Ghoti: Orthographic Literacy in Focus: Discussion by Samejon (2015)

There is a noticeable advantage then for emerging L2 or L3 literates in accessing novel material if they are adequately exposed in the TL or L2. Among these exposures, the ability to distinguish words is highly considered to be a product of ‘self-teaching’ since phonological awareness cannot satisfy the bounds of beginning learners (Cunningham, et al., 2002). This notion was affirmed by the study of Pae, Sevcik, and Morris (2010) when they recognized that phonological awareness is a good predictor of reading success but as the reading task grows more in complexity automated recall facilitates fluency. With the script-dependence hypothesis, we can ascertain that in reading and word choice drills, GPC familiarization of a language contributes to fewer errors. These errors, however, are orthography-specific, which is very telling of what cognitive process may be at work in varying orthographic depth. This consideration would positively lead to modeling a more efficient orthographic literacy. Linguistic transfer processes also surfaced along withour survey of hypotheses. Linguistic and orthographic proximity hypothesis exemplifies the linguistic aspect of orthography acquisition as the proponents identified transfers in phonology and shared linguistic constraints between two orthographies. This reminds us of Cummins’ cognitive academic language proficiency but with a greater curve to literacy and orthography. Meanwhile, Schwartz, Kahn-Howitz, and Share (2014) comprehensively knit altogether the aforementioned variables. To put it simply, the case of emergent Circassian English literates demonstrated extensive use of prior literacy experience and execution of cognitive repertoire (p. 43) in advancing literacy in the English language. This study duly affirmed the script-dependence hypothesis (Geva & Siegel, 2000) and linguistic and orthographic learning hypothesis (Kahn-Horwitz et al., 2011).

Apart from the works highlighted above, the studies surveyed for this paper have given us several emphases. These emphases, though varying, still yielded to a whole and more meaningful understanding of the constructs of orthography in literacy.

Significantly, we came to understand that mapping principles, as in phonology and morphology, are different across orthographies (Wang, 2005) and though this may be valid, cognitive processes enable predictive literacy development via sound-to-symbol identification, manipulation, and fluency of retrieval (Caravolas, Lervåg, Mousikou, Efrim, Litavsky, Onochie-Quintanilla, Salas, Schöffelova, Defior, Mikulajova, Seidlová-Málkova, & Hulme, 2012). While there are observable limitations to linguistic transfer, we can see that cognates and shared alphabetic features may facilitate literacy in the TL (Deacon, Wade-Woolley, & Kirby, 2009). Also, this shared orthographic function aids sound perception to the early stages of learning (Pytlyk, 2011). Moreover, literature on the segmented phases of cross-orthographic literacy (Seymour, 2006) provided future direction to the role of instruction time in teaching deep or shallow orthographies (Zaretsky, Kraljevic, Core, & Lencek 2009; Everson 2011). In sequential literacy, a universal thread in phonology and writing can be attested whatever their orthographic depth is (Rose, et al., 2010; case of Korean-American). By and large, studies similarly echo the importance of exposure to the target orthography for the development of strategies in literacy (Matsumo, 2013; Geva & Siegel, 2000; Cunningham, et al., 2011; Schwartz et al., 2014). With the hope to unify the studies in cross-orthographic literacy, Frost (2012) attempted to establish a universal model for literacy (cf. Seymour, 2006). This is to overarch previous studies and promote a universal model of reading primarily harnessing frequency of letter distribution and cues in facilitating acquisition. In a practical sense, this statistical direction is a natural cognitive strategy for 'resource saving' (p. 274).

            This field is still developing and with present methodologies results of several theories converged. With this, perhaps the recent study of Goodwin, August, and Calderon (2015) must be mentioned. The study utilized several theories including grain-size theory and the ODH, among others. They drew our attention to two things: (a) the actual processing of an orthography, in this case Spanish as L1 and English as TL, via large grain, i.e., morphology, such as affixes, and small grain, sound bits or phonology; and (b) the role of teaching instruction to this process. Firstly, this implies that emerging literates process the TL in larger chunks and eventually shifts to the phonological tasks of the orthography, i.e., small grain. More specifically, in a more transparent orthography, granulation tends to be processed via small grain while in deep orthography via large grain. And, secondly, instructional support in the underlying process of acquisition directly contributes to cohesive TL processing (p. 623). Language instruction used by teachers then should not be arbitrary but in itself dedicated especially to beginning and emerging literates in their formative years.

Ghoti: Orthographic Literacy in Focus: Acquisition by Samejon (2015)

The ODH has given us a primer on the concepts of how various orthographies are rendered. On the same note, unlocking these conventions across orthographies, not only on how they are written but also as to why they are written in such manner, is matter of interest to literacy studies. Studies in the field of literacy have grown from understanding reading and writing abilities to a contextualized phenomenon involving social models and cultural dimensions (Warriner, 2013, pp. 530-532). The latter is the observable social turn in the study of literacy and literacies, whose concepts are not examined within this paper. Considering our present aims, the former definition and the subcomponents/skills of literacy, i.e., phonological awareness, decoding, fluency, comprehension and vocabulary (Li et al., 2014) will prove to be useful as a working definition. This premise leads us to consider three hypotheses that are instrumental to unlocking literacy irrespective of orthographic variances: self-teaching, script-dependence, and the linguistic and orthographic proximity.

The self-teaching hypothesis rallies that that phonological processing, though vital, is not enough for orthographic learning (Cunningham, Perry, Stantovich, & Share, 2002) that is, in forming, storing, and accessing orthographic components. Therefore, already existing orthographic knowledge facilitates maximum decoding ability rather than simply relying on phonological process as seen in emerging L2 literates. They (Cunningham et al., 2002) tested the hypothesis in a deep orthography, English, from the initial testing via Hebrew (Share, 1999). This time, general cognitive ability in recall and orthographic discrimination ability was considered.  Interpretation of data shows that in reading and orthographic choice tasks, orthographic knowledge relies on previous success of exposure (Cunningham, et al., p. 196; see Share, 1999, methodology), thus, ‘self-teaching’ happens and a faint reliance on phonological processing is observed.

The second hypothesis in consideration is the script-dependence hypothesis. It posits that variance between L1-L2 orthography possibly contributes to the ease of word decoding accuracy, which is due to cross-linguistic transfer (Geva & Segiel, 2000). In a study to prove such claim, Geva and Siegel (2000) observed English L1 grade students acquiring Hebrew as L2. They found those favorable circumstances including ample exposure and language of instruction, and the lesser demand in Hebrew orthography allayed L2 acquisition more effectively than the students’ L1. Cross-linguistically, they argue that once familiarity in the L1 GPC rules is acquired, it approximates to the L2 up to the point where sub-skills and strategies need to be developed.

Sensing the need for a more ‘linguistic’ approach came the linguistic and orthographic proximity hypothesis. This hypothesis is conceptualized by Kahn-Horwitz, Schwartz, and Share (2011), who, when having tested script-dependence hypothesis, discovered the challenges faced by emerging triliterates in decoding short vowels and consonant clusters in English as the TL. It was first thought that cognitive abilities such as phonological awareness and working memory are independent of specific language experiences (p. 138). This study, however, found that these two interplay with the orthography of the language being acquired. Novel phonemes and the observed variations in (a) writing the sound clusters /ʃ/ and /tʃ/ as well as short vowel and (b) silent ‘e’ reading cannot just be accounted by script dependency. Concerning literacy, they went on to conclude:

[…] that the specific advantage of emerging Russian–Hebrew- speaking triliterates on short vowel decoding and spelling is a result of the similar way the Russian and English scripts graphically represent vowels and not due to a triliteracy experience per se. (p. 152)

More attention is then attributed to cross-linguistic processes than literacy transfers, as in this case. They further invited researchers to expand this hypothesis by examining languages that have phonemic properties in common but greater writting system differences and vice versa.

Ghoti: Orthographic Literacy in Focus: Representation by Samejon (2015)

Orthography is sometimes confused with writing system and script. Consequently, this section will start by distinguishing these three terms.

There are two types of writing systems: morphographic, which is logographic, and phonographic, which is phonemic. Along the meaning-to-sound continuum, a morphographic writing system represents idea plus sound in their written characters, i.e., Mandarin. The second type, phonographic, consists of subtypes such as: syllabic or moraic, which is written according to syllable vowel plus consonant, e.g. Devanagari; alphabetic is according to phoneme vowel or consonant, e.g., Greek; abugida is according to phoneme vowel diacritic or consonant; and abjad is according to phoneme consonant only, e.g. Arabic (Peckham, 2015; Li, McEntee-Atalianis, & Lorch, 2014).

While a writing system is particularly concerned with the written abstraction of meaning-to-symbol or sound-to-symbol identification, orthography encompasses a writing system being governed by language-specific conventions, e.g. sound and formation rules and variation (cf. Wei et al., 2014). A script then is the “alphabet” of a writing system. To this effect, Koda (2012) provided this paradigm:

To illustrate, English and Russian writing systems are both alphabetic, but they differ in their scripts—the former employs the Roman script and the latter uses the Cyrillic. English and Spanish are alphabetic, both employing the Roman script, but differ orthographically in spelling conventions. (p. 1)

            Multilingualism, being the norm in the global landscape, has brought studies of literacy to give great interest to bi-, tri-, and multi-literacy. In the process of acquiring literacy in another language, emerging literates are immediately confronted with a different orthography. As mentioned earlier, the second language (L2) or target language (TL) may share similar scripts but can be proven novel via the L2’s orthography. It is sensible to ask at this point: what then would be the bases of such variation in orthography? Frost (1994; 2012) may well give us clues through the Orthographic Depth Hypothesis (ODH).

According to the ODH (Frost, 2012), languages can be classified as having deep or shallow orthography. A language with deep orthography has a relatively distant grapheme-phoneme correspondence (GPC) where a word should be processed through their morphology. Also, a deep orthography’s letter-to-sound representation is not straightforward. In a playful note, “ghoti” may be read the way you read “fish” if English would remain straight-out faithful to the sound of “gh” in “enough”, “o” in “women”, and “ti” in “nation”. Retrieval of such sound repertoire is possible through mental awareness, or simply, familiarity of the language’s phonology. In contrast, a language with shallow orthography has strong grapheme-phoneme identification. Sample languages below will show us how these concepts are embodied in natural languages.

Chinese has limited phonemes and in order to represent their language they added tones that are codified in their writing system. This has not simply provided readers of Chinese a means for retrieving pronunciation but meaning as well. The case of Japanese is quite different. Having borrowed the writing system of Chinese, Japanese developed ‘kanji’ to disambiguate homophones and further developed hiragana and katakana. Broadly, both hiragana and katakana function as their phonographic scripts. Finnish has a very transparent orthography while English and Hebrew, on the other hand, have a deep orthography.

What we can learn from these orthographic variances is that: orthography accommodates phonology and meaning using minimal orthographic units for processing (Frost, 2012; p. 267) primarily directed to the access of a particular group. Minimal in this sense is the optimization of symbols to facilitate efficient representation and retrieval. However, this is not the only means of how a specific orthography is represented or developed. With this in mind, it is imperative to mention Cahill (2014) who explores the non-linguistic factors in the development of orthography and Sebba (2015) who traces script alternations brought by the events in a language group’s history. Nevertheless, this multimodal/social aspect of orthographic representations deviates from the intentions of this paper, thus, such views deliberately merits a separate paper.

While conventions across cultures develop specific orthography, this difference can mean that bi-, tri- or multiliteracy is something complex but, ultimately, not impossible. At present, modern reflections offer insights as to how literacy, in view of differences in orthographies, can be facilitated. The next section will deal with these developments.

Ghoti: Orthographic Literacy in Focus: Introduction by Samejon (2015)

Orthography is rendered in varying conventions across cultures.  It may not be culturally dependent but certainly carries cryptic and linguistic characteristics. How an outsider breaks through and gain access to another orthography has been the fascination that brought this paper to be. We may then say that gaining literacy in another language is actually a cross-linguistic endeavor with challenges attached. This same fascination, however, is not novel yet still to be fully understood. Orthographic development and processing had long propelled scholars on studies, which led to insights for accessing literacy in another orthography.

Following this brief introduction, this paper focuses on reviewing orthographic convention variances and orthography acquisition hypotheses, which will culminate in this paper’s discussion and conclusion.