Alphabetic writing systems, though ubiquitous are not universal. Non-alphabetic writing systems are not, however, anachronistic and, suggest Rayner et al. (2012,) add to our understanding of the alphabetic principle.
Writing systems in China developed about 3500 years ago and, although logographic, Chinese today is not a picture writing system. Around 8000 characters are used in writing modern Chinese (Martin, 1972) as the language does not use the inflected forms that create many more words in English. Furthermore, a Chinese character represents a morpheme and a syllable simultaneously rather than a word, and characters fall into six different classifications with only 5% being simple pictographs and the majority being semantic-phonetic compounds (Martin, 1972) which indicate pronunciation and meaning simultaneously. The characters are not, however, the smallest unit as these may contain semantic radicals which enable characters to be built with more economy as well as phonetic radicals to indicate pronunciation. It would be wrong then, according to Rayner et al. (2012), to categorise Chinese characters as pictures as most characters are not images of morphemes but contain indications of pronunciation thus embedding the principle of sound coding across much of the language. They conclude that it would, therefore, appear that even this logographic language applies some of the principles of sound to symbol correspondence.
The development of the early reading of Chinese further enlightens the pedagogy of sound to symbol correspondence in an alphabetic system as it lacks the latter’s letter-sound mapping organisation. Chinese characters do not explicitly encode pronunciation with only an element of the sound cue indicated by the phonetic radical. This means that the character name and pronunciation must be memorised with about 3000 characters expected to be memorised by the end of primary school (Sheridan, 1981). It is the requirement of this huge memory feat of a script ‘suitable only for an elite’ (Mair, 2007:2) that Unger (1977) suggested was the cause of such high levels of illiteracy in China. Teaching, Unger (1977) noted, was reliant on memorisation of not only characters but whole passages such that he quotes Mao as likening reading lessons to ‘stuffing students like Peking ducks’ (1977: 5). Flash cards, choral reading, visual aids and charts were used to aid memorisation mirroring the word method of the teaching of the reading of English throughout much of the twentieth century. The complexity and difficulty of learning 3000 Chinese characters inherently undermines the learning of English’s 100,000-word vocabulary through the shape of words (Dehaene, 2010). Crucial, was the introduction of the Chinese Phonetic Alphabet (Pinyin) in 1958 with 21 consonants and 35 rimes which uses the Roman alphabet letters to transcribe the exact pronunciation of a character including its lexical tone.
Although Chinese characters are not eschewed as part of the teaching of reading, children learn the Pinyin alphabetic coding system in the first few months of instruction to bridge the gap between characters and pronunciation (Sheridan, 1981). By the end of the fourth grade, pupils are expected to be able to read and write Pinyin moving to characters alone by the sixth grade (Sheridan, 1981). The alphabetic principle of Pinyin enables pupils to identify unfamiliar characters by reactivating spoken sound knowledge and linking it to the character (Lin et al., 2010) and indicates, according to Perfetti and Tan (1998) that phonology is activated in word recognition across writing systems, even those that do not operate with an alphabetic principle. Lin et al. (2010) found that for Chinese children aged six, mastery of Pinyin was the clearest predictor of word reading skills the following year. Although the dramatic improvements in China’s literacy rates since the introduction of Pinyin (Peterson, 1994) cannot be entirely ascribed to the use of the alphabetic principle, in bridging the gap between phonology and logography these data, suggest Rayner et al. (2012), offer converging evidence for a universal phonological principle in reading.
Although the Japanese language uses characters derived from Chinese in its written construction, the two bear little relation as Japanese makes extensive use of syllabaries. However, with only 100 syllables, most morphemes are more than one syllable, and most words are polysyllabic. Thus, characters could either represent morphemes or syllables but not both as in the Chinese system (Rayner et al., 2012). What evolved was a hybrid of the two, with Kanji characters representing morphemes and simplified Kana characters representing syllables (Taylor, 1981). However, this is further complicated by Kana being divided further into two different forms serving two different purposes: Katakana, which represents words borrowed from other languages and Hiragana to represent prefixes, suffixes and function words. The system is made even more intricate by the inclusion of occasional Roman letters and Arabic numerals (Taylor, 1981).
What is illuminating, is that despite this convoluted, hybrid system of writing, reading rates (speeds) in Japanese are comparable to those in English (Ikeda and Saida, 1978) and that literacy rates in Japan are amongst the highest in the world (OECD, 2015) suggesting that learning to decipher the apparently convoluted Japanese writing system is no more demanding than learning an alphabetic one and perhaps less (Rayner et al., 2012). It has been argued by Rozin and Gleitman (1977) that the lower reported rates of reading problems in Japan are a result of Japanese reading instruction initially focussing on the syllabic symbols of Kana prior to embarking on the Kanji symbols. They go on to postulate that using a syllabary might be the best way to introduce children to reading in cultures with alphabetic languages. However, they ignore the fact that Japanese has far fewer syllables than western languages and that Japanese can in principle be written entirely using Kana making it a highly efficiently structured system complicated by its hybrid nature. Furthermore, they fail to account for the effects of Japan’s standardised teaching model and standardised Course of Study along with Japan’s far greater equity of schooling and social structure (Murata and Yamaguchi (2010).
In contrast, Korean, which has nearly a thousand syllables, utilises characters which represent the morphemes as well as an alphabetic system that represents phonemes and articulatory features which when blocked together forms a syllabary. What is intriguing is that it is not an evolved system, but one created by scholars in the fifteenth century who, having studied both logographic and alphabetic systems, specifically designed this combined system. Whether this system is an improvement over alphabetic systems is unclear according to Rayner et al. (2012); however, Korea reports very high levels of literacy and low rates of reading problems (OECD, 2015). Again, like Japan, it has a standardised national curriculum and far greater equality of provision and has the highest levels of its population educated to tertiary level (OECD, 2019).
Whether a language is encoded using a logographic, alphabetic or hybrid writing system, what is evident is the importance of phonological cues to assist rapid reading and the instant recognition of words. What is also clear is the assistance these cues afford to the relief of memory. No system can rely on the memorisation of shapes alone - vocabularies are just too big. Thus the pedagogy of word shape recognition, without reference to phonological cues, is fallacious. What is perhaps most remarkable is the flexibility of the human brain to automatise the decoding of these phonological cues to such an extent that 300 words per minute can be read in any of the systems. What is equally remarkable is the assumption by some that this skill does not require explicit instruction by highly trained pedagogues.