English
language findings: Access to sub-lexical phonological units
including
phonemes is a
significant predictor of later reading and, when trained, has a significant and
289
durable impact on later reading ability. Over the last decade, this conclusion has
been adopted by the Israeli Ministry of Education and incorporated into the Israeli
curriculum both in pre-school and in grade school. More recent work has explored
phonological awareness in the upper-elementary grades, and also the question of
precisely
which
phonological units are the crucial ones in Hebrew. Related to this, is
the growing interest in the question of awareness of vowels versus consonants.
Several cross-linguistic studies by Tolchinsky, Levin, and colleagues provide a
valuable backdrop for understanding the development of phonological awareness in
the Hebrew abjad as opposed to alphabetic and non-segmental writing systems.
Evidence is accumulating that phonemes are not the only sub-lexical unit
important for Hebrew, and, furthermore, among phonemes, awareness of
consonants (particularly syllable-final consonants) may be more important than
awareness of vowels. Several studies across a range of ages from pre-schoolers to
adults have shown that CV (body) units (
not
rimes, as in English) are more
accessible than isolated phonemes among native Hebrew speakers (Ben-Dror et al,
1995; Bentin et al., 1990; Share & Blum, 2005). In addition, access to single
(consonant) phonemes has been found to shift from an early pre-literacy advantage
for initial phonemes to a literacy-engendered preference for final phonemes (codas)
(Saiegh-Haddad, 2007; Share & Blum, 2005). Bar-Kochva (2013) recently reported
that final consonant isolation tended to be more strongly correlated with first grade
word identification than initial phoneme isolation. CV deletion also correlated well
with Grade 1 reading reaffirming that both consonantal phonemes
and
sub-syllabic
(CV) bodies are important for Hebrew reading.
290
Several studies have now shown that young children more easily identify and
isolate consonants than vowels (Lapidot et al., 1995–1996; Tolchinsky & Teberosky,
1998). In a study of phonological awareness in 5-year-old native speakers of
Spanish, Hebrew, and Cantonese, Tolchinsky et al. (2011) found no significant
differences between the three languages in initial phoneme (consonantal) isolation,
but a significant advantage emerged in final phoneme isolation in Hebrew. Both final
(but not initial) phoneme isolation and CV isolation correlated well with Hebrew
word writing and reading. It seems that syllable-
initial
consonant awareness may be
less critical than awareness of syllable-final consonants which are not bound to a
vowel in a CV conglomerate. A CV response was frequently supplied in the initial
phoneme isolation task, but rarely a VC (rime) in the final phoneme isolation task. In
learning to read Hebrew, it appears that individual consonants (particularly final
consonants)
and
CV units are the critical elements of phonology.
Recently, a number of cross-linguistic studies have questioned whether the
PA-reading is equally strong in deep and shallow alphabets. S&L99 cited several
studies reporting weaker correlations between end-of-Grade 1 Hebrew reading and
phonological awareness compared to English but they also noted that Bentin and
Leshem (1993) reported a strong correlation between Hebrew reading and PA in
the
middle
of Grade 1 when most children are still learning basic letter-sound
correspondences. Share (2008) went on to suggest that, at this point in
development, decoding skill may be more comparable to those of English speakers
later in the year. This led to the
functional opacity hypothesis
(Share, 2008),
according to which the PA-reading association is strongest when the spelling-sound
code is opaque owing either to intrinsic spelling-sound irregularities in the
291
orthography or to incomplete mastery of a regular code. This implies that the
intrinsic relationship between phonological awareness and reading may be equally
strong in both languages, the only difference being one of timing and duration.
Several recent Hebrew studies of phonological awareness in older upper-
elementary children have reported data bearing on this issue (Bar-Kochva, 2013;
Bar-Kochva & Breznitz, 2014; Cohen-Mimran, 2009; Katzir et al., 2012).
Unfortunately the findings conflict, probably owing, at least in part, to the use of
different measures and different sample ages. One possible interpretation is that PA
correlates with reading accuracy but not fluency or comprehension. This tentative
interpretation, if correct, would seem to refute Share’s
functional opacity hypothesis
because both Katzir et al. (2012) and Bar-Kochva & Breznitz (2014) found
substantial correlations in Grade 4 between reading accuracy in
pointed
(highly
regular) and not just unpointed text. However, the declining knowledge of vowel
diacritics after Grade 1 (see next section), essentially re-instates a level of functional
opacity (for vowels and occasionally consonants too) that places new demands on
phoneme awareness among these older readers.
Summing up, research undertaken over the past 15 years has confirmed the
importance of phonological awareness in learning to read Hebrew. It appears that
both individual phonemes (particularly final consonants)
and
CV bodies are the
critical phonological elements in learning Hebrew. The hypothesis that the PA-
reading link is strong only in the presence of decoding ambiguity may be sustained
if a U-shaped developmental curve is assumed: During initial learning of consonant
and vowel correspondences, decoding ambiguity is high, declining by the end of
Grade 1 when the code is mastered, then re-emerging again later in Grade 2 (and
292
beyond) when knowledge of the vowel diacritics diminishes thereby re-instating
functional opacity for vowels. The role of CV bodies is now well-established. The
integral CV syllable block has the advantage of obviating the well-known problems
of co-articulating consonants and vowels within a syllable, that is, blending
(Feitelson, 1988) thereby reducing working memory demands (see also Abu-
Ahmad, 2015). It appears that whereas initial consonants are intimately linked to
the following vowel in an indivisible CV unit, access to final singleton phonemes is
important because these phonemes still need to be integrated into closed (CVC)
syllables. Future research will need to track the development of vowel and
consonant awareness across the early years and examine its role in both silent and
oral reading of pointed and unpointed text.
3.1.2
3.1.23.1.2
3.1.2
Morphological development and morphological awareness
Morphological development and morphological awarenessMorphological development and morphological awareness
Morphological development and morphological awareness
Given the central role of morphology in spoken Hebrew (Berman, 1985; Ravid,
2001) and especially the unique Semitic root-and-pattern morphology, it is not
surprising that the association between morphological awareness and early reading
and spelling has received a good deal of attention in the reading research literature.
The earlier pre-S&L99 work focused on the general question of whether
morphological awareness/knowledge is related to reading. The answer then, as
now, is a resounding yes. Post-S&L99 studies have almost unanimously reported
significant and unique contributions of morphological awareness to reading and
spelling in pointed/unpointed, silent/oral, and older/younger samples (e.g., Bar-
Kochva, 2013, Bar-Kochva & Breznitz, 2014; Cohen-Mimran, 2009; Gur, 2005; Ravid
293
& Mashraki, 2007; Schiff et al., 2011). Indeed, some studies have found evidence that
morphological awareness in Hebrew has a stronger association with reading than
phonological awareness. Today, measures of morphology are now included as
standard practice in all investigations of Hebrew literacy development. The task
ahead will be to fill in the finer details—
which
aspects of morphology are related to
which
aspects of reading and spelling and
when
. To date, different measures of
morphology (and often reading too) have typically been used interchangeably.
3.2
3.23.2
3.2
Development of word identification
Development of word identificationDevelopment of word identification
Development of word identification
3.2.1
3.2.13.2.1
3.2.1
Word decoding development
Word decoding developmentWord decoding development
Word decoding development
The rapid early mastery of Hebrew’s phonologically transparent pointed script,
discussed in S&L99, has again been confirmed in numerous studies conducted in
recent years, notably Shany, Bar-On, and Katzir’s (2012) definitive study of a large
nationally representative sample.
S&L99 reviewed a number of studies indicating that vowel diacritics
facilitated oral reading speed and accuracy among young readers, mainly by
reducing vowel errors. More recent work has taken a more genuinely
developmental approach posing the question of how pointing influences reading at
different
developmental points.
Gur’s (2005) longitudinal study found that the advantage of oral text reading
accuracy for pointed (as opposed to unpointed) script in Grade 1 was substantial
(pointed 94%, unpointed 85%), narrowing in Grade 2 (95% – 92%), before
disappearing by the end of Grade 3 (96%, 95%). This trend has since been
294
replicated for accuracy and extended to reading rate (Bar-Kochva & Breznitz, 2014;
Schiff, Katzir & Shoshan, 2013). These data reaffirm that pointed word reading is
mastered by the end of first grade and maintained thereafter at consistently high
levels. The less familiar unpointed script seems to take some getting used to
initially, but is eventually mastered by late Grade 3. What is typically labeled an
“advantage” of pointing (for speed and accuracy) might perhaps better be
characterized as developmental “catch-up” with the initial “disadvantage” for
(unfamiliar) unpointed text overcome by Grade 3. Simply asking whether diacritics
help or hinder may be missing the fuller developmental picture.
The direct investigation of the diacritical system per se has recently emerged
as a topic worthy of study its own right, rather than merely as a tool for addressing
the classic debate on the role of phonology in word reading.
3
Gur (2005) examined
the issue of vowel diacritics in a longitudinal sample followed from Grade 1 to Grade
3 at a single school. Children read aloud vowel diacritics under seven conditions;
first in isolation, then in a variety of morphological, lexical, and sentential contexts
(in additive fashion).
Overall knowledge of diacritics presented in print actually
diminished
over
the three grades (Share & Gur, in preparation). The same trend was reported by
3
In what appears to be a case of “reverse-Anglocentrism”, it is important to
recognize that the difference between pointed and unpointed Hebrew is primarily
vowel phonology, and accordingly plays a very different linguistic role to the
“phonology” tapped in the classic regular/exception word contrast in English. The
unspoken assumption underlying numerous Hebrew studies that the role of
“phonology” in word identification, as gauged by within-item pointed/unpointed
comparisons, is functionally equivalent to the classic regular/exception word
contrast in English may be ill-founded.
295
Shany et al. (2012) in their large nationally representative (cross-sectional) sample.
This is a startling finding when one considers the fact that steadily
improving
knowledge of the spelling-sound code as witnessed in increasing letter-sound
knowledge and pseudoword reading accuracy (among normally developing
populations at least) is an axiomatic indeed
unquestioned
finding in the English-
language literature. These data underscore the pedagogical role of Hebrew diacritics
in providing full phonological specification for the learner. Gradually, the diacritics
become less crucial as reading skill advances and the reader relies to a greater
extent on higher-order word-level lexical and morphological information.
Another unexpected finding emerged in both these investigations. Contrary
to popular belief, the easiest diacritic was not the ubiquitous
kamats
(
ָ
) /a/ which
accounts for some 25% of all vowel signs, but the
hiriq
(97%) /i/. Gur hypothesized
that the advantage of the
hiriq
is due to its spelling-sound consistency; it alone
among the vowel diacritics has a single phonemic value (/i/).
4
In addition to
consistency, Gur found frequency correlated strongly with reading accuracy but not
visual-spatial complexity (number and position of dots) (see also Bar-Kochva,
2013).
Gur also found that diacritical mastery correlated. 86 with pseudoword
reading in both Grades 1 and 2, showing unambiguously that pseudoword reading
in the Hebrew abjad depends largely on vowel diacritics rather than consonants.
4
The
kamats
has two inconsistent values. One is the so-called
kamats katan
which appears in a single but very high frequency item: /kol/ ‘all’. The second is the
hataf kamats
/o/ which
appears in a few dozen fairly common nouns. It appears that
even a very high frequency diacritic, if not entirely consistent, creates ambiguity for
the beginning reader
296
Consistent with the increasingly role of lexical-morpho-orthographic processing in
developing readers, the correlation between word decoding and vowel diacritic
mastery diminished (to.61) by Grade 3. The decreasing reliance on phonology and
growing lexical-morphological influences is also seen in the fact that two measures
of morphological awareness were uncorrelated in Grade 1 with any of the CV
reading tasks, but did show significant moderate correlations by Grade 3.
The S&L99 review included several studies attesting to a growing influence
of lexico-morphological knowledge across age. For example, reading accuracy
increases across age for pseudowords that conform to a familiar morpho-
phonological pattern as opposed to “unpatterned” pseudowords, as well as the
tendency to misread incorrectly pointed items in accordance with the correct
vocalization pattern. But before surveying more recent work, it is important to
recap Ravid’s (1996) groundbreaking study examining the oral reading of highly
literate forms that are unfamiliar to young children. First graders read these forms
more accurately than fourth grade readers, suggesting that novice readers rely
mostly on bottom-up serial spelling-sound (assembled) decoding, whereas the older
more skilled readers rely more on lexical-orthographic (addressed) phonology
disregarding the vowel diacritics and generating the more common spoken forms.
This general developmental progression has been confirmed in the newer studies.
Gur (2005) compared the reading of morphologically
unpatterned
pseudowords that were phonologically and orthographically legal at the syllable
(CV) level, but did not conform to a Hebrew morpho-phonological pattern with a
matched list consisting of
patterned
pseudowords that conformed to a familiar
297
morpho-phonological pattern. A list of real words with the identical morpho-
phonological patterns to the patterned pseudoword list was also presented.
In Grade 1, children’s pronunciation accuracy was unaffected by either
morphological patterning (i.e., well-formedness) or even word-specific lexical
support (real words), but by Grade 2 there was a clear differentiation manifest
mainly as declining accuracy for unpatterned pseudowords together with lexical
benefits for real words (i.e., word superiority effect). This pattern of findings was
reproduced in the reaction time data. Once again, (and confirming the earlier study
by Ravid (1996), pure (“unassisted”) grapheme-phoneme decoding actually
declines after Grade 1 hand in hand with growing sensitivity to higher-order (word-
level) morphological and lexical constraints.
Turning to unpointed orthography, the two major challenges for the
developing reader are the pronunciation of unpointed words and the rampant
homography. Only a single study in Hebrew was published at the time of the earlier
S&L99 review. Bentin, Deutsch, and Liberman (1990) identified young disabled
readers who experienced selective difficulties with unpointed but not pointed text.
This group were characterized by difficulties in the domain of syntax.
More recently, Bar-On and Ravid (2011) investigated the role of morphology
in the reading of (non-homographic) unpointed words across a wide range of ages.
Participants read aloud unpointed pseudowords constructed from a pseudoroot and
a genuine pattern (e.g.,
ןסלכ
KaLSan).
Responses were scored as either a true Hebrew
morpho-phonological pattern (KaLSan), as unpatterned responses that constituted
a permissible grapho-phonemic decoding (at the syllable level) but did not conform
298
to any Hebrew morpho-phonological pattern (e.g., KaLeSan), or basic grapho-
phonemic errors (KaLMaN).
In early Grade 2, unpatterned responses predominated (50%), followed by
grapho-phonemic errors (20%) with 30% correct pattern responses. Thereafter, the
trend shifted to mostly correct pattern identification, with around one third
unpatterned responses, and grapho-phonemic errors extremely rare. The authors
labeled this a transition from phonological to morphological reading, and described
the change from early Grade 2 to late Grade 2 as a “watershed” period in reading
acquisition (cf. Share & Shalev, 2004). The authors concluded that morphology is
the key to determining lexical identity in unpointed script, eliminating multiple
alternative pronunciations.
Bar-On and Ravid’s second study addressed the issue of homography. They
presented the homographic pseudoword
ףגשמ
which can be assigned four different
pronunciations depending on context. The most common forms – present tense verb
and a noun, were correctly identified by the end of Grade 2, but the more complex
and lower frequency alternatives were only mastered around Grade 7.
In a follow-up study, Bar-On (in press) showed that extensive homography
poses little problem for the reader of unpointed text (indeed skilled readers were
found to be largely oblivious to these ambiguities) because almost all alternatives
were found to possess distinct morpho-syntactic features such as nouns versus
verbs (consider
wind/wind
in English). Bar-On identified several sources of
homography including the interwoven (non-linear) root-plus-pattern structure of
most content words, and linear affixation. Bar-On concludes with the insightful
observation that the context-dependent nature of printed words in standard
299
unpointed Hebrew text raises questions about the role of models and assessment
practices founded on isolated word recognition.
Consistent with the context-dependent nature of unpointed reading, several
correlational studies have demonstrated that semantic and syntactic measures
contribute selectively to unpointed but not pointed reading (Shatil & Share, 2003),
whereas morphological knowledge is important for both scripts (Cohen-Mimran,
2009).
3.2.2
3.2.23.2.2
3.2.2
Word spelling development
Word spelling developmentWord spelling development
Word spelling development
The pioneering work of Levin, Tolchinsky, and colleagues on pre-school
writing/spelling aimed at tracing the developmental stages characterizing children’s
early attempts at producing written language prior to school entry. More recent
research has extended this work by delving into the social/environmental factors
(e.g., maternal mediation) that help or hinder growth in early writing (e.g., Aram &
Levin, 2004; Korat, 2005; Schiff & Lotem, 2011) as well as cross-linguistic
comparisons between Hebrew and other orthographies (e.g., Levin et al., 2013;
Tolchinsky et al., 2011). A new strand of research led by Ravid and Schiff has
focused on the role of morphology among school-aged children thereby providing a
more complete picture of writing development both before and after school entry.
A series of studies summarized by Levin (1997) documented the early
emergence of both the universal properties of writing systems and Hebrew-specific
features. This work culminated in a developmental scale of Hebrew writing (Levin,
300
Korat, & Amsterdamer, 1996; Shatil, Share & Levin, 2000) which has proven popular
with researchers and educators alike.
In all these investigations, children’s representation of consonants has been
found to be more advanced than that of vowels. Newer developmental (e.g., Ravid &
Kubi, 2003) and cross-linguistic work (e.g., Tolchinsky et al., 2012) has confirmed
the advantage of consonants over vowels.
A new generation of research into spelling in school has also addressed the
biggest challenge in Hebrew spelling—producing orthographically correct
(conventional) spellings given the many options for transcribing so many phonemes
(see Table 1).
In an analysis of naturally occurring spelling errors in compositions
produced by children from Grades 1 to 6, Ravid (2001) compared content words
with grammatical words, and,
within
content words, root letters versus non-root
letters. As anticipated, Ravid (2001) found that errors in content words far
outnumbered errors in grammatical words with content word errors high in Grades
1 (29%) and Grade 2 (20%) declining steeply thereafter. Grammatical errors were
rare after Grade 1 (19%). Letter-level analyses of errors within content words
replicated this pattern.
In a follow-up study in Grades 2, 3, and 4, Ravid dictated a set of words each
containing root and function letters. Again, root letter errors for homophonous
letters were far more numerous than function letter errors. This finding has since
been replicated and extended by Ravid and Bar-On (2005) and Gillis and Ravid
(2006).
301
3.2.3
3.2.33.2.3
3.2.3
Reading and spelling difficulties
Reading and spelling difficultiesReading and spelling difficulties
Reading and spelling difficulties
The defining feature of reading difficulty in Hebrew is not reading accuracy (as in
English) but reading speed. Hebrew dyslexics are typically found to read at half the
rate of age-matched normal readers (see, e.g., Breznitz, 1997). Phonological
awareness deficits are also characteristic of poor readers in Hebrew, but
morphological deficits seem to be equally important. Syntax and semantics are not
crucial given Hebrew’s phonologically transparent script, but are essential for
dealing with ambiguity in unpointed text.
The influential English-language subtyping taxonomy spawned by the dual
route model (phonological versus surface dyslexics) may have limited applicability
to Hebrew. Recently Shany and colleagues (e.g., Shany & Share, 2011; Shany &
Breznitz, 2011) have proposed a promising subtyping scheme based on rate versus
accuracy, adducing evidence for true double dissociations between rate-disabled
and accuracy-disabled readers.
Spelling difficulties in Hebrew are rarely manifest in phonologically
implausible spellings (see, e.g., Lamm & Epstein, 1994) but in orthographically
incorrect spellings involving the many homophonic letters (Ravid, 2012).
3.3
3.33.3
3.3
Reading Compreh
Reading ComprehReading Compreh
Reading Comprehension
ensionension
ension
3.3.1
3.3.13.3.1
3.3.1
Predictors of reading comprehension
Predictors of reading comprehensionPredictors of reading comprehension
Predictors of reading comprehension
Only a few studies have addressed this topic in Hebrew. In a large-scale longitudinal
study, Shatil and Share (2003) examined the contribution of a host of Kindergarten
302
predictors to first grade decoding and reading comprehension (of pointed text).
Predictors included both “domain-specific” variables (visual-orthographic
processing, phonological awareness, phonological memory, and early literacy) and
“domain-general” (general intelligence, metacognitive functioning, and oral
language). Whereas domain-general measures (which included oral vocabulary and
syntax) explained very little of the variance in Grade 1 decoding (a speed and
accuracy composite), both domain-specific and domain-general blocks each
contributed significant and substantial portions of variance to the prediction of
Grade 1 reading comprehension.
In the research thus far, morphological awareness has repeatedly been found
to explain significant and substantial variance in reading comprehension
throughout the elementary years. In a longitudinal study, Bar-Kochva (2013) found
that morphological awareness and phonological working memory were the best
Kindergarten predictors of Grade 1 (pointed) reading comprehension, followed by
vocabulary and phonological awareness. RAN correlated well with fluency measures
but somewhat less with comprehension. Visual-spatial measures using non-
symbolic material were uncorrelated with any reading measure.
In a second longitudinal study, Bar-Kochva and Breznitz (2014) compared
the predictors of pointed and unpointed comprehension in Grades 3 and 4. Pointed
comprehension in Grade 3 was most strongly predicted by phonological awareness
(.4), and also, but more weakly, by Raven, Peabody vocabulary, morphology, RAN,
and digit-span (all around .3), but again uncorrelated with visual processing speed.
Interestingly, neither pseudoword reading accuracy nor speed was correlated with
comprehension (reinforcing early comments about the questionable relevance of
303
vowel diacritics beyond first grade. The picture was largely unaltered for unpointed
comprehension except that Raven and Peabody were more strongly correlated than
with pointed suggesting that unpointed text depends more on higher-order
reasoning.
In Grade 4, the results for pointed comprehension were roughly similar,
although this time Raven did not correlate more strongly with unpointed
comprehension and phonological awareness was a little weaker. Pseudoword
reading accuracy in this grade correlated significantly (but very weakly) with
unpointed comprehension. On the whole, therefore, the predictors of pointed and
unpointed comprehension were fairly similar, consistent with the idea that the
common consonantal infrastructure of both versions of the orthography is the core
of reading in both orthographies.
3.3.2
3.3.23.3.2
3.3.2
Word level
Word levelWord level
Word level
effects in comprehending text
effects in comprehending texteffects in comprehending text
effects in comprehending text
A small number of studies have focused on word level effects in comprehending test.
Ravid and Mashraki (2007) found that an aggregated score on morphology
(combining five separate morphology measures) correlated. 64 with fourth grade
reading comprehension. It also correlated well with prosodic reading fluency but
made a unique contribution to comprehension even after partialling out prosodic
fluency. Word-level comprehension questions correlated very highly with higher-
order comprehension (.81) indicating that lexical/morphological processing is
central to higher-order reading comprehension, consistent with Perfetti’s “lexical
quality” hypothesis (Perfetti, 2007; Perfetti & Hart, 2002).
304
Schiff, Schwartz-Nahshon and Nagar (2011) also found moderate to strong
correlations between morphological awareness and reading comprehension in
Grades 3 and 7. Furthermore, the contribution of morphology remained significant
after controlling both word decoding and phonological awareness.
3.4
3.43.4
3.4
Conclusion
ConclusionConclusion
Conclusion
Learning to read Hebrew, like all segmental orthographies, depends on phonological
awareness, but one in which the critical phonological units only partly overlap with
European alphabets; the supra-phonemic subsyllabic CV unit, and, at the phoneme
level, consonants more than vowels. Morphology has also been shown to be crucial
in Hebrew for word-level, sentence-level, and text-level reading. The initial pointed
and highly phonologically transparent script, like other transparent scripts, is
acquired rapidly. Interestingly, the challenges of learning the diacritic-like vowel
system appear to stem not from its visual-spatial complexity but from phonological
inconsistency. The “all-phonological” and sequential nature of beginning reading
gradually shifts in Grade 2 to a more rapid, more parallel lexico-morphological-
orthographic mode of word identification that lays the foundations for faster more
fluent reading. It also provides the platform for a successful transition to unpointed
text (with its rampant homography) which depends heavily on syntactic and
semantic-pragmatic contextual support that would not be possible without a certain
level of word reading fluency (Perfetti, 1985).
305
Discussion
DiscussionDiscussion
Discussion
4.1
4.14.1
4.1
Challenges in learning to read Hebrew
Challenges in learning to read HebrewChallenges in learning to read Hebrew
Challenges in learning to read Hebrew
A wide-angled overview of Hebrew reading acquisition reveals important
commonalities with alphabetic reading acquisition—this follows from the fact that
both writing systems are primarily segmental. Close up, at a finer grain, there are
fundamental differences; this too follows from the fact that one is alphabetic, the
other an abjad. Both perspectives, the wide-angled lens and the close-up, provide
valuable insights.
The wide-angled view reveals important universals (at least for segmental
writing systems) including the indispensability of phonological awareness and a
developmental shift from small-unit sequential phonological recoding to a higher-
order or larger-unit morpheme-level or morpho-orthographic mode of processing
which is faster and more efficient. One way to characterize this general shift without
making commitments to specifics such as unit size or timing issues is the
unfamiliar-
to-familiar/novice-to-expert
dualism (Share, 2008). This is a shift, common to all
skill learning, from slow, effortful, piece-by-piece unskilled performance to rapid
one-step large-unit skilled performance. This dualism also converges with the
dualistic nature of an efficient orthography which must serve the needs of the
novice for
decipherability
(via phonological transparency) as well as the expert’s for
unitizability
and
automatizability
(via morphemic transparency). Hebrew performs
this balancing act with aplomb, employing diacritics to supply the beginner with a
consistent, phonologically well-specified script, while helping the expert-to-be
306
unitize the basic meaning-bearing units (roots) of the orthography by transforming
abstract, discontinuous spoken roots into concrete, continuous and therefore
unitizable orthographic units. However, a dualistic approach focused exclusively at
the level of individual words falls short of providing a complete picture of Hebrew
reading acquisition: There appear to be three rather than two major aspects of
learning to read Hebrew. These can be broadly characterized as a cumulative
progression from lower-order phonological (
sub-lexical
) sequential spelling-to-
sound translation through higher-order string-level (
intra-lexical
) lexico-
morphological processing up to a
supra-lexical
contextual level essential for
unpointed script with its pervasive homography. Each aspect or mode of processing
adds to, not replaces the previous one.
4.2
4.24.2
4.2
Implications for instruction
Implications for instructionImplications for instruction
Implications for instruction
In order to draw out the instructional implications of this three-phase model of
reading development, a closer examination of the three modes is required. First, the
sub-lexical phonological recoding level of the beginner is
not
the separate segments
(vowels and consonants) of an alphabet, but primarily CV syllables and syllable-final
consonants. Vowels appear to play only a subordinate role, if any. This initial entry-
level spelling-sound translation is a relatively “pure”, almost exclusively low-level
decoding, akin to the bottom-up processes characteristic of regular alphabetic
orthographies. But unlike alphabets, blending is completely bypassed in the case of
open syllables, although, for closed syllables, final consonants may have to be
appended to CV units. Direct observational studies of beginner’s decoding will be
307
needed to confirm this picture and determine how it varies across instructional
practices. Note that even the common term “grapheme-phoneme” is theoretically
loaded; there are no digraphs in Hebrew hence no need for a term other than
“letter”, and the basic graphic units may not only be phonemes, but multi-phonemic
CV units as already noted.
The second phase of this progression also brings to light fundamental
differences between the Hebrew abjad and European alphabets with major
implications for teaching. In the course of the second grade, the “watershed” year of
reading development, the Hebrew reader shifts gear to a faster, more efficient word-
level or morpheme-level lexico-morphological-orthographic reading strategy based
on larger, higher-order units that primarily reflect consonantal roots and morpho-
phonological patterns. The highly productive word-formation structures in Hebrew,
both non-linear and linear, underpin reading at this point even though children are
still reading mostly pointed script. The term “lexical” comes to mind here but, if
conceptualized in the classic dual-route sense, is problematic for Hebrew because
most words, even the early lexicon of beginning readers, are not mono-morphemic
but multi-morphemic, representing the intersection of two abstract entities, a root
and pattern (each discontinuous and unpronounceable). Although the result is a
specific pronounceable unit, the dozens of content word inflections and affixes raise
doubts about the utility of the notion of an orthographic lexicon (as conceptualized
in the English-language literature) as a list or directory of discrete self-contained
(“encapsulated” or “autonomous”) word-specific entries. Rather than the minimally
inflected, hence largely constant string of letters in an English word, most Hebrew
308
words (if that term can be applied) are far less “stable” but more chameleon-like
entities or families of overlapping resemblances (recall yiXToV, KTuBa).
The most startling and “unalphabetic” feature of this second phase of reading
development, and one with profound implications not only for instruction but also
assessment, is a decline in knowledge and use of vowel signs (as witnessed in
deteriorating pseudoword reading accuracy), a finding that stands in complete
opposition to a universal almost axiomatic feature of normal alphabetic reading
development. At this point, the role of morphology in Hebrew emerges again and
again as crucial, even outweighing phonological awareness in some studies.
The transition to unpointed Hebrew (around Grades 3 and 4) with its
pervasive homography, again highlights glaring alphabet/abjad differences with far-
reaching implications not only for instruction, assessment and even the definition of
reading disability. In Hebrew, the significance of supra-lexical contextual
information sets it well apart from English and even more so from regular European
alphabets. Given the ubiquity of homographs in unpointed text, reliance on
contextual information appears to be both central and indispensable (even among
highly skilled readers, see, e.g., Benuck & Peverly, 2004). By contrast, the
developmental literature in English describes the marginal and
declining
role of
context that accompanies growing word recognition skill (Stanovich, 1986; 2000),
with increasing word reading modularity made possible by self-contained written
forms (spellings). On a note of concord, it is interesting to observe that the
diminishing role of vowel diacritics in Hebrew and extra-lexical context in English
(particularly in the case of irregular words, Ricketts, Nation, & Bishop, 2007;
309
Tunmer & Chapman, 2012) both appear to serve a similar assistive function for
developing readers.
Final conclusion
Final conclusionFinal conclusion
Final conclusion
The study of Hebrew reading acquisition offers valuable insights into one of the
fundamental dilemmas of all writing systems; the trade-off between the needs of the
novice versus the expert (Share, 2008). This is the antinomy between the
decipherability
needed by learners encountering massive numbers of unfamiliar
words, as opposed to the expert-to-be’s need for
automatizability
and
unitizability
,
that is, between phonological transparency and morphemic transparency. The
ingenious bi-scriptal solution achieved by means of optional diacritics provides both
a phonologically well-specified script to assist the novice (so-called
pointed
Hebrew), and later, by shedding its vowel diacritics (and pedogogical intent), a
phonologically under-specified script (
unpointed
Hebrew) emphasizing
morphology.
The study of Hebrew literacy acquisition also widens the researcher’s angle
of vision on many issues on the global research agenda including the phonological
awareness-reading connection in a non-alphabetic orthography, the transition from
a phonologically well-specified orthography to an under-specified one, pervasive
homography, morphemic analysis of poly-morphemic strings, the processing of
diacritics, and the classic problems learning to spell a writing system that, like
English, preserves many now-neutralized consonant and vowel distinctions in
spelling.
310
Above all, the study of Hebrew reading offers all reading researchers some
refreshing insights into literacy learning outside the familiar box of European
alphabets, one that provides a wider angle of vision on all orthographies.
chapter
chapterchapter
chapter-
--
-references
referencesreferences
references
References
ReferencesReferences
References
Abu-Ahmad, H. (2015).
Learning to read Arabic syllabically versus alphabetically:
psycholinguistics, grammatological and pedagogical aspects
(Unpublished
doctoral dissertation). University of Haifa, Israel.
Aram, D. (2005). Continuity in children’s literacy achievements: A longitudinal
perspective from kindergarten to school.
First Language, 3
, 259–289.
Aram, D., & Levin, I. (2004). The role of maternal mediation of writing to
kindergartners in promoting literacy achievements in second grade: A
longitudinal perspective.
Reading and Writing, 17
, 387–409.
Bar-Kochva, I. (2013). What are the underlying skills of silent reading acquisition? A
developmental study from kindergarten to the 2nd grade.
Reading and
Writing, 26
, 1417–1436.
Bar-Kochva, I., & Breznitz, Z. (2014). Reading scripts that differ in orthographic
transparency: A within-participant-and-language investigation of underlying
skills.
Journal of Experimental Child Psychology, 121
, 12–27.
311
Bar-On, A. (in press). [in Hebrew]
kriya betzel hahomografia: haba’aya hi gam
hapitaron
, Reading in the shadow of homography: The problem is the
solution.
Safa Ve’Oriyanut
.
Bar-On, A., & Ravid, D. (2011). Morphological analysis in learning to read
psseudowords in Hebrew.
Applied Psycholinguistics, 32
, 553–581.
Ben-Dror, I., Bentin, S., & Frost, R. (1995). Semantic, phonologic, and morphologic
skills in reading disabled and normal children: Evidence from perception and
production of spoken Hebrew.
Reading Research Quarterly, 30
, 876–893.
Bentin, S., Deutsch, A., & Liberman, I. Y. (1990). Syntactic competence and reading
ability in children.
Journal of Experimental Child Psychology, 48
, 147–172.
Bentin, S., & Leshem, H. (1993). On the interaction of phonologic awareness and
reading acquisition: It’s a two-way street.
Psychological Science, 2
, 271–274.
Benuck, M. B., & Peverly, S. T. (2004). The effect of orthographic depth on reliance
upon semantic context for oral reading in English and Hebrew.
Journal of
Research in Reading, 27
, 281–299.
Berman, R. (1985). Hebrew. In D. I. Slobin (Ed.),
The cross-linguistic study of
language acquisition
. Hillsdale, NJ: Erlbaum.
Berman, R. (1997). Modern Hebrew. In R. Hetzron (Ed.),
The semitic languages
(pp.
312–333). Abingdon, Oxon, United Kingdom: Routledge.
Berman, R. (2003). Children’s lexical innovations: Developmental perspectives on
Hebrew verb-structure. In J. Shimron (Ed.),
Language processing and
Comment [INT26]:
Unlinked
312
acquisition in languages of Semitic root-based morphology
(pp. 243–291).
Amsterdam, the Netherlands: Benjamins.
Beveridge, M. E. L., & Bak, T. H. (2012). Beyond one-way streets: The interaction of
phonology, morphology, and culture with orthography.
Behavioral and Brain
Sciences, 35
, 18–19.
Breznitz, Z. (1997). Effects of accelerated reading rate on memory for text among
dyslexic readers.
Journal of Educational Psychology
,
89
, 289–297.
Chomsky, W. (1941). The history of our vowel-system in Hebrew.
The Jewish
Quarterly Review, 32
, 27–49.
Cohen-Mimran, R. (2009). The contribution of language skills to reading fluency: A
comparison of two orthographies for Hebrew.
Journal of Child Language
,
36
,
657–672.
Daniels, P. T. (1990). Fundamentals of grammatology.
Journal of the American
Oriental Society, 110
, 727–731.
Daniels, P. T., & Bright, W. (Eds.) (1996).
The world’s writing systems
. New York,
NY: Oxford University Press.
Diringer, D. (1948).
The Alphabet: A Key to the History of Mankind
. London, United
Kingdom: Thames & Hudson.
Evans, N., & Levinson, S. C. (2009). The myth of language universals: Language
diversity and its importance for cognitive science.
Behavioral and Brain
Sciences, 32
, 429–492.
Feitelson, D. (1988).
Facts and fads in beginning reading
. New York, NY: Ablex.
313
Frost, R. (2012). Towards a universal model of reading.
Behavioral and Brain
Sciences, 35
, 263–279.
Frost, R., Siegelman, N., Narkiss, A., & Afek, L. (2013). What predicts successful
literacy acquisition in a second language?
Psychological Science, 24
, 1243–
1252.
Geva, E. (2008). Facets of metalinguistic awareness related to reading development
in Hebrew: Evidence from monolingual and bilingual children. In K. Koda &
A. Zehler (Eds.),
Learning to read across languages
(pp. 154–187). New York:
Routledge.
Gillis, S., & Ravid, D. (2006). Typological effects on spelling development: A
crosslinguistic study of Hebrew and Dutch.
Journal of Child Language, 33
,
621–659.
Glinert, L. H. (1994). Israeli Hebrew. In R. E. Asher & J. M. Y. Simpson (Eds.),
The
encyclopedia of language and linguistics
(pp. 1538–1541). Oxford, United
Kingdom: Pergamon Press.
Gnanadesikan, A. E. (2009).
The writing revolution: Cuneiform to the internet
.
Chichester, United Kingdom:Wiley-Blackwell.
Goerwitz, R. L. (1996). The Jewish scripts. In P. T. Daniels & W. Bright (Eds.),
The
world’s writing systems
(pp. 489–494). New York, NY: Oxford University
Press.
314
Gur, T. (2005).
Reading Hebrew vowel diacritics: A longitudinal investigation from
Grade 1 to Grade 3
(Unpublished doctoral dissertation). University of Haifa,
Israel.
Henrich, J., Heine, S. J., & Norenzayan, A. (2010). The weirdest people in the world?
Behavioural and Brain Sciences
,
33
, 61–135.
Hetzron, R. (1987). Hebrew. In B. Comrie (Ed.),
The world’s major languages
(pp.
578–593). London, United Kingdom: Croom Helm.
Hoberman, R. D., & Aronoff, M. (2003). The verb morphology of Maltese: From
Semitic to Romance. In J. Shimron (Ed.),
Language processing and language
acquisition in root-based Semitic, morphology
(pp. 61–78). Amsterdam, the
Netherlands: John Benjamin.
Katzir, T., Schiff, R., & Kim, Y.-S. (2012). The effects of orthographic consistency on
reading development: A within and between cross-linguistic study of fluency
and accuracy among fourth grade English- and Hebrew-speaking children.
Learning and Individual Differences
,
22
, 673–679
Korat, O. (2005). Contextual and non-contextual knowledge in emergent literacy
development: A comparison between children from low SES and middle SES
communities.
Early Childhood Research Quarterly, 20
, 220–238.
Lamm, O., & Epstein, R. (1994). Dichotic listening performance under high and low
lexical work load in subtypes of developmental dyslexia.
Neuropsychologia,
32
, 757–785.
315
Lapidot, M., Tubul, G., & Wohl, A. (1995–1996). [in Hebrew] Mivchan eranut
fonologit kekli nibui lerechishat hakri’a [A test of phonological awareness as
a predictor of reading acquisition].
Chelkat Lashon, 19–20
, 169–188.
Leikin, M., Schwartz, M., & Tobin, Y. (2011).
Current issues in bilingualism: Cognitive
and socio-linguistic perspectives
. Dordrecht, the Netherlands: Springer.
Levin I. (1997). [in Hebrew] Hitpatxut haktiva hamukdemet be’Ivrit [The early
development of writing in Hebrew]. In J. Shimron (Ed.),
Mexkarim
bepsixologia shel halashon beyisrael: rexishat lashon, kriya vektiva
[Studies
in the psychology of language in Israel: language acquisition, reading and
writing]. Jerusalem, Israel: Magnus.
Levin, I., Aram, D., Tolchinsky, L., & McBride, C. (2013). Maternal mediation of
writing and children’s early spelling and reading: The Semitic abjad versus
the European alphabet.
Writing Systems Research, 5
, 134–155.
Levin, I., Korat, O., & Amsterdamer, P. (1996). Emergent writing among
kindergartners: Cross-linguistic commonalities and Hebrew-specific issues.
In G. Rijlaarsdam, H. van der Bergh, & M. Couzijn (Eds.),
Current trends in
writing research: Theories, models and methodology
(pp. 398–419).
Amsterdam, the Netherlands: Amsterdam University Press.
McCarthy, J. J. (1981). A prosodic theory of nonconcatenative morphology.
Linguistic Inquiry
, 12, 373–418.
Miller, P. (2009). Learning in a world of silence: Insight from 30 years of work and
research with deaf and hard of hearing children. In D. Aram & O. Korat (Eds.),
316
Literacy and language: Interactions, bilingualism, and difficulties
. Jerusalem,
Israel: Magnes Press.
Naveh, J. (1982).
Early history of the alphabet: an introduction to West Semitic
epigraphy and palaeography
. Jerusalem, Israel: Magnes Press.
Perfetti, C. A. (1985).
Reading ability
. New York, NY: Oxford University Press.
Perfetti, C. A. (2007). Reading ability: Lexical quality to comprehension.
Scientific
Studies of Reading, 11
, 357–383.
Perfetti, C. A., & Hart, L. (2002). The lexical quality hypothesis. In L. Verhoeven, C.
Elbro, & P. Reitsma (Eds.),
Precursors of functional literacy
(pp. 189–213).
Amsterdam, the Netherlands: John Benjamins.
Ravid, D. (1990). Internal structure constraints on new-word formation devices in
Modern Hebrew.
Folia Linguistica, 24
, 289–346.
Ravid, D. (1996). Accessing the mental lexicon: Evidence from incompatibility
between representation of spoken and written morphology.
Linguistics, 34
,
1219–1246.
Ravid, D. (2001). Learning to spell in Hebrew: Phonological and morphological
factors.
Reading and Writing, 14
, 459–485.
Ravid, D. (2012).
Spelling morphology
. New York, NY: Springer.
Ravid, D., & Bar-On, A. (2005). Manipulating written Hebrew roots across
development: The interface of semantic, phonological, and orthographic
factors.
Reading and Writing, 18
, 231–256.
317
Ravid, D., & Kubi, E. (2003). What is a spelling error? The discrepancy between
perception and reality.
Faits de Langues, 22
, 87–98.
Ravid, D., & Mashraki, Y. E. (2007). Prosodic reading, reading comprehension and
morphological skills in Hebrew-speaking fourth graders.
Journal of Research
in Reading, 30
, 140–150.
Ravid, D., & Schiff, R. (2006). Roots and patterns in Hebrew language development:
Evidence from written morphological analogies.
Reading and Writing, 19
,
789–818.
Ricketts, J., Nation, K., & Bishop, D. V. M. (2007). Vocabulary is important for some,
but not all reading skills.
Scientific Studies of Reading, 11
, 235–257.
Rogers, H. (1995). Optimal orthographies. In I. Taylor & D. R. Olson (Eds.),
Scripts
and Literacy
(pp. 31–43). Dordrecht, the Netherlands: Kluwer Academic
Publishers.
Saiegh-Haddad, E. (2007). Epilinguistic and metalinguistic phonological awareness
may be subject to different constraints: Evidence from Hebrew.
First
Language, 27,
385–405.
Schiff, R., Katzir, T., & Shoshan, N. (2013) Reading accuracy and speed of vowelized
and unvowelized scripts among dyslexic readers of Hebrew: The road not
taken.
Annals of Dyslexia, 63
, 171–185.
Schiff, R., & Lotem, M. (2011). Effects of phonological and morphological awareness
on children’s word reading development from two socioeconomic
backgrounds.
First Language, 31
, 139–163
318
Schiff, R., Schwartz-Nahshon, S., & Nagar, R. (2011). Effect of phonological and
morphological awareness on reading comprehension in Hebrew-speaking
adolescents with reading disabilities.
Annals of Dyslexia, 61
, 44–63.
Schwarzwald, O. R. (2002).
Modern Hebrew morphology
[in Hebrew]. Tel Aviv,
Israel: Open University.
Shany, M., Bar-On, A., & Katzir, T. (2012). Reading different orthographic structures
in the shallow-pointed Hebrew script: A cross-grade study in elementary
school.
Reading and Writing, 25
, 1217–1238.
Shany, M., Breznitz, Z. (2011). Rate- and accuracy-disabled subtype profiles among
adults with dyslexia in the Hebrew orthography.
Developmental
Neuropsychology, 37
, 889–913.
Shany, M., & Share, D. L. (2011). Subtypes of reading disability in a shallow
orthography: A double dissociation between accuracy-disabled and rate-
disabled reading in Hebrew.
Annals of Dyslexia, 61
, 64–84.
Share, D. L. (2003). Dyslexia in Hebrew. In N. Goulandris (Ed.),
Dyslexia in different
languages: Cross-linguistic comparisons
(pp. 208–234). London, United
Kingdom: Whurr.
Share, D. L. (2008). On the Anglocentricities of current reading research and
practice: The perils of over-reliance on an “outlier” orthography.
Psychological Bulletin, 134
, 584–616.
Share, D. L. (2012). Frost and fogs, or sunny skies? Orthography, reading, and
misplaced optimalism.
Behavioral and Brain Sciences, 35
, 45–46.
Comment [INT27]:
Unlinked
319
Share, D. L., & Blum, P. (2005). Syllable splitting among literate and pre-literate
Hebrew-speakers.
Journal of Experimental Child Psychology, 92
, 182–202.
Share, D. L., & Gur, T. (in preparation). When less is more:
Declining
code knowledge
signals progress in reading in a Semitic abjad.
Share, D. L., & Levin, I. (1999). Learning to read and write in Hebrew. In M. Harris &
G. Hatano (Eds.),
Learning to read and write
(pp. 89–111). Cambridge, United
Kingdom: Cambridge University Press.
Share, D. L., & Shalev, C. (2004). Self-teaching in normal and disabled readers.
Reading and Writing, 17
, 1–31.
Shatil, E., & Share, D. L. (2003). Cognitive antecedents of early reading ability: A test
of the modularity hypothesis.
Journal of Experimental Child Psychology, 86
,
1–31.
Shatil, E., Share, D. L., & Levin, I. (2000). On the contribution of kindergarten writing
to Grade 1 literacy: A longitudinal study in Hebrew.
Applied
Psycholinguistics, 28
, 1–25.
Shimron, J. (1993). The role of vowels in reading: A review of studies of English and
Hebrew.
Psychological Bulletin, 114
, 52–67.
Shimron, J., & Sivan, T. (1994). Reading proficiency and orthography: Evidence from
Hebrew and English.
Language Learning, 44
, 5–27.
Stanovich, K. E. (1986). Matthew effects in reading: Some consequences of
individual differences in the acquisition of literacy.
Reading Research
Quarterly, 21
, 360–406.
Comment [INT28]:
Unlinked
Comment [INT29]:
Unlinked
320
Stanovich, K. E. (2000).
Progress in understanding reading: Scientific foundations
and new frontiers
. New York, NY: Guilford.
Tolchinsky, L., Levin, I., Aram, D., & McBride-Chang, C. (2012). Building literacy in
alphabetic, abjad and morphosyllabic systems.
Reading and Writing, 25
,
1573–1598.
Tolchinsky, L., & Teberosky, A. (1998). The development of word segmentation and
writing in two scripts.
Cognitive Development, 13
, 1–14.
Tunmer, W. E., & Chapman, J. W. (2012). Does the set for variability mediate the
influence of vocabulary knowledged on the development of word recognition
skills.
Scientific Studies of Reading, 16
, 122–140.
... Verbs are always formed by a non-linear combination of a root and a verbal pattern (Ashkenazi et al., 2016 ). There are seven verbal patterns or conjugations (binyanim ), each conveying a specific aspect such as transitivity, reciprocity, reflexivity, causativity, and inchoateness (Share, 2017 ). In contrast to the small number of verbal patterns, there are over 100 nominal patterns, some of them clearly signaling semantic categories (e.g., CaCaC for professions; CaCoC for colors; CaCeCet for diseases), but most are unpredictable. ...
... That is, consonants are fully represented by letters arranged linearly from right to left. In contrast to alphabetic writing systems in which both consonants and vowels have full and equal representation, Hebrew vowels are subordinate to consonants; either they are omitted or have secondary status (Share, 2017;Shimron, 1993). ...
... Beginning readers are taught to read in pointed Hebrew, a near perfectly regular or transparent orthography in which each letter and vowel sign corresponds to a single phoneme (Bar-On & Ravid, 2011 ;Share & Bar-On, 2018 ). This phonological transparency affords a high degree of decipherability (Share, 2008(Share, , 2017, accounting for the rapid mastery of decoding by the end of first grade (Shatil & Share, 2003 ). Despite the one-symbol-one-sound transparency, the CV unit ("core syllable") is a common unit of instruction (Kogan & Saiegh-Haddad, 2023 ;Mizrachi, 2024). ...
Article
Full-text available
The study of Hebrew, a non‐European language written in a non‐alphabetic (abjadic) script offers valuable insights into the science of reading beyond the well‐studied alphabetic scripts. Because reading development in Hebrew is shaped by the uniquely Semitic root‐and‐pattern morphology and the abjadic (predominantly consonantal) orthography, our review begins by providing some basic information about the Hebrew language and its writing system. We then outline the developmental phases through which the developing Hebrew reader progresses, as proposed in Share and Bar‐On's Triplex Model, and the central role of phonological and morphological awareness. Finally, we review the research literature on developmental dyslexia in Hebrew within the context of current theories on the underlying perceptual/neurobiological basis of dyslexia (e.g., temporal processing, magnocellular deficits, visual attention span etc.). Here, we draw attention to the problems of comorbidity and the need to go beyond the conventional “whole‐group” approach and address the issue of heterogeneity and subtypes.
... The word pattern, consisting of a sequence of vowels or vowels and consonants, determines vocal, syntactic-semantic properties, and the word's classification in terms of speech parts (Ravid and Schiff 2006). Despite this complexity, Hebrew orthography maintains high morphological transparency, with root and pattern morphemes clearly represented, even more so than in speech (Ravid and Schiff 2006), and are considered cornerstones for understanding Hebrew orthography and developing reading skills (Share 2017;Share, Shany, and Lipka 2019). Early exposure to this complex yet transparent morphological system, profoundly influences Hebrew speakers' linguistic and cognitive development, fostering an early reliance on morphological information from the onset of reading acquisition ( Bar-On and Ravid 2011;Frost 2012;Ravid and Schiff 2006;Share 2017;Share and Bar-On 2017). ...
... Despite this complexity, Hebrew orthography maintains high morphological transparency, with root and pattern morphemes clearly represented, even more so than in speech (Ravid and Schiff 2006), and are considered cornerstones for understanding Hebrew orthography and developing reading skills (Share 2017;Share, Shany, and Lipka 2019). Early exposure to this complex yet transparent morphological system, profoundly influences Hebrew speakers' linguistic and cognitive development, fostering an early reliance on morphological information from the onset of reading acquisition ( Bar-On and Ravid 2011;Frost 2012;Ravid and Schiff 2006;Share 2017;Share and Bar-On 2017). Morphological deficiencies among Hebrew readers have been documented in several studies (e.g., Cohen and Schiff 2019;Raveh and Schiff 2008;Schiff et al. 2016;Schiff and Raveh 2007), and it has recently been argued that 'PA deficits are also characteristic of poor readers in Hebrew, but morphological deficits seem to be equally important' (Share 2017, 170). ...
... Recent evidence demonstrates the centrality of MPI also in reading the transparent pointed version from second grade onwards, superseding reliance on phonological information (Yinon and Shaul 2024). Thus, second grade represents a "watershed" period in Hebrew reading acquisition, marking a gradual shift from phonological to morphological processing (Bar-On and Ravid 2011;Share 2017). This underscores the significant need for further exploration into the influence of morphological deficits, particularly in MPI, on the developmental trajectory of RD in Hebrew. ...
Article
Full-text available
While the multiple cognitive deficits model of reading difficulties (RD) is widely supported, different cognitive‐linguistic deficits may manifest differently depending on language and writing system characteristics. This study examined cognitive‐linguistic profiles underlying RD in Hebrew, characterised by rich Semitic morphology and two writing versions differing in orthographic consistency—a transparent‐pointed version and a deep‐unpointed version. A two‐step cluster analysis grouped 96 s graders and 81 fourth graders based on their phonological awareness (PA), rapid naming (RAN), orthographic knowledge (OK) and morphological‐pattern identification (MPI) abilities. Word‐reading accuracy and fluency in both writing versions were also examined. The analysis revealed substantial variation in cognitive‐linguistic functioning levels, identifying distinct homogeneous profiles at both ages. PA and RAN deficits were most significant across grades. A distinct OK deficit profile emerged among fourth graders, particularly manifested in Hebrew's deep‐unpointed script. Moreover, the findings converge on the conclusion that while the phonological deficit is primary and persists over time, its importance is secondary to the main morphological deficit, reflecting a central independent factor in explaining RD in Hebrew. These results support a hybrid‐dynamic perspective, emphasising how the relevance of different cognitive‐linguistic deficits underlying RD varies with development, influenced by Hebrew's unique orthographic and morphological characteristics.
... The present study defines phonological awareness as one's ability to detect, manipulate, or analyze auditory speech sounds (Richardson & Nieminem, 2017). Share and Levin (1999) concluded that "phonology may well be a universal and inescapable feature of early reading and writing" (p.107). Research has shown a prominent role played by phonological awareness in higher achievement in early writing skills and inefficient performance at different stages of handwriting acquisition (Ray et al., 2021). ...
... Based on an earlier research review, we focused on several factors that may contribute to early handwriting performance. Research suggested that linguistic skills (phonological awareness and orthographic knowledge) (Leppanen et al., 2006;Mathwin et al., 2022;Pritchard et al., 2021;Richardson & Nieminem, 2017;Share & Levin, 1999), as well as graphomotor and fine-motor skills, (Ray et al., 2021) are related to early handwriting performance in several languages. It is hypothesized that linguistic skills best predict early handwriting performance (Bazyk et al., 2009;Ray et al., 2021), followed by graphomotor and fine-motor skills. ...
Article
Full-text available
This study aimed to delve into the under-explored domain of early handwriting performance among Arabic-speaking kindergarten children, focusing on the potential factors influencing early handwriting competency. The research encompassed 218 children from diverse socioeconomic backgrounds in Israel. The underlying skills assessed were divided into linguistic skills (phonological awareness and orthographic knowledge) and graphomotor and fine-motor skills. Hierarchical regression analyses were utilized to evaluate the contributions of these skills. Results indicated that, within the Arabic orthographic context, orthographic knowledge stood out as a paramount contributor to early handwriting performance, more so than phonological awareness. Furthermore, graphomotor and fine motor skills significantly influenced letter-copying speed and legibility, but not the accuracy of letter-writing to dictation. In conclusion, while orthographic knowledge is paramount, the importance of graphomotor and fine motor skills for early handwriting performance in Arabic cannot be understated. The study suggests that a focused approach to these skills can lead to more effective interventions and teaching methodologies tailored for Arabic-speaking kindergarteners.
... Hebrew is also a language with two scripts, pointed and unpointed script and uses the abjad writing system, or consonantal writing system, where only consonantal sounds are represented, and vowels are omitted (Ravid, 2006;Share, 2017). The pointed script uses consonantal letters plus 12 diacritic marks (nikud) to represent the five vowel sounds of modern Hebrew (Ravid, 2006;Share & Bar-On, 2017). ...
... Children learning to read are expected to use the pointed version during the first two years of instruction before moving to the unpointed version during the third year of instruction (Share, 2017). Due to dual scripts, the Triplex Model of Hebrew Reading Development (Share & Bar-On, 2017) outlines three stages of reading acquisition from sublexical phonological spelling-to-sound translation with the pointed script (Grade 1) to larger units of lexico-morpho-orthographic processing (Grade 2) and supralexical contextual reading. ...
Article
Full-text available
There is evidence that dyslexia occurs in many languages around the world. As many have attempted to develop theories that encompass the multidimensionality of dyslexia across languages, they fall short when accounting for variation due to orthographic depth while also focusing on European alphabetic languages. Many students with dyslexia demonstrate difficulties in languages that are neither European nor alphabetic. Therefore, a more comprehensive model is needed to provide consistency in understanding, assessment, and identification for a greater number of students. We employ the ten dimensions of orthographic complexity proposed by (Daniels and Share, Scientific Studies of Reading 22:101–116, 2018) to discuss how orthographic features (e.g., spoken-written distance, spatial arrangement and nonlinearity, historical change, and omission of phonetic elements) are associated with reading acquisition and difficulties. We review and synthesize literature to provide a background for reading acquisition and dyslexia in alphabetic languages (e.g., Finnish, Spanish, German, Portuguese, French, and English; Seymour et al., British Journal of Psychology 94:143–174, 2003) that range from transparent to opaque in orthographic depth as well as other languages that are Semitic (e.g., Arabic and Hebrew), syllabic-alphabetic (e.g., Korean), and morphosyllabic (e.g., Chinese). Implications for assessment, identification, and intervention, as well as future directions will be discussed. This approach could aid in a shift in perspective from ‘Anglocentric’ and ‘alphabetocentric’ (Daniels and Share, Scientific Studies of Reading 22:101–116, 2018; Share, Psychological Bulletin 134:584–615, 2008, Share, Reading Research Quarterly 56:S391–S402, 2021) to a broader view that addresses a wider variety of orthographies.
... All 22 letters represent consonants, with vowels represented (i) by four non-optional dual-purpose letters that double as vowels and (ii) a set of 13 optional diacritic-like vowel signs, primarily sub-lineal, which are the standard form of Hebrew for beginning readers, children's literature, poetry, and liturgical texts. When fully "vowelized" with the vowel diacritics, Hebrew text is referred to as pointed: when the optional vowel diacritics are omitted, this form of Hebrew is referred to as unpointed which is the standard form for skilled readers (Share, 2017). It is also important to point out that Hebrew has a rich and complex non-concatenative derivational and inflectional morphology. ...
... We suspect that our threeletter words actually had more degrees of freedom (or ambiguity) regarding vowel patterns because far more morphological patterns exist for these shorter words (which contain only root letters) than for the longer fiveletter words which in our study included many affixed consonants that specify the morpho-phonological pattern and hence the specific vowelling pattern. Although in our study, all words were fully pointed (i.e., all vowels (as well as consonants) are unambiguously marked), it is well known that after Grade 1, when children are taught all the consonantal letters and vowel signs, knowledge of the vowel signs declines rapidly (see, for example, Shany et al., 2012) and is replaced by reliance on higher-order information such as morphological patterns and specific lexical knowledge (Share, 2017). With regard to the age by word length interaction, we speculate that although this effect operates on both age groups, it is stronger in adults compared to children. ...
Article
Full-text available
Previous studies examining the link between visual word recognition and eye movements have shown that eye movements reflect the time-course of cognitive processes involved in reading. Whereas most studies have been undertaken in Western European languages written in the Roman alphabet, the present developmental study investigates a non-European language—Hebrew, which is written in a non-alphabetic (abjadic) script. We compared the eye-movements of children in Grades 4 to 6 (N = 30) and university students (N = 30) reading familiar real words and unfamiliar (pseudo)words of 3 letters and 5 letters in length. Using linear mixed models, we focused on the effects of word familiarity, word length, and age group. Our results highlight both universal aspects of word reading (developmental and familiarity (lexicality) effects) as well as language-specific word length effect which appears to be related to the unique morphological and orthographic features of the Semitic abjad.
... Consequently, uency and speed are oen the more discriminating measures of reading difficulty. Indeed, various studies conducted in Dutch (Yap and van der Leij, 1993), Finnish (Lyytinen et al., 2004), German (Wimmer, 1993), Greek (Porpodas, 2006), Hungarian (Csepe, 2006), Hebrew (Breznitz, 1997;Share and Levin, 1999), Italian (Zoccolotti et al., 1999), and Norwegian (Lundberg and Høien, 1990) showed that many dyslexics attain high levels of reading accuracy but remain slow. ...
Article
Full-text available
CITATION Jabbour-Danial M, Share DL and Shalhoub-Awwad Y (2024) Not just phonology: a longitudinal study of dyslexia subtypes based on the distinction between reading accuracy and reading rate. Introduction: Previous cross-sectional investigations by Shany and colleagues have provided evidence of double dissociation among dyslexics between word reading accuracy and (pure) word reading rate. A rate-disabled subtype (with intact reading accuracy) evinced deficits only in rapid naming (RAN); An accuracy-disabled subtype (with intact reading rate) showed deficits in phonological awareness (PA) and morphological awareness (MA) but not RAN. Method: The present longitudinal study followed 639 Palestinian Arabic-speaking children from preschool to Grade 1 with the aim of determining (1) whether a dissociation between PA C MA and RAN is apparent among pre-literate preschoolers, (2) whether the PA C MA-disabled subgroup constitutes a mild form of Developmental Language Disorder (DLD), (3) whether our three disabled subgroups (PA C MA-only, RAN-only, and PA C MA-plus-RAN) can be differentiated on preschool early literacy measures and, (4) whether the three at-risk subgroups develop into selectively rate-disabled, accuracy-disabled, or doubly-disabled (accuracy C rate) readers in Grade 1? Results: Our findings confirmed the existence of two distinct selectively disabled/at-risk subgroups in preschool: a RAN-only subgroup with intact PA and MA and a PA C MA subgroup with broad impairments across language measures but intact RAN. Grade 1 reading data also confirmed that the RAN-disabled subgroup became slow but accurate readers, whereas the PA C MA subgroup developed into inaccurate and slow readers. Discussion: Our study indicates partial dissociation between early dyslexia subtypes, each displaying distinct and non-overlapping cognitive-linguistic profiles in preschool. The study also revealed a strong association between reading accuracy and reading rate among beginning readers. This study emphasizes the importance of considering heterogeneity in reading outcomes as well as multiple oral language skills beyond the well-documented role of PA.
Article
Full-text available
The relative importance of phonological versus morphological processes in reading varies depending on the writing system's orthographic consistency and morphological complexity. This study investigated the interplay between phonology and morphology in Hebrew reading acquisition, a language offering a unique opportunity for such examination with its rich, complex Semitic morphological system and dual writing versions differing in orthographic consistency—transparent-pointed and deep-unpointed versions. Ninety-eight second graders and 81 fourth graders participated in pseudoword-reading tasks designed to distinguish between the different processes: pointed morphologically based pseudowords (pointed MPW), reflecting phonological and morphological processing; unpointed morphologically based pseudowords (unpointed MPW), reflecting only morphological processing; and pointed non-words (pointed NW), with no internal morphological structure, reflecting only phonological processing. Real pointed-word reading accuracy and fluency were also assessed. Results showed the highest accuracy in reading unpointed MPW, with a similar accuracy level observed between unpointed MPW and pointed MPW in second grade, while a significant difference emerged in fourth grade. An age-by-processing type interaction revealed decreasing accuracy in pointed MPW and increasing accuracy in unpointed MPW with age. Additionally, morphological processing significantly enhanced the accuracy and fluency of reading pointed words beyond phonological processing, despite the comprehensive phonological information provided by the transparent, pointed script. These findings suggest that the contribution of morphology exceeds that of phonology as early as second grade, with this trend strengthening through fourth grade, emphasizing children's early prioritization of morphological transparency over orthographic consistency in learning to read Hebrew Semitic orthography.
Article
Full-text available
It is well known that learning to spell is a complex and challenging process, especially for young learners, in part because it relies on multiple aspects of linguistic knowledge, such as phonology and morphology. The present longitudinal study investigated the role of morphology in early spelling in two Semitic languages, Hebrew and Arabic, that are structurally similar but differ in the phonological consistency of phoneme to letter mappings (“backward consistency”). Whereas Arabic mappings are mostly one-to-one – allowing children to rely mainly on phonology to spell words correctly, Hebrew has numerous one-to-many phoneme-to-letter mappings that are governed by morphological considerations, thereby precluding a purely phonological spelling strategy. We, therefore, predicted that morphology would make a more substantial contribution to early Hebrew spelling than to Arabic spelling. We tested this prediction in a longitudinal study of two large parallel samples (Arabic, N = 960; Hebrew, N = 680). We assessed general non-verbal ability, morphological awareness (MA), and phonological awareness (PA) in late Kindergarten and spelling in the middle of the first grade with a spelling-to-dictation task. Hierarchical regression analyses revealed that after controlling for age, general intelligence, and phonological awareness, morphological awareness contributed a significant additional 6% variance to Hebrew spelling but only 1% to Arabic word spelling. The results are discussed within the framework of the Functional Opacity Hypothesis (Share, 2008), which we extend to spelling.
Thesis
Full-text available
L’objectif de cette thèse était de comprendre le rôle de la syllabe pour la mise en relation des premières relations ortho-phonologiques. La question de la taille de l’unité psycholinguistique (Ziegler & Goswami, 2005) à prendre en compte lors de la création des premières connexions entre le langage oral et le langage écrit a amené Doignon-Camus et Zagar (2009, 2014) à proposer un modèle (Developmental Interactive Activation Model with Syllables, DIAMS) qui s’appuie sur la disponibilité phonologique précoce de la syllabe chez les pré-lecteurs (Liberman et al., 1974) et la fonctionnalité de celle-ci pour le traitement du langage écrit (i.e., les enfants perçoivent visuellement la syllabe au cours de l’apprentissage de lecture, Doignon-Camus & Zagar, 2014). Ce modèle suppose que les premières relations entre le langage écrit et le langage oral s’établiraient grâce à la construction d’un pont syllabique, c’est-à-dire entre les lettres et les syllabes correspondantes. Pour tester l’hypothèse du pont syllabique, comme moyen de construction des relations ortho- phonologiques, nous avons mis en place une expérimentation comportementale auprès de plus de 400 enfants pré- lecteurs scolarisés en dernière année d’école maternelle (Grande Section). Les résultats ont mis en évidence le rôle de la syllabe pour acquérir les relations ortho-phonologiques, grâce notamment au développement significatif des compétences en conscience phonémique après un enseignement des relations lettres-syllabe, par rapport à un enseignement des relations lettre-phonème. D’autres études complémentaires nous ont permis de mettre en évidence la facilité et la rapidité avec laquelle le pont syllabique peut être construit pour développer l’acquisition du code alphabétique. A partir des résultats obtenus chez une population d’enfants tout-venant, nous avons également testé l’hypothèse du pont syllabique auprès d’enfants à risque et en difficulté de lecture plus avancés en âge et dans leur parcours scolaire, à travers une méthodologie de cas unique. Les résultats de ces études ont mis en exergue l’efficacité de l’intervention basée sur la syllabe sur les compétences en conscience phonémique et en lecture (i.e., précision et automatisation) chez un public à risque et en difficulté de lecture. Cette thèse tentait donc de répondre à la question de la taille de l’unité lors de l’acquisition des relations ortho-phonologiques, et a amené de nouvelles perspectives de recherche. D’un point de vue pédagogique, les résultats ont consolidé les pratiques enseignantes en se basant sur des preuves issues de la recherche expérimentale, et ont également apporté des propositions de remédiation pédagogique et de rééducation orthophonique.
Article
Full-text available
This book puts together contributions of linguists and psycholinguists whose main interest here is the representation of Semitic words in the mental lexicon of Semitic language speakers. The central topic of the book confronts two views about the morphology of Semitic words. The point of the argument is: Should we see Semitic words’ morphology as “root-based” or “word-based?” The proponents of the root-based approach, present empirical evidence demonstrating that Semitic language speakers are sensitive to the root and the template as the two basic elements (bound morphemes) of Semitic words. Those supporting the word-based approach, present arguments to the effect that Semitic word formation is not based on the merging of roots and templates, but that Semitic words are comprised of word stems and affixes like we find in Indo-European languages. The variety of evidence and arguments for each claim should force the interested readers to reconsider their views on Semitic morphology.
Article
Full-text available
Maternal writing mediation and children's literacy were analysed in two writing systems; the Semitic abjad and the European alphabet. Forty Israeli Hebrew-speaking and 43 Spanish-speaking mother-child dyads participated in this study. The children, aged: M=68.58 months, had not yet been exposed to formal reading instruction. Israeli kindergartners embark on their initial steps in reading and spelling in an abjad—a consonantal writing system that deficiently and inconsistently marks vowels by letters. Spanish kindergartners are introduced to a shallow alphabetic writing system that consistently marks consonants and vowels. This study assessed: (1) children's code-based skills (letter knowledge and phonological awareness), spelling, and reading, and (2) mothers' word writing mediation. The groups were basically similar in code-based skills, but the reading and spelling of the Israeli children were substantially lower than those of their Spanish counterparts. Maternal writing mediation was lower among Israeli than Spanish mothers with respect to vowels. Regression analyses showed that children's spelling in Hebrew and in Spanish were predicted by children's code-based skills and by maternal writing mediation. Children's reading in Hebrew was uniquely predicted by code-based skills and in Spanish by maternal writing mediation. This study sheds light on the importance of writing mediation and its relation to writing systems.
Article
Full-text available
In focusing on English and Hebrew writing systems, the author suggests that writing systems could function without vowels, but once introduced the vowels are useful for (1) disambiguating homophones and homographs, (2) marking the pronunciation of other phonemes that may be pronounced in various manners, (3) facilitating the parsing of sublexical elements, and (4) signifying phonological information. Relating the studies reviewed to a new approach in word-recognition research, it is argued that the role of vowels is additive (sometimes complementary, sometimes redundant) to the roles of other cues facilitating word recognition. The decoding of unvoweled words does not necessarily exclude phonological processing, and the decoding of voweled words may benefit from lexical input. The author also suggests that the viability of reading unvoweled print is restricted by the features of word morphology. (PsycINFO Database Record (c) 2012 APA, all rights reserved)
Chapter
This paper examines the question of what constitutes a good writing system. I have taken it as given that some writing systems are better than others. This question could certainly be debated. The opposite position would be that all writing systems are equally good for all languages, that it simply makes no difference which writing system is used. As far as I know, very few people have held this position, either academics or lay people. Consider the presence of proposed spelling reforms for English, the recent character simplifications of China, and the considerable thought that is often given to establishing an orthography for a previously unwritten language. All these are clear testimony to the fact that people generally feel that some writing systems are better than others. (Berry, 1977;Justeson, 1988;Sampson, 1985;Winter, 1983)
Article
A very influential suggestion in I. J. Gelb's Study of Writing concerns the nature of the West Semitic scripts. Usually thitherto called "alphabets," the Hebrew, Phoenician, etc., scripts are for Gelb "syllabaries" whose characters have values like bx; a corollary is that the Ethiopian script is not a syllabary but an alphabet. While this notion has been widely adopted, it remains counter-intuitive. A recent suggestion of P. Swiggers-to distinguish between "denoting" and "standing for" a syllable-helps to clarify Gelb's thinking but does not alleviate the terminological difficulties. A characteristic of Gelb's work in many fields may account for his view of the evolution of writing systems and seems to have led to the mischaracterization of some scripts noted by previous commentators. Here a typology of scripts is proposed that recognizes both a new type intermediate between syllabary and alphabet and a fundamental distinction between two kinds of syllabary.
Article
THE ABILITY to consciously use linguistic knowledge in the semantic, phonologic, and morphologic domains was tested in fifth-grade children with reading disabilities and compared with that of normally reading age-matched controls and with a 2-years-younger group of controls matched with the reading disabled for vocabulary size. Ail testing involved speech. Children with reading disabilities were inferior to both control groups in their ability to assign words to semantic categories, to identify the first phoneme in spoken words, and to judge the morphologic relationship between word pairs. Their relative impairment was most conspicuous in the morphologic domain. However, the number of words produced by the reading disabled according to semantic, phonologic, or morphologic constraints was similar to that produced by the younger vocabulary-matched group, and both groups were inferior to the age-matched controls. It is concluded that the developmental reading-disabilities syndrome includes a deficit in the linguistic skills necessary for conscious application of linguistic rules in word perception and production. This linguistic impairment may explain some of the reading comprehension problems as well as the difficulty in deciphering printed words.