An Introduction to Applied Linguistics

see also Chapter 3, Vocabulary). An increasing amount of L2 research has focused 
more specifically on processes of L2 word recognition, with the findings having 
real implications for instruction.
Koda (2005) makes the case that word recognition in second language reading 
must be viewed as a major factor in its own right and not just as a facet of overall 
second language proficiency. She argues that L2 word recognition is affected 
by the amount of L2 orthographic processing experience, the distance between 
the orthographies of the L1 and L2, and the interaction between L1 and L2 
orthographic knowledge.
Earlier word recognition studies have shown differences in word recognition 
efficiency among learners with different amounts of L2 experience. Favreau and 
Segalowitz (1983) showed that even for otherwise fluent bilinguals, if a second 
language is weaker than the first language and reading is slower in the second 

---

219
Reading
language, word recognition in the L2 is less automatic than in the L1. In a follow-
up study of the development of automaticity in French speakers learning ESL, 
Segalowitz and Segalowitz (1993) showed that practice on word recognition tasks 
leads to faster and more stable (less variable) responses. These faster and more 
stable responses indicated that processing had not merely become faster across the 
board, but that a qualitative change or restructuring of processing had occurred. 
They maintained that this reflected the attainment of automatization, not just a 
simple speeding up of the processing mechanisms.
Although Segalowitz’s various studies have examined mature, literate adult 
readers, Geva, Wade-Woolley and Shany (1997) have focused on younger learners 
learning to read simultaneously in English (L1) and Hebrew (L2). Geva et al. (1997) 
conclude that steps associated with the development of L1 reading efficiency (that 
is, accuracy attained before speed) may be applicable to the development of word 
recognition skills in L2, but they do not emerge concurrently in both languages. 
They also conclude that linguistic features, such as ‘orthographic depth’ (the degree 
to which the written system of a language corresponds to its spoken system) and 
morphosyntactic complexity ‘may interact with more global L2 proficiency effects’ 
(Geva et al., 1997: 119) to determine the course of early L2 reading development 
(see also Geva and Wang, 2001; Lesaux, Lipka and Siegal, 2006).
Implications: Word recognition exercises are probably useful for both older and 
younger L2 readers, enhancing fluency and raising student awareness of the 
processing demands of extended independent reading.
L2 Word Recognition Differences across L1s
According to the orthographic depth hypothesis (Frost, 2005; Seymour, 2006), 
pre-lexical phonology (the immediate and automatic matching of graphemes 
and phones to produce word recognition) plays a more important role in lexical 
access in ‘shallow orthographies’, where the correspondences of graphemes 
to phonemes are more direct and consistent (for example, Finnish, Spanish, 
Turkish) than in ‘deep orthographies’, where the mapping of letters to sounds is 
less direct and less consistent (for example, English, unmarked Arabic, Chinese). 
Cross-linguistic research comparing L2 learners with different L1 backgrounds 
has consistently demonstrated superior word recognition performance for those 
with L1 orthographic backgrounds more similar to the L2. For example, Koda 
(1989) found better word recognition for L2 learners of Japanese with related L1 
backgrounds (Chinese and Korean) than she did for unrelated ones (English). 
Muljani, Koda and Moates (1998) showed this effect again for ESL learners from 
related (Indonesian [Roman alphabet]) versus unrelated (Chinese [logographic]) 
L1 orthographic backgrounds (see also Koda, 2005).
Green and Meara (1987) showed differences between three groups of ESL 
learners with contrasting L1 orthographic backgrounds: Spanish speakers (Roman 
alphabetic orthography), Arabic speakers (non-Roman alphabetic orthography) 
and Chinese speakers (non-alphabetic orthography). The researchers concluded 
that the three groups used different visual processing strategies when pursuing 
a search task not only in their L1s but also in their L2s. Green and Meara (1987) 
concluded that L1 writing systems have a deep and lasting effect on the ways 
in which L2 materials are processed. Ryan and Meara (1991) investigated the 

---

220 An Introduction to Applied Linguistics
hypothesis that Arabic speakers, because of the emphasis on consonants in the 
lexical structure and orthography of their L1, would also tend to rely heavily on 
consonants when attempting to recognize L2 English words. In a task that required