There has been a recent upsurge of interest in languages generally thought of as less commonly taught in traditional world language education. In the United States, for example, languages such as Chinese, Japanese, and Arabic, while still only a small part of the foreign language educational landscape, have begun to experience steady or growing enrollments in schools (Modern Language Association, 2007), and are beginning to be attractive alternatives as curricular offerings at all educational levels. Aside from the enrollment trends, there is interest in expanding instruction in less commonly taught languages such as Chinese in the lower grades, with the eventual goal to have longer articulated learning sequences (Asia Society, 2010).

With selected U.S. states developing roadmaps to meet the demands of diverse populations, there is a new importance placed upon the nurturing of heritage language learners, or those who grow up speaking a second language at home but who often do not develop literacy skills in these languages. Within the American context, heritage learners are now considered an important priority in maintaining national language capacity (Brecht & Rivers, Reference Brecht and Rivers2000; Brecht & Walton, Reference Brecht and Walton1993) with specific instructional needs that must be addressed (He, Reference He, He and Xiao2008; Koda, Zhang, & Yang, Reference Koda, Koda and Zehler2008; Kondo-Brown, Reference Kondo-Brown2010; Montrul, Reference Montrul2010). To meet these challenges, states are also beginning programs (Asia Society, 2010; Falsgraf & Spring, Reference Falsgraf and Spring2007) in the early grades that employ immersion and bilingual models of language education whereby content is learned in the target language, thus providing alternatives to more traditional foreign language education models.

As well, the Flagship language initiative makes a variety of less commonly taught languages such as Chinese, Hindi, Urdu, Korean, Persian, and Russian available with the goal of taking students to superior levels of proficiency, levels that have not been attainable through traditional sequences of language exposure offered in university settings. As less commonly taught languages begin their ascent, initiatives designed to prepare teachers have begun with programs such as STARTALK, a U.S. government initiative which over the past three years has begun summer programs nationwide to train teachers in Chinese, Arabic, Hindi, Swahili, Turkish, Persian, and Urdu.

The majority of these languages are also of interest for educators and researchers because they share something in common that is different from the more traditional foreign language offerings of Spanish, French, and German—they do not employ the Roman alphabet, but instead use a totally different alphabetic system as in the case of Arabic and Hebrew, a non-alphabetic logographic system in Chinese, or a mix of native syllabary systems with borrowed Chinese characters, in the case of Japanese. If these languages are ever to take their place as more commonly taught languages in world languages education, it will be essential that we come to understand how these languages are mastered by students of different ages, who come from different literacy backgrounds, and who learn them in a variety of different learning settings.

To this end, this article will highlight some of the theoretical issues that underpin the reading of selected languages employing non-Roman orthographies. While there are many languages that employ such writing systems, this article will highlight the languages of Chinese, Japanese, Arabic, and Hebrew. The selection of these languages is not entirely arbitrary—Chinese employs a writing system that is non-alphabetic in nature, a system that has been responsible for transmitting through its written artifacts a culture that goes back millennia. Japanese, while also employing Chinese characters, employs syllabary systems that make it one of the most unique multi-script writing systems in the world. Arabic and Hebrew, on the other hand, employ non-Roman alphabets that share the characteristic of vowels being sometimes, but not always, marked in text, thus presenting unique challenges in the learning of these languages. Arabic is also unique in that its diglossic nature employs a written language that is different from the dialects commonly spoken across the Arab world. These languages were chosen also because of burgeoning research traditions that have developed to investigate reading in these writing systems as second languages (L2), among learners of different ages who are learning these languages in different types of settings. This article, then, includes short summaries of research to illustrate some of the methodologies used to investigate various topics. The article will then discuss the search for best practices for teaching learners to read in these languages, practices that have yet to find consensus among their practitioners.

THE ORTHOGRAPHIC VARIABLE

Researchers have been able to document that certain languages take American learners longer to learn. For example, Jackson and Malone (Reference Jackson and Malone2010) report that State Department employees are enrolled in highly intensive daily instruction for almost 6 months (600 class hours) in a Western European language (e.g., French, Dutch, or Spanish), about 10 months (1100 class hours) for a so-called hard language (e.g., Russian, Hindi, or Thai), and 2 years for the most challenging languages (Arabic, Chinese, Japanese, and Korean) in order to attain professional competence. These expectations would be much lower in secondary and post-secondary language programs where the intensity of instruction as well as the number of contact hours would be much less. While languages such as Chinese, Japanese, and Arabic differ linguistically and culturally in a number of ways from languages such as French and Spanish, the writing systems employed in many of the languages stated above are either non-alphabetic in nature, or employ non-Roman alphabets that present special challenges for learners whose first language (L1) employs the Roman alphabet.

The Chinese writing system, for example, is often termed a logography because its primary unit, the character, has been thought in its structure to represent meaning instead of sound, although characters do often contain phonetic cues, with phonology thought to play an important role in Chinese character recognition and reading (Perfetti & Dunlap, Reference Perfetti, Dunlap, Koda and Zehler2008; Perfetti & Liu, Reference Perfetti and Liu2005; Perfetti, Liu, & Tan, Reference Perfetti, Liu, Tan, Kao, Leong and Gao2002, Reference Perfetti, Liu and Tan2005). In fact, approximately 85% of all Chinese characters are termed compound characters (Zhu, Reference Zhu and Yuen1988), whereby they contain a component that contains a radical conveying semantic information, and a phonetic element that hints at how the entire character is pronounced. For instance, the left hand part of each of the following characters is a semantic radical termed “three dot water” in Chinese, and classifies each of these characters as having something to do with water: 海 [hǎi] meaning sea, 洋 [yáng] meaning ocean, and 池 [chí] meaning pond. With regard to phonetic elements, the character 工 in Chinese by itself is pronounced [gōng], and is found in the following characters as the phonetic element, giving a hint as to the pronunciation of the entire character: 空 [kōng] empty, 貢 [gòng] contribute, and 紅 [hóng] red. Unfortunately, this phonetic element is not very reliable in giving a systematically precise pronunciation of the overall character.

In the Chinese writing system, the character constitutes the smallest pronounceable unit in the written language (Cheung, McBride-Chang, & Chow, Reference Chikamatsu, Cook and Bassetti2006) with each character pronounced as a syllable with generally only one pronunciation per character. Chinese is also written using the Roman alphabet, termed romanization. Although there have been many romanization systems in use to either transcribe or teach Chinese to foreign language learners, the pinyin system is currently the most often used for foreign language teaching purposes in America. Consequently, the topic of how students who come from an alphabetic system in English to learn Chinese characters has received a great deal of research attention, as researchers are interested in knowing whether or not phonological processing strategies brought from their first reading experience transfer when learning a writing system whereby phonology is less systematically portrayed in the actual written unit.

Japanese, on the other hand, employs multiple scripts, thus offering a series of different challenges to American foreign language learners. Two of the scripts are syllabary systems whereby sound maps onto the printed symbol at the syllable level. These two syllabaries consist of 46 basic symbols and also have somewhat specialized functions, with one (hiragana) being used primarily for function words such as case-marking particles, and the other (katakana) being used for words borrowed from other languages. To complicate matters, Japanese also uses Chinese characters termed kanji. These characters operate differently in Japanese than in Chinese, however, in that a character can be pronounced with more than one syllable, and with different pronunciations depending upon the contextual use of the specific kanji. Japanese can also be written using the Roman alphabet, called romaji, which is used for road signs, as well as introducing foreign language learners to Japanese vocabulary before the native writing systems are introduced. To give an example, the sentence: ジョンスミスは私のアメリカ文学の教授である means “John Smith is my American literature professor.”: ジョンスミス and アメリカ are foreign words meaning “John Smith” and “America,” so are written in katakana. 私, 文 学 and 教 授 are written in kanji, and mean “I,” “literature” and “professor.” The other symbols are hiragana, serving the grammatical functions of subject marker, possessive marker, and verb. Consequently, similar research questions have been asked about how foreign language learners approach the learning of Chinese characters, as well as the syllabary scripts that are more finite.

Non-Roman alphabets such as Hebrew and Arabic also present their own sets of challenges. Hebrew, for example, is written right to left, with 22 letters, most of which denote consonants and an ancillary system of 13 vowel diacritic marks (Ravid, Reference Ravid, Joshi and Aaron2006). Texts designed for proficient readers of Hebrew omit vowels, but texts designed to teach reading to children and L2 learners include the vowels, as they provide a transparent pronunciation of syllables involving very few exceptions in terms associated with specific letter strings, position in a word, and vowel combinations (Geva & Wade-Woolley, Reference Geva, Wade-Woolley, Yucesan Durgunoglu and Verhoeven1998), facilitating accurate word decoding even by novice learners (Geva & Siegel, Reference Geva and Siegel2000).

Challenges in learning to read Modern Standard Arabic (MSA) stem from the fact that MSA is the literary language of the Arab-speaking world, with newspapers, most books, and magazines written in MSA. Children will learn MSA in school, but in their homes grow up speaking one of the many spoken dialects of Arabic which are generally not written. MSA, therefore, can serve as a means of oral communication between educated Arabic speakers whose regional or national dialects are mutually unintelligible. It is also used for certain television shows such as news broadcasts, and the formal language of education such as school lectures. The downside of learning MSA for Arab children is that they generally have no oral/aural experience with MSA when they begin learning it in school, making it more akin to learning a second language (Fender, Reference Fender, Koda and Zehler2008). To complicate matters, the Arabic writing system as used in texts for proficient readers, does not represent short vowels, meaning that the reader will be aided by the surrounding context to determine the proper pronunciation. Children learning to read MSA, however, are helped with this process, as diacritics are used in the text to mark the presence of short vowels, thus giving learners the means to phonologically decode words. As children develop more proficiency in MSA, these diacritics are dropped so that their materials resemble those more commonly used by experienced MSA readers. Foreign language learners of Arabic whose first language writing system is alphabetic, then, may be aided in learning vowelled Arabic given their L1 experience in deriving phonology from print. Vowelled Arabic will also be something foreign language learners and Arab children have in common, as they both are learning to read in a language they are also just learning to speak. Consequently, similar to Hebrew learners, the support they receive from diacritical vowel markings, and the support Chinese learners receive from romanization such as pinyin, will be a necessary component of best practice.

Because the theoretical frameworks that support foreign language reading are usually based on L2 learners who have already learned to read in their first language (L1), theoretical underpinnings have arisen largely from data derived from post-secondary learners of Chinese and Japanese, and often deal with various aspects of how students learn to read and the strategies they employ to cope with Chinese and Japanese scripts. That is, even though there are studies that involve learning to read Chinese and Japanese beyond the character and/or word level (Chang, Reference Chang, Everson and Shen2010; Everson & Ke, Reference Everson and Ke1997; Lee-Thompson, Reference Lee-Thompson2008; Warnick, Reference Warnick and Nara2001), many of the findings center on the character or word. In her review of word recognition, Koda (Reference Koda1996, Reference Koda, Zhang, Yang, He and Xiao2008) discusses frameworks stemming from connectionist theories that stress the importance of the effects of L2 processing experience, thus focusing on the effect of different proficiency levels as a variable of interest. Processing experience takes into account not only linguistic knowledge but also processing efficiency, factors that may develop separately. The other important framework takes into consideration L1 and L2 orthographic distance effects, or the theory that the extent to which the L2 shares orthographic properties with the learner's L1 may aid the development of word recognition efficiency in the L2.

These theoretical foundations are important for a number of reasons when examining post-secondary school learners of these languages. First, they have focused much of the research on script-specific topics, such as whether or not learners who come from L1 alphabetic backgrounds access the phonological aspects of print in orthographies such as Chinese that have been heretofore conceptualized as being more meaning-based scripts that do not lend themselves to phonological access by their readers. Second, they have encouraged a line of research investigating the strategies that learners employ to memorize, access, and infer the meaning of Chinese characters, as well as investigating the value placed on characters by both teachers and learners. Third, these foundations have taken into account the development of reading proficiency in studies involving mostly beginning but some advanced learners as research subjects. Lastly, it should be noted that this research has spotlighted the importance of word recognition and lower-level processing in an era of foreign language research that has been dominated by more schema-theoretic perspectives that emphasize the role of factors such as background knowledge in comprehension processing. One of the major contributions of this research, then, is that it has been able to document the difficulties L1 readers of English experience at the more micro-level of the reading process when learning to read in writing systems that are not alphabetically-based.

RESEARCH ON CHINESE

Given this backdrop of how the writing systems used in Chinese and Japanese differ so dramatically from the Roman alphabet learned in school by native English speakers, it should come as no surprise that much of the research has investigated how students of these languages learn Chinese characters. This is testimony to the complexity of the task facing foreign language learners of Chinese when they begin to learn characters, as their rudimentary knowledge of the spoken language combined with a yet undeveloped sense for how the orthography, morphology, and phonology operate, all contribute to making the task of learning Chinese characters a burgeoning area of research.

McGinnis (Reference McGinnis and Chu1999), for example, surveyed first year college-level students of Chinese about the strategies they employed to learn characters, and found that initially learners preferred repeated practice and character-specific mnemonic aids in the form of personal stories to memorize a character's sound, meaning, or shape. At this stage, students did not deconstruct characters into their radical or phonetic elements, but focused on memorizing characters as holistic units. In a similar study, however, Ke (Reference Ke1998) found that as beginning students progressed, those who perceived the learning and use of character components as more important strategies performed better in tasks of character recognition and production. These learners also felt that practicing characters in context so that they were combined to form multi-character words was more effective than learning the characters individually. It seems, then, that students early on in the character learning process develop an understanding of how Chinese characters are composed and the need to look for recurring elements such as radicals and phonetics, but due to the sheer number of recurring components in Chinese characters, have not yet learned enough characters to generate rules that will help them decompose characters systematically. As their orthographic awareness increases, however, learners become more reliant on their knowledge of character component structures, a finding corroborated by Shen (Reference Shen2005) who determined through survey research that the use of strategies whereby students used character components as aids to figure out character pronunciation and meaning was among the most heavily replied upon by beginning learners.

A number of other studies corroborated this research and lent more evidence to the importance of radical knowledge as an aid in a variety of character identification tasks. Taft and Chung (Reference Taft and Chung1999), for example, found that novice learners of Chinese used their knowledge of Chinese radicals as an aid to recall the meanings of characters that were newly learned, especially when the target characters were initially presented to the learners with more explanation about the meanings and use of radicals. As well, Jackson, Everson, and Ke (Reference Jackson, Everson, Ke, McBride-Chang and Chen2003) found that students completing their first year of Chinese were able to guess the meaning of unfamiliar characters in a multiple-choice format based on the meaning of the character's semantic radical. In other words, early on in their experience, beginning learners of Chinese who knew the meaning of a Chinese radical by itself could apply the radical to guessing the meaning of characters they had not yet formally learned, though this ability was somewhat variable.

Wang, Liu, and Perfetti (Reference Wang, Liu and Perfetti2004) also found that beginning learners learned the structure of Chinese orthography early in their learning experience, perhaps due to their scrutiny of the characters to compensate for their inability to readily access the pronunciation of the characters, and that explicit instruction hastened the learning process. These findings were extended by Shen and Ke (Reference Shen and Ke2007), who investigated learners’ awareness of radicals and this effect on word acquisition among students of three levels of Chinese proficiency. They discovered a moderate correlation across all levels between their overall Chinese character vocabulary and their ability to apply knowledge of radicals. As with the previously cited research, they also found that radical knowledge developed early among beginning learners, yet also found that it advanced rapidly during their first year of study, and that an overwhelming amount of their participants (93%) found the knowledge of radicals to be helpful in learning characters.

The finding that knowledge of radicals was helpful in learning characters was also found in Shen (Reference Shen, Everson and Shen2010) who investigated the learning of individual radicals by beginning learners of Chinese. Her findings indicated that learners considered the sounds and shapes of individual radicals to be more difficult to learn than their meanings, and that aural-oral practice and repetition was a learning technique most favored by the students.

In addition to this research that investigated the students’ ability to infer the meanings of Chinese characters, studies employing think aloud protocols have discovered the need for students to attempt to recover the pronunciation of the characters when they are reading (Everson & Ke, Reference Everson and Ke1997; Lee-Thompson, Reference Lee-Thompson2008) or learning characters (Winke & Abbhul, Reference Winke and Abbuhl2007). In his book decrying specific myths about Chinese, DeFrancis (Reference DeFrancis1984) referred to one as “the ideographic myth,” (p. 133) or the notion that Chinese as a writing system was able to bypass the intermediary of speech, a theory that DeFrancis (Reference DeFrancis1989) disputed. He even called for Chinese to be termed a logographic phonetic, instead of a logography to underscore the importance of the phonological aspects of Chinese characters. DeFrancis stated that there is no writing system that is read without phonology, a view shared by Perfetti, Zhang, and Berent (Reference Perfetti, Zhang, Berent, Frost and Katz1992) who have put forth the universal phonological principle in reading which states that there is no writing system that is read without phonology and that Chinese is no exception.

The think-aloud protocol evidence of laborious vocal decoding when Chinese learners read different authentic texts in Chinese led Everson (Reference Everson1998) to conduct a study that confirmed the importance for CFL learners of achieving a sound-symbol match when pronouncing and identifying the meaning of two character words. In this study, learners who had completed one year of Chinese were given a naming task whereby they would pronounce in Chinese a series of two-character words displayed to them on a computer monitor that were part of their first-year curriculum that they had already studied. Later, they were required to give the meaning for these and other words so that the relationship between the two tasks could be correlated. The results indicated that when learners could identify the meaning of a Chinese two-character word, there was about a 90% likelihood that they also could pronounce it, suggesting that the retrieval of meaning for these learners is not exclusively a visual process, and that learners use their spoken language resources to anchor the meaning of the characters. Lee-Thompson (Reference Lee-Thompson2008) found a common strategy among her intermediate learners when reading Chinese text was to write pinyin by the characters they were reading as an aid to recovering and remembering their pronunciation.

It should be noted that when involved in reading at the text level, intermediate level students (Everson & Ke, Reference Everson and Ke1997: Lee-Thompson, Reference Lee-Thompson2008) experienced difficulty isolating meaningful word units in the text, with students explaining in their think-aloud protocols or as witnessed by their marking of the text that the lack of spacing between characters and words in running text presents major challenges to students even at the intermediate level. That is, the lack of visible boundaries between words was a significant hurdle for students in processing the overall text. Follow-up research by Shen (Reference Shen2008) in which learners were asked to judge whether or not certain characters in a text went together as words indicated that these learners used a variety of strategies for determining which characters went together, while providing evidence that for even advanced learners, this is a difficult task. Bassetti (Reference Bassetti, Cook and Bassetti2005) puts forth the explanation that what constitutes a word in Chinese for Chinese language learners is not straightforward, possibly due to their L1 perceptions of what words are in English, and the effect of the complex system of word morphology that the Chinese language and orthographic system impose upon them as L2 learners. In a task where both third- and forth-year learners of Chinese as well as L1 readers of Chinese were asked to circle the characters that formed words in a text, her results indicated that the L2 learners and L1 readers of Chinese differed in their perceptions of what constitutes a word in Chinese, with many L2 learners expressing difficulty in the task due to the qualitative differences between English and Chinese word formation.

While more research is needed that investigates reading beyond the word and sentence level, data from advanced learners engaged in think-aloud protocols indicated that they possess a greater depth of understanding of Chinese orthography, morphology, and language use in general, with evidence that they could more easily isolate problems in word identification and text comprehension, thus making less random decisions than the less advanced learners about how to remedy problematic situations (Everson & Ke, Reference Everson and Ke1997). Again, advanced learners used a strategy of attempting to pronounce characters they did not recognize, while also employing surrounding context, text structure, background knowledge of the topic and the visual make-up of the unknown characters, thus testifying to an impressive repertoire for working through problems in both textual understanding and word recognition.

RESEARCH ON JAPANESE

Given that Japanese employs two syllabary scripts in the form of katakana and hiragana as well as Chinese characters, it is no wonder that much of the research in Japanese as a foreign language investigates how learners deal with a multi-script orthography. Early research involving beginning learners indicated that symbol identification speed distinguished good from poor readers (Koda, Reference Koda1990), and that lower level processing skills contributed significantly to successful text comprehension (Koda, Reference Koda1992), findings that set the stage for more research in this area. Such findings were important to start investigation into so-called lower level processing as much of the research to date at that time emphasized the importance of top down processing and not attention to the character or word level.

As with Chinese, an early strand of research investigated the extent to which learners use the phonological information available to them when recognizing words (Matsunaga, Reference Matsunaga1995, Reference Matsunaga1996, Reference Matsunaga and Nara2001), harkening back to the theory that readers whose L1 reading experience was with alphabetic languages, would be disadvantaged in reading languages whose orthographies are non-alphabetic, and therefore do not portray phonological information in a systematic, accessible manner.

Chikamatsu (Reference Chikamatsu1996), for example, investigated the word recognition strategies of Chinese and American learners of Japanese. Her purpose was to determine whether or not these learners employed strategies more akin to those used in their L1 (i.e., phonological processing for the American learners and visual processing for native speakers of Chinese). In this study, the subjects were shown a series of words from their vocabulary experience written in varying forms of kana, the Japanese syllabary scripts, and asked whether or not these were legal words. Her results indicated that the Chinese students employed more visual strategies and the American students, more phonologically-based strategies. Chikamatsu (Reference Chikamatsu2006) replicated this experiment with American L2 learners of varying proficiency levels, and found that the higher proficiency L2 learners of Japanese employed more visual information than the beginning level learners who, again, tended to use phonological information to a higher degree. This was interpreted to mean that beginning learners of Japanese whose L1 was English transferred their L1 processing strategies to the L2, but that this interference dissipated as proficiency in the L2 increased.

The predominance of L1 phonological strategies was also documented by Mori (Reference Mori1998) in a test designed to determine whether American learners would differ from Chinese and Korean learners in a short-term memory task when presented with pseudo-kanji characters that differed in the degree to which phonological elements were presented as pronunciation aids in the characters. Her results indicated that the American learners were hampered when the pseudo-characters prevented the learners from using a phonological analysis to remember the characters, whereas Korean and Chinese learners employed Chinese characters exclusively or partially in the Chinese and Korean writing systems.

As with the research in both L1 and L2 Chinese, researchers of L2 Japanese are also interested in how well learners can figure out unknown kanji encountered in isolation or in running text (Kondo-Brown, Reference Kondo-Brown2006b; Mori, Reference Mori2002; Reference Mori2003; Mori & Nagy, Reference Mori and Nagy1999; Toyoda, Reference Toyoda2000). Mori and Nagy (Reference Mori1999), for example, set up varying contextual conditions whereby students could guess the meanings of unknown (two) kanji word combinations. In their study, the learners employed an integrated strategy of inferring meaning from the composition of the kanji themselves, as well as the surrounding context. But when only one of these sources was available to the subjects, their ability to guess the meanings of the kanji decreased significantly. It also seemed that the learners’ strategy of using the clues within the kanji and their ability to use context was not correlated, indicating that the ability to extract information derived from morphological analysis of the kanji, and the ability to extract information derived from context may entail different processing skills.

Further investigation (Mori, Reference Mori2002) indicated that these two sources of information were additive, thus giving inferencing advantage to students who could integrate both sources of information. In a further study (Mori, Reference Mori2003), these results were borne out when students’ guesses with actual kanji meanings were compared. Mori (Reference Mori1999) indicated that the strategies readers chose for this task seemed to be related to their beliefs about learning in general, and their beliefs about language learning in particular. Kondo-Brown (Reference Kondo-Brown2006b) also investigated the inferencing of unknown kanji among two groups of Japanese learners, finding that inferencing ability improved with proficiency, although even among the higher proficiency group, inferencing ability was variable.

Inferencing ability also improved when learners could retrieve partial or full phonological renderings of one of the characters. In a study conducted by Toyoda (Reference Toyoda2000), the researcher had beginning learners of Japanese guess either the pronunciation or meaning of two-kanji word combinations they had not yet learned after undergoing several training sessions designed to heighten their awareness as to the purpose of Chinese character components. Her analysis indicated that even at this beginning stage, learners showed evidence of using their knowledge of the meaning and pronunciation components of kanji to make successful guesses, though they were more successful at inferring meaning than pronunciation.

Another interesting line of research has sought to determine attitudes both teachers and learners of Japanese have towards kanji learning, and how these attitudes may be related to learning strategies. Shimizu and Green (Reference Shimizu and Green2002) investigated the beliefs of Japanese teachers’ views of learning and teaching kanji, employing a principle component analysis that revealed six similar dimensions accounting for the variance. Teachers’ beliefs centered around factors dealing with beliefs that kanji were (a) useful in the Japanese reading process; (b) difficult; (c) embedded with cultural meaning; (d) potentially motivating to the students’ learning experience; (e) required a specific aptitude for learning; and (e) were not about to be replaced by a phonological script. Their analysis indicated that belief systems surrounding the importance of teaching kanji are complex and possibly influence their choice of kanji teaching strategies. The overall analysis indicated that the preferred strategies for teaching kanji were rote learning (strategies emphasizing drill and practice and repeated writing of kanji), memory strategy learning (linking the learning of new kanji with kanji already taught), and context strategy learning (emphasizing that the meaning of kanji in authentic reading situations is largely context-dependent). Particularly interesting findings were indications that teachers who emphasized the utility and cultural meaning of kanji, and who enjoy teaching kanji to students who enjoy learning them, seem less likely to emphasize rote learning strategies to their students.

Mori and Shimizu (Reference Mori and Shimizu2007) employed a similar measurement instrument used in the previous study to investigate students’ attitudes and beliefs towards learning kanji as well as their self-reported kanji learning strategies, and again found that these belief systems were complex and multi-dimensional. In summary, their analysis yielded six attitudinal factors (kanji is fun; kanji is difficult; kanji has cultural value; kanji has a future; kanji is useful; kanji learning requires special abilities) and six strategy belief factors (morphological analysis, rote memorization, context-based strategies, association strategies, metacognitive strategies, and helplessness). The complexity of the students’ belief systems is reflected in the different dimensions captured in the study, from attitudinal factors as to how difficult and enjoyable kanji are to learn to the students’ beliefs as to how useful their learning strategies seem to be, all the way working towards an appreciation of kanji as a purveyor of culture over time. Although correlational in nature, there were indications that certain relationships existed between belief systems and strategy use. For example, students who exhibited an appreciation of the cultural value of kanji seemed to have a more comprehensive view of effective learning strategies and were less likely to feel lost in the kanji learning process. There also was a relationship between enjoyment of learning kanji and belief in the efficacy of morphological analysis and the use of creativity in coming up with imaginative memorization strategies. On the other hand, perceptions of kanji difficulty were associated with strategies of rote memorization and a sense of feeling lost in the kanji learning process. Overall, however, students considered rote memory of kanji to be the most effective strategy for learning, and metacognitive strategies the least effective, indicating that students may in the end feel that kanji learning is a question of brute-force memorization, repeated writing and practice, and that ultimately, the overall structure of the character and/or its surrounding context, are not all that important in deriving kanji meaning. The researchers speculated that this might be due to teachers falling back on the way they learned to write Japanese, which potentially exerts a powerful influence on the way their students approach kanji learning. Attitudes and strategies were also the focus of a study (Mori, Sato, & Shimizu, Reference Mori, Sato and Shimizu2007) investigating learners’ ability to use context and morphological information to learn novel kanji words. Their results indicated a significant relationship between belief in the effectiveness of their learning strategies and their success in employing morphological analysis, thus highlighting the power of task-specific beliefs and metacognitive awareness in the word recognition process.

RESEARCH ON HEBREW AND ARABIC

Much of the research involving learning to read in Hebrew and Arabic focuses on children learning these languages in bilingual settings, both in FL and L2 settings, and some studies involving L1 Arabic learners of Hebrew. The research has also been framed in terms of the script dependent hypothesis (Geva & Siegel, Reference Geva and Siegel2000) and orthographic depth hypothesis (Katz & Frost, Reference Katz, Frost, Frost and Katz1992) which state that orthographies differ in how regularly they portray sound-to-symbol relationships, thus facilitating reading development in languages that employ more shallow orthographies. The reading of Hebrew and Arabic, therefore, presents unique opportunities for research in that Hebrew and Arabic orthographies are considered shallow when vowels in words are marked in text for beginning L1 and L2 learners, and deep when presented without vowels in texts designed for proficient readers.

From the viewpoint of L1 Arabic reading development, Abu-Rabia (Reference Abu-Rabia2002) has highlighted the vowelling of text and the presentation of words in context as critical components of L1 Arabic reading development. Studies investigating the role of vowelling and context among good and poor L1 readers of Arabic (Abu-Rabia, Reference Abu-Rabia1997), and reading employing different text types (Abu-Rabia, Reference Abu-Rabia1998b) were influenced by whether or not the texts were vowelled. Abu-Rabia (Reference Abu-Rabia2001) also found among L1 Arabic readers of L2 Hebrew performing reading tasks in both languages, that vowelling affected isolated and contextual word recognition as well as silent reading comprehension in both languages.

An alternative view termed the central-processing hypothesis states that skilled reading is not influenced by the orthographic system involved, but instead depends on linguistic and cognitive skills such as working memory, naming speed, and lexical processing, without which reading development will be difficult regardless of the orthographic system (Baluch, Reference Baluch, Joshi and Aaron2006). Much of the research testing these theories, then, is characterized by testing bilingual or trilingual (Abu-Rabia & Siegel, Reference Abu-Rabia and Siegel2003; Schwartz, Geva, Share, & Leikin, Reference Schwartz, Geva, Share and Leikin2007) learners of various ages in varying learning contexts (Abu-Rabia, Reference Abu-Rabia1998a, Reference Abu-Rabia1999) over a variety of reading performance and intelligence measures.

Researchers have found evidence that supports both of these frameworks, leading some to state that these hypotheses are not either/or, but in fact complementary. Abu-Rabia and Siegel (Reference Abu-Rabia and Siegel2002), for instance, tested the reading, memory, and language skills of heritage language children aged 9–14 whose home language was Arabic and whose language of school instruction was English but who also learned to write Arabic in a heritage language school program. The researchers found significant correlations between reading skills, syntactic awareness tests, and working memory tests, indicating that these qualities were not language specific. Geva and Siegel (Reference Geva and Siegel2000) tested children in grades 1–5 learning to read concurrently in their L1 English and L2 Hebrew. General intelligence tests, L1 and L2 memory tasks, and word recognition and pseudoword recognition tasks were administered in both languages. Profiles in word and pseudo-word decoding indicating steeper development in English than in Hebrew along with orthographic-specific decoding errors supported the script dependent hypothesis. The researchers, however, argued for a complementary relationship between these two hypotheses, as individual differences in memory explained a small portion of the variance between L1 and L2 reading measures.

Geva, Wade-Woolley, and Shany's (Reference Geva, Wade-Woolley and Shany1997) longitudinal study of children learning to read simultaneously in L1 English and L2 Hebrew during grades 1 and 2 found that the speed and accuracy of letter naming, word reading, and reading of these words in text across languages was highly correlated, though in the early stage of language acquisition, text reading was not more efficient than reading of isolated words. The researchers believe that orthographic depth and morphosyntactic complexity may interact with L2 proficiency effects in determining L2 reading development, and that the steps associated with L1 reading efficiency (accuracy before speed) may apply to L2 word recognition development although these steps may not emerge concurrently.

Other studies (Leikin, Share, & Schwartz, Reference Leikin, Share and Schwartz2005; Schwartz, Laikin, & Share, Reference Schwartz, Leikin and Share2005; Schwartz, Share, Leikin, & Kozminsky, Reference Schwartz, Share, Leikin and Kominsky2008) were conducted comparing performance on a number of reading measures among immigrant first grade Russian speaking children learning Hebrew bilingually, and immigrant first grade Russian speaking children who were also literate in Russian. The studies also included a sample of monolingual/monoliterate Hebrew-speaking children. One study (Leikin, Swartz, & Share, Reference Leikin, Schwartz and Share2010) found that (a) the bi-literate bilinguals outperformed mono-literate bilingual and monolingual peers on measures of phonemic awareness, word identification and pseudoword decoding accuracy; (b) Russian speaking children not literate in this language were indistinguishable from monolingual counterparts on Hebrew phonemic awareness, decoding, and spelling skills; and (c) mono-literate bilinguals were inferior to their bi-literate peers on Russian phonemic awareness measures. The researchers concluded that Russian literacy impacted both Russian (L1) and Hebrew (L2) literacy skills and highlights the relationship between literacy acquisition and explicit phonemic awareness, with transfer involving the interplay between Russian syllabic complexity, the nature of the Russian script, and the training that typically accompanies the onset of literacy acquisition in this language. Phonological processing skills were also found to be correlated with word recognition skills in English and Arabic among Canadian-Arabic bilinguals (Abu-Rabia & Siegel, Reference Abu-Rabia and Siegel2002).

TOWARDS BEST PRACTICE

In his review of research-based implications for L2 reading, Grabe (Reference Grabe2004, Reference Grabe, Hong and Anderson2009) cites several areas that need to be considered in L2 reading instruction: (a) ensure fluency in word recognition, (b) emphasize the learning of vocabulary, (c) activate background knowledge, (d) ensure acquisition of linguistic knowledge and general comprehension, (e) teach recognition of text structures and discourse organization, (f) promote development of strategic readers rather than mechanical application of strategy checklists, (g) build reading fluency and rate, (h) promote extensive reading, (i) develop intrinsic motivation for reading, and (j) contribute to a coherent curriculum for student learning. Grabe states that there is empirical support for each of these implications, although at the same time he emphasizes the need for additional research to further identify aspects of effective L2 reading instruction in particular settings. It is hoped that this article has provided some of this additional research so as to highlight that while these areas apply to reading in the languages covered in this review, orthographic issues particular to these languages require unique awareness and attention on the part of their educational practitioners.

INTRODUCING CHARACTERS

For example, one enduring controversy among Chinese and Japanese pedagogues is when the native scripts should be introduced to learners. Because learners are unable to determine the pronunciation of Chinese characters, they begin their study of Chinese by learning vocabulary through romanization, or systems that present the language using the roman alphabet, with diacritics to represent the tone of the word. Although there are a number of systems in use, pinyin has become the romanization system of choice for most textbooks used in the United States. As an example, the word “friend,” is a two-character word written 朋 友 in Chinese, and is presented as péngyou for beginning learners in pinyin, with the diacritic over the first syllable indicating the rising tone. One school of thought contends that it is best to introduce vocabulary words, written dialogues, and spoken drills in pinyin until learners acquire a significant amount of spoken language through this medium, thereby giving the student a firm base of spoken proficiency upon which to associate the written characters, much the way a child learns to read after learning to hear and speak the spoken language. As a learner's proficiency advances, new vocabulary will be introduced in both characters and pinyin, but reading texts will be exclusively written in Chinese characters. An opposing view is that students should be introduced to characters immediately, as characters are a motivating factor for learners of Chinese, and represent the Chinese language in the most authentic way. The concern, of course, is that overwhelming the student with characters is demotivating and so labor intensive that Chinese character learning becomes the primary purpose of the course, resulting in dispirited and demoralized students not learning to speak the language who end up dropping the course. (For more discussion on the use of romanization in learning Chinese and Japanese, see Christensen & Warnick, Reference Christensen and Warnick2006; Nara, Reference Nara, Nara and Noda2003; Unger, Reference Unger and Nara2001; Xing, Reference Xing2006).

More recent theory pointing to the need to develop a metalinguistic understanding of the linguistic structure of the language that the new script will map onto, as well as learning the properties of the script itself, seems to call for a firm foundation of spoken language to be taught to the learner. This has led some to temper the enthusiasm for exclusive use of authentic texts to beginning learners, recommending that pedagogical reading materials designed around what students can say in the spoken language also have their place (Everson, Reference Everson, Everson and Xiao2009). Encouraging in this regard is the study by Hitosugi and Day (Reference Hitosugi and Day2004) that demonstrated that through the use of Japanese children's books, even beginning learners of Japanese can make progress in reading in as little as ten weeks through a structured reading program employing texts that were well within the proficiency range of the learners, while also improving learners’ attitudes towards the learning of Japanese.

In Japanese, there has also been controversy about when to introduce the kana scripts as well as the kanji, the Chinese characters. Japanese learners are aided by the fact that any words in Japanese can be written in hiragana or katakana, two systems that have a relatively small number of symbols and can be learned by students in a relatively short time. Therefore, vocabulary can be introduced to students in the native scripts of hiragana and katakana almost immediately, and kanji can be introduced at the teacher's discretion. A study in Chinese (Packard, Reference Packard1990) compared two groups of Chinese language learners in the same program, one which learned Chinese characters immediately, and one which started learning characters after a lag of three weeks. The lag group did better on tests of phonetic discrimination after one semester, and measures of oral production after one year. A similar study in Japanese (Hatasa, Reference Hatasa2002) compared one group that learned the kana scripts immediately, to a group learning Japanese first in romaji and introducing kana scripts eight weeks later. In this study, there was no significant difference in terms of achievement between the groups. In a similar study, Dewey (Reference Dewey2004) found that how students value the introduction of the writing systems is related to how the teacher views this issue, thus indicating a loyalty to the teacher influencing the learning views of novice learners. In-service and pre-service training of teachers, therefore, will need to highlight these findings so that teachers can learn to develop a sense of balanced instruction between learning to speak and learning to read in these languages.

TEACHING THE COMPONENTS OF CHINESE CHARACTERS

In line with this recommendation is the amount of research already cited in this review that suggests that learning the structural properties of Chinese characters facilitates a learner's word recognition development and ultimate reading achievement. This impacts a number of the categories presented by Grabe (Reference Grabe2004, Reference Grabe, Hong and Anderson2009) such as fluency in word recognition, and enabling the learner to acquire vocabulary. More specifically, the research among foreign language learners of Chinese indicates that they develop an appreciation of the component parts of Chinese characters as aids to character and word recognition early on in their learning experience (Ke, Reference Ke1998; Taft & Chung, Reference Taft and Chung1999; Wang, Liu, & Perfetti, Reference Wang, Liu and Perfetti2004) with students reporting their heavy reliance upon this strategy (Shen, Reference Shen2005; Shen & Ke, Reference Shen and Ke2007). This ability to use the phonetic and semantic elements of the Chinese character seems to advance rapidly among beginning learners, and is used as an aid for both recall and inferencing unknown characters (Jackson, Everson, & Ke, Reference Jackson, Everson, Ke, McBride-Chang and Chen2003; Shen, Reference Shen2005). Perhaps because of their limited control of the spoken language, and consequent inability to access phonological properties of the characters (Wang, Liu, & Perfetti, Reference Wang, Liu and Perfetti2004), the ability to exploit the semantic elements of characters develops before an ability to attend to the phonetic components of characters, although learners demonstrate a need to recover the pronunciations of character when they read (Everson & Ke, Reference Everson and Ke1997; Lee-Thompson, Reference Lee-Thompson2008), identify (Everson, Reference Everson1998), and learn characters (Winke & Abbhul, Reference Winke and Abbuhl2007).

For best practice, then, we must ask about the best way to present Chinese character instruction in both our classes and in our learning materials so that students can develop strategies that facilitate both the initial learning and use of Chinese characters in the reading process. As well, given the variable nature of students ability to actually use these components successfully, it appears this is no longer nice-to-know information to be taught from a cultural or historical perspective, but information that must be woven systematically into the fabric of Chinese reading classroom pedagogy.

A number of research studies suggest that how these principles are taught can make a difference in how students learn characters, with some researchers (Shen, Reference Shen2004) recommending that class time be used to introduce characters by highlighting their etymological background and the various components making up the characters. This leads to the learner appreciating and exploring characters in a concept-driven way, an approach that may be more powerful for retention than rote memory-type strategies, strategies still used by students and advocated by some teachers. It should also be noted that technological advances that provide the capability to animate characters, rapidly deconstruct the characters into their component parts through animation and the use of color, and supply the pronunciation through audio files, can highlight etymological and character components more as primary information and less as historical or background information, as has been done with the teaching of characters through traditional textbooks (Fan, Reference Fan2010). Textbooks, then, should include more of this information so that students can begin to view characters as forms not composed of random strokes, but instead as a principled orthographic system.

It should be noted, however, that much of our research involves beginning learners of Chinese, a more plentiful subject pool of students just beginning to learn characters. While initial results indicate the need to continue teaching strategies that build upon our learners’ proficiency as they move into more advanced levels (Chang, Reference Chang, Everson and Shen2010; Warnick, Reference Warnick and Nara2001), more research needs to be conducted to understand the reading strategies and perception of character components by advanced learners. Indeed, Koda (Reference Koda and Nara2001) has argued that perhaps a better way to view this complex issue is not to concern ourselves with when to introduce characters, but instead to determine how to enhance the learning of Chinese characters for learners at different levels of proficiency.

JAPANESE KANJI LEARNING: CHARACTER COMPONENTS, STRATEGIES, AND BELIEFS

Investigating the role of the component parts of Japanese kanji is as central to the Japanese L2 research and pedagogical community as it is to those who research and teach Chinese. Kanji present different challenges to Japanese learners, for although they often present the same semantic meaning as their Chinese counterparts, they can be pronounced in multiple ways, and unlike the general monosyllabic pronunciation of characters by speakers of Chinese, they can be pronounced with more than one syllable. Yet, there are some similarities on what the research has revealed concerning learners of Japanese that are similar to the findings from Chinese learners.

Researchers have found, for example, that learners in some studies prefer a very rote approach to memorizing characters, and that their strategic orientation to learning characters is related to their beliefs linked to their perceptions of kanji (Mori & Shimizu, Reference Mori and Shimizu2007), and even to the beliefs of their teacher (Shimizu & Green, Reference Shimizu and Green2002). Different types of kanji are perceived by learners to be easier to learn than others (Yamashita & Maru, Reference Yamashita and Maru2000), with learners attending to basic construction principles and meanings before sound (Toyoda, Reference Toyoda2009). While beginning students can be taught the basics of inferencing unknown characters (Toyoda, Reference Toyoda2000), inferencing kanji is thought to be a complex process involving not only the component elements of the kanji, but also the use of the character's surrounding context (Mori, Reference Mori2002, Reference Mori2003; Mori & Nagy, Reference Mori and Nagy1999). Given these findings where inferencing seems to require a balance between use of the context as well as the morphological properties of the Chinese characters, best practice calls for reading exercises that develop in the learner a heightened awareness of the different sources that are available in a text that will help them construct meaning as they read, thus pointing to Grabe's (Reference Grabe2004, Reference Grabe, Hong and Anderson2009) categories of text structure and discourse organization as important areas of best practice. Research employing think-aloud protocols among advanced learners (Everson & Ke, Reference Everson and Ke1997) indicates they employ a highly orchestrated and fluent system of exploiting both linguistic and contextual elements when reading text and solving word identification problems. Instead of trusting that these abilities will develop in our students on their own, best practice must reserve a place in the curriculum for explicitly developing these skills through exercises that teach the importance of contextual information in reading, as well as teaching our learners that morphological and phonological information derived from the characters themselves are potentially useful in retrieving meaning.

It seems as well that there is a complex relationship between attitudes towards learning Chinese characters and the strategies students employ to learn them, with research evidence suggesting that positive attitudes and knowledge about characters interact in a powerful way to inspire learning. The evidence also suggests that teachers serve as powerful purveyors of both language and culture, and can influence students’ views that their difficulties in learning kanji are not because they are somehow ill-equipped for the learning task which is somehow beyond them (Mori, Sato, & Shimizu, Reference Mori, Sato and Shimizu2007). The classroom, then, should also be a place where teachers take an active role in getting the students to examine and develop their learning strategies (Gamage, Reference Gamage2003), with the role of the teacher also being as a builder of student confidence in addition to one who imparts content knowledge. This expanded role for the teacher of developing strategic readers (Grabe, Reference Grabe2004, Reference Grabe, Hong and Anderson2009) will then also need to be included as part and parcel of pre-service and in-service teacher development so as to increase teacher awareness of their multi-faceted role in the classroom when teaching reading. As well, best practices should include being aware of our students’ attitudes, as there seems to be connections between how our students view themselves as readers and their interest in reading in Japanese (Kondo-Brown, Reference Kondo-Brown2006a), indicating that getting to know our students and how they feel about learning these languages and what they believe the difficulties are will be helpful to remedy these problems.

It should also be emphasized that with all of the attention that is paid to the teaching of characters in the Chinese and Japanese curriculum, the ultimate goal of our reading programs should be to where our learners actually get to the point to where they can read Chinese and Japanese texts. All too often, teachers run the risk of focusing their reading instruction on selected aspects of reading without providing and guiding their students with opportunities to read extended texts. In other words, best practice dictates that students be provided opportunities to read more materials that are not far above their proficiency level, thus pointing to Grabe's (Reference Mori, Sato and Shimizu2004, Reference Grabe, Hong and Anderson2009) recommendation that students be given plentiful opportunities for extended reading. All too often, our learners spend too much time in translation or other tasks that do not require them to actually read, but often decode their way through densely written texts containing far too many unknown words.

While there is rightfully an emphasis placed upon our students learning to read authentic materials, learning to read in different orthographic systems requires pedagogical interventions that take into consideration the learners’ relative unfamiliarity with the new orthography they are learning, as well as the language system that underlies it. It has been noted that Hitosugi and Day (Reference Hitosugi and Day2004) devised an extensive reading curriculum for second semester university students of Japanese that employed books designed for Japanese children that started with texts employing the kana systems of Japanese writing, and slowly progressing to texts employing kanji. An integral part of their program was explaining to the students the purpose for extensive reading, and to give them the opportunity to read texts that were specifically designed to be at their reading level. In this particular program, evidence suggests that those participating in this extensive reading exhibited gains in reading proficiency, as well as positive attitudes towards reading in Japanese. Such pedagogical interventions hold promise for remedying some of the problems typically associated with reading in Chinese and Japanese—a lack of automaticity when processing both kana and Chinese characters, an inability to develop reading fluency, and opportunities to read texts within one's proficiency level so as to develop the beginnings of the stamina necessary down the road if one is to read authentic materials for longer periods of time. Clearly, if our goals dictate that we want to take our learners beyond the intermediate levels of reading, our classrooms will need to provide opportunities for our learners to develop into extensive readers, with pedagogy designed to take into consideration their current level of reading proficiency.

THE ROLE OF WRITING

A final word should be given to the role of requiring students to learn to write Chinese characters from memory. It should come as no surprise that research among foreign language learners of Chinese indicates that there is a strong relationship between the ability to recognize characters and the ability to write them from memory (Ke, Reference Ke1998). Yet, a high price is paid in time and effort to attain this proficiency, with evidence indicating that for learners of Japanese at the end of their second year of study, there is a considerable gap between learners’ ability to recognize and produce kanji (Chikamatsu, Reference Chikamatsu, Cook and Bassetti2005). Moreover, with the proliferation of affordable Chinese and Japanese word processing programs that show promise in helping learners to be more accurate writers (Chikamatsu, Reference Chikamatsu, Cook and Bassetti2005), and the need to handwrite beyond basic character formation being questioned as unauthentic, some have dismissed this part of the curriculum as being a waste of time (Allen, Reference Allen2008).

While the importance of being able to reproduce characters correctly with proper balance and stoke order is not questioned, an emphasis on writing characters from memory by hand may be reserved for students whose professional needs dictate the development of this special ability. Yet, it should be noted that the idea of abandoning handwriting will most likely be contested among native teachers of Chinese, as the writing of characters is not only viewed by them as useful for communication, but as a gateway into Chinese and much of Asian culture, where people have been learning Chinese characters this way for thousands of years. Or as Walker (Reference Walker and Lepke1989) has pointed out, for native teachers of languages such as Chinese, Japanese, and Korean, “. . . the writing system often embodies the very image of their language” (p. 119).

An important cornerstone, then, is for teachers and program directors to be aware of how best practice addresses the pedagogical cultures that reside in the minds of teachers who come to the classroom as native or non-native speakers of the language. Current reading pedagogy stresses that we strive to make our learners strategic readers by developing a complex array of skills that goes beyond a curriculum that spends inordinate amounts of time having our learners memorize how to write characters from memory. With time as a resource that no teacher can afford to squander, teachers will need to reevaluate the goals of their particular programs to determine the extent to which writing characters from memory will be addressed.

HEBREW AND ARABIC

Lastly, research in learning to read in Arabic and Hebrew takes us to a somewhat different dimension in L2 learning, as much of the research investigates the reading development in these languages among children. In this regard, the bilingual context of this learning indicates a complex relationship between L1/L2 language proficiency, orthographic depth with vowelled and unvowelled text, learner age, and contextual settings of language use and instruction (Saiegh-Haddad & Geva, Reference Saiegh-Haddad and Geva2010). In her synthesis of research, Geva (Reference Geva, Tremblay, Barr and Peters2006) asserts the need to give L2 learners sustained and systematic opportunities to develop proficiency in their L2 oral proficiency, while not overlooking the fact that students experiencing difficulty may be having problems in the acquisition of basic literacy skills. As with Chinese and Japanese, this need also falls within a number of categories referred to in Grabe's (Reference Grabe2004, Reference Grabe, Hong and Anderson2009) synthesis, especially the need to ensure the acquisition of linguistic knowledge. The importance of vowelled texts in Hebrew and Arabic for beginning learners also draws upon the need for developed language resources among our learners, not only to take advantage of their experience with alphabetic reading if their L1 is represented by alphabets, but also to facilitate and harness the power of phonological processing derived from more explicit phonological representation provided by marking vowels.

As the demand for opportunities to learn Arabic increases, there are some in the Arabic teaching profession who support moving towards proficiency-based instruction, yet note the challenges accompanying an effort to teach a diglossic language where a literary-based written language (MSA, or Modern Standard Arabic) and a host of Arabic spoken dialects reside side-by-side. In its attempt to move towards standardized instruction of Arabic, members of the Arabic teaching community in the United States have written that to prepare learners of Arabic, “teachers of Arabic disagree on whether their students should be taught only MSA or a dialect or a mixture of both. . .”, though there is “undisputed agreement by teachers and scholars that a thorough understanding of MSA is important” (Standards, 2006, p. 115). There is a likelihood, then, that Arabic reading programs will be standardized in teaching their students to read in MSA, though how and what variant of the spoken language other than MSA will be taught (if at all) might vary from program to program. For those interested in how language fields develop and meet the instructional challenges facing their members (Brecht & Walton, Reference Brecht and Walton1993), Arabic will provide a compelling case study for evaluating the evolution of best practice.