A “Radical” Approach to Reading Development in Chinese: The Role of Semantic Radicals and Phonetic Radicals
Ho, Connie Suk-Han
Two studies investigating the significance of radical knowledge in Chinese reading development are reported in this paper. Study 1 examined the semantic radical knowledge of 20 Grade 1, 20 Grade 3, and 20 Grade 5 Chinese children in Hong Kong. It was found that various types of semantic radical knowledge, including the position and semantic category of semantic radicals, correlated significantly with Chinese word reading and sentence comprehension. Study 2 examined phonetic radical knowledge with another three groups of 20 Chinese children in Grades 1, 3, and 5 respectively. It was found that various measures of phonetic radical knowledge, including the function and sound value of phonetic radicals, correlated significantly with Chinese word reading. These studies found that, developmentally, the children started acquiring the knowledge of character structure, position, semantic category, and sound value of radicals from about Grade 1. However, they did not understand that the function of semantic radicals is to provide meaning cues in reading until Grade 3. The authors concluded that the radical is an important orthographic processing unit in reading development in Chinese.
Reading is an important means of acquiring knowledge in today’s literature world. However, not all children master this important skill without difficulty. It may be difficult for a child to learn to read if he or she treats every new word as a new visual configuration without discovering the underlying regularities of the script. Researchers like Ehri (1991, 1992) and Frith (1985) propose that children initially learn to read holistically at the logographic stage by associating some visual features of a word with its sound. Later, when children discover the “cipher” of the language system at the alphabetic stage, they solve the problem of having to memorize many different words and are able to read new words (Cough, Juel, & Griffith, 1992). What Cough et al. (1992) mean by “cipher” is the phonological regularity of a writing system (e.g., letter-sound correspondence rules in an alphabetic language). At the third and orthographic stage, letter strings and whole words are processed as whole units and this allows efficient and automatic word recognition.
In the opinion of Nagy and Anderson (1999), learning to read is “fundamentally metalinguistic.” Metalinguistic awareness is the ability to reflect on and manipulate the structural features of language. Past research has shown a strong association between metalinguistic awareness and reading development. For instance, awareness of phonemes has been found to be important in learning to read alphabetic languages. However, many studies have shown that apart from awareness of the phonological regularity of a writing system, awareness of the orthographic regularity (i.e., the spatial and sequential redundancy of orthographic units) (e.g., Berninger, 1987, 1990, 1994; Corcos & Willows, 1993) and morphological regularity (i.e., the correspondence between morphological units and meaning) (e.g., Anderson & Nagy, 1992; Anshen & Aronoff, 1988; Freyd & Baron, 1982; Nagy, Anderson, Schommer, Scott & Stallman, 1989; Tyler & Nagy, 1989; Wysocki & Jenkins, 1987) is also important in reading. Since most of these studies have been conducted with alphabetic languages, one may ask whether the same findings apply to learning to read Chinese, a non-alphabetic script. The present studies were conducted with the aim of providing an answer to this question. Before we proceed to review related literature in Chinese, we will first describe the main characteristics of the Chinese orthography.
The Main Characteristics of the Chinese Orthography
The basic graphic unit in Chinese is a character. Almost all Chinese characters represent morphemes, and all characters are monosyllabic. That is why Chinese script is often described as morphosyllabic. There are about 3,000 Chinese characters in daily use in Mainland China (Foreign Languages Press Beijing, 1989) and about 4,500 frequently used characters in Taiwan (Liu, Chuang, & Wang, 1975).
According to Shu et al. (2003), the average rate of phonological consistency for phonetic radicals in elementary grades is 64%, which is higher than that of phonological regularity for phonetic radicals. However, since the size of phonetic families (i.e., number of characters with the same phonetic radical) is small, it takes a long time for children to develop phonetic consistency awareness. There are 563 phonetic families in Shu et al.’s corpus and the average family size is 3.23. Children’s awareness of both phonetic regularity and consistency (as reflected in the use of direct derivation and analogy strategies respectively) was examined in the present studies.
Shu et al. (2003) have also suggested that children develop semantic radical awareness at an early age because of the high transparency, high frequency and large family size of semantic radicals. There are 124 semantic families in their corpus and the average family size is 14.99. One of the objectives of the present studies was to examine whether children develop the awareness of semantic radicals earlier than phonetic radicals.
The Radical as an Important Orthographic Unit in Chinese Character Recognition
A number of recent studies have shown that the radical is an important processing unit for adult skilled readers in the recognition of Chinese characters (e.g., Chen et al., 1996; Feldman & Siok, 1997, 1999a, 1999b; Li, 1997; Taft & Zhu, 1997). However, few studies have been conducted to examine systematically the role of radicals in Chinese reading development. The aim of this paper is to fill this gap.
Some researchers have reported that the position and function of radicals are the main factors that determine how a radical affects character recognition in Chinese skilled readers (e.g., Feldman & Siok, 1999a; Li, 1997; Li & Chen, 1997). We inquired whether the same types of radical knowledge are important in learning to read Chinese.
The Positional Regularity of Radicals
Many researchers have proposed that automatic word recognition in English depends on a reader having knowledge of the orthographic regularity of a script (e.g., Barker, Torgesen, & Wagner, 1992; Berninger, 1994; Corcos & Willows, 1993; Venezky & Massaro, 1979). The orthographic knowledge is based on statistical probability or rules (e.g., letter probabilities, positional frequencies, and sequential redundancy patterns). This knowledge is acquired through repeated exposure to the orthographic structure of words in text.
In the case of Chinese, the positional regularity of radicals is the crux of a character’s orthographic structure. Like the frequency effect of letter sequences in English, the positions of all the radicals in a Chinese character mostly determine whether the character is legal or not. If all the radicals in a Chinese character are in their legal positions (i.e., in positions where the radical may appear in real characters), that character may either be a real character or a pseudocharacter. If any of the radicals is not positioned legally, the character is a noncharacter. The most common way of testing Chinese readers’ knowledge of radical position or character structure is by a character decision task (e.g., Cheng & Huang, 1995; Feldman & Siok, 1997; Peng, Li, & Yang, 1997; Taft & Zhu, 1997). For instance, Cheng and Huang (1995) asked 71 children in grades 2 through 6 in Taiwan to judge whether an item had been learned before as a legal character, and, if not, to judge whether it looked like a legal character. The stimuli were frequently used characters, rare characters, pseudocharacters and noncharacters. They found that most children judged frequently used characters as real characters and noncharacters as illegal characters. The second and third graders judged rare characters and pseudocharacters as character-like at chance level. From fourth grade onward, more children judged the two types of characters as looking like characters. It appears that even second graders have some rudimentary knowledge of the orthographic structure, and this knowledge grows with age. The children have gained more accurate positional knowledge of radicals by the time they reach Grade 4.
In two other studies, one conducted by Chan and Mimes (1998) on 60 preschoolers and primary school children in Hong Kong and the other by Shu and Anderson (1999) on 143 primary school children in Beijing, it was found that even first graders could use the positional rule to reject noncharacters and judge pseudocharacters as orthographically acceptable. This suggests that basic character structure or radical position knowledge develops quite early in Chinese children.
However, it is noteworthy that the character decision tasks described above are considered as tests of knowledge of character structure and implicit knowledge of radical positions. The above findings therefore did not show anything about explicit knowledge of radical positions or specific knowledge regarding semantic radical position and phonetic radical position. The present studies examine such explicit positional knowledge and whether the knowledge of orthographic structure in Chinese, as represented in the positional regularity of radicals, is related to a child’s reading development.
The Functional Regularity of Radicals
As pointed out by Feldman and Siok (1997), the position of radicals is often confounded with its function. Feldman and Siok (1999a) reported that about 75% of Chinese ideophonetic compound characters have their semantic radicals on the left. In other words, in characters of left-right structure, most left radicals serve the semantic function and right radicals serve the phonological function. Therefore, a reader’s knowledge of the position of radicals to some extent helps him/her determine the semantic radical and the phonetic radical that is in a Chinese character.
The Phonological Function. The impact of phonological regularity in learning to read alphabetic languages has been the focus of many researchers for years. Goswami (1986) has proposed that beginning readers read words by analogy, even before they are able to decode words phonologically. Ehri and Robbins (1992) have further demonstrated that some decoding skills are needed in order to read words by analogy. Similarly, both direct derivation and analogy strategies are used in reading by Chinese beginning readers (Ho & Bryant, 1997a; Ho, Wong, & Chan, 1999).
Similar to beginning readers of English, 90 Chinese first and second graders were found to name phonologically regular Chinese characters better than irregular ones (Ho & Bryant, 1997a), and 30 Chinese third and sixth graders were found to name consistent Chinese characters better than inconsistent ones (Yang & Peng, 1997). Phonetic-related errors were also found to be the most dominant type in reading Chinese characters and words among Chinese first and second graders (Ho & Bryant, 1997a). Chan and Siegel (2001) tested 94 primary school children in Hong Kong and found that older and normally achieving students made more phoneticrelated errors, whereas younger normal and poor readers made more semantic and visual errors. These findings suggest that even Chinese first-graders start to make use of the phonetic radical for sound cues, and the reliance on the phonetic principle seems to be related to reading proficiency. However, the development of phonetic knowledge and its relationship to reading development has not been studied systematically.
The Semantic Function. Shu and Anderson (1999) have suggested that morphological awareness (i.e., the awareness of the morphemic structure of words and the ability to reflect on and manipulate the structure) is an important aspect of metalinguistic awareness that is closely related to literacy in both alphabetic and non-alphabetic scripts. The semantic radical is the orthographic unit in Chinese that encodes or specifies the meaning of a character. Similar to the findings on phonological consistency, significant semantic consistency effects were reported with adult skilled readers in performing a semantic categorization/decision task (Chen & Weekes, 1997; Leek, Weekes, & Chen, 1995; Miao & Sang, 1991; Zhang, Zhang, & Peng, 1990), a lexical decision task, and a priming task (Feldman & Siok, 1997). Only a limited number of studies have examined the awareness of the function of semantic radicals. One was conducted by Cheng and Huang (1995), who used a semantic-relatedness judgment (SRJ) task to assess Taiwan children’s knowledge of the function of semantic radicals. The task required participants to identify among three choices the character that was semantically related to the target character. The correct choice was the one that shared the same semantic radical as the target. They tested 71 Chinese children in grades 2-6 and reported that only the sixth graders used the radical as a reference for getting at the meaning of characters in the SRJ task.
Shu and Anderson (1997) assessed 220 Beijing children’s semantic radical awareness by presenting them with two-syllable words familiar from oral language. One syllable was written in character and the other was written in Pinyin (an alphabetic system used to write the sounds of Putonghua, the national spoken language in Mainland China). The children were asked to circle a character to replace the Pinyin in each word. The correct character contained a semantic radical consistent with the meaning of the two-syllable word. This task appeared to provide more contextual cues to the participants than Cheng and Huang’s (1995) SRJ task. Shu and Anderson found that both third and fifth graders used the semantic radicals to derive meaning of unfamiliar characters and recently learned characters, but the first graders did not. As compared with the findings in Cheng and Huang’s study, younger children displayed some awareness of the function of semantic radicals when given more contextual cues. Shu and Anderson also found that children who were rated as good readers by their teachers displayed more awareness of radicals than children rated as poor readers. However, they based this finding solely on teachers’ ratings without actually measuring the children’s reading performance.
The age when children appear to have acquired the knowledge of the function of semantic radicals depends very much on how the knowledge is assessed. In a creative writing task, Chan and Nunes (1998) presented six pictures of objects to 60 children. The children were asked to make up a name for each object by creating a new character with two radicals. Six familiar semantic and six familiar phonetic radicals were provided as choices for the six pictures. They found that from age six, some of the children were able to use semantic radicals to represent the meaning of a pseudocharacter. It is noteworthy that the radicals in their study were of very high frequencies. Therefore, some children seem to have acquired the function and meaning of familiar semantic radicals quite early.
In summary, the orthographic, phonological, and morphological regularities of the Chinese writing system are mainly represented in the positional frequencies, functional regularities, and consistencies of radicals in Chinese characters. Some past studies have examined the implicit knowledge of positional regularities of radicals, but explicit or specific knowledge regarding semantic radical position and phonetic radical position is seldom examined. Previous researchers have also failed to examine separately a reader’s awareness of the function of radicals and his/her knowledge of the specific sound value or semantic category of the radicals. The development of these various aspects of radical awareness and how they relate to reading development has also yet to be studied systematically. The authors addressed all these issues.
Aims of the Present Studies
Since the radical is an important unit in processing Chinese characters that relates to orthographic, phonological, and morphological regularities, the present studies were designed to examine systematically and comprehensively Chinese children’s development of radical knowledge. The aims of the studies were to investigate (1) whether Chinese children’s radical knowledge (both positional and functional) was related to their reading performance, and (2) when children acquired different aspects of radical knowledge. Two studies were conducted in Hong Kong to examine these issues: Study 1 on semantic radical knowledge and Study 2 on phonetic radical knowledge. Although a sentence comprehension test was included in Study 1, these studies focused more on word-level than text-level processing because individual differences in word-level processing skills are a major source of individual differences in text-level processing skills (e.g., Jackson & Coltheart, 2001; Perfetti, 1985; Shankweiler et al., 1999). Hong Kong children, including the participants in the present studies, are taught to read Chinese characters by the whole word approach without much emphasis on the use of the knowledge of radicals. Teachers may teach the meaning of some semantic radicals explicitly but not the sound of phonetic radicals. Students also learn to use semantic radicals to look up characters in a Chinese dictionary from about Grade 3 onwards. However, the role or function of semantic and phonetic radicals is not taught systematically.
As reviewed above, Chan and Nunes (1998) and Shu and Anderson (1999) have shown that the knowledge of basic character structure develops quite early in Chinese children. Young children seem to have a rudimentary understanding of the function and meaning of highly familiar semantic radicals as well, especially with contextual cues. However, these studies did not examine explicit knowledge of semantic radical positions, or separate a reader’s awareness of the function of radicals from his^er knowledge of the specific semantic category of radicals.
Study 1, therefore, was designed to examine comprehensively Chinese children’s development of semantic radical knowledge. Specifically, we looked at when children acquired three aspects of semantic radical knowledge: namely the position (both implicit and explicit knowledge) and function of semantic radicals and the specific semantic categories of different radicals. We also looked at the relationship between the children’s semantic radical knowledge and their reading development.
Participants. A total of 60 Chinese children, 20 each in Grade 1 (mean age = 7 years, 2 months, mean IQ = 112), Grade 3 (mean age = 9 years, 2 months, mean IQ = 115), and Grade 5 (mean age = 11 years, mean IQ = 122), were recruited from one government-subsidized ordinary primary school in Hong Kong. Parents of these children responded positively to our invitation letter and consented to allow their children to participate in the study. The kind of reading instruction they received was typical ofthat which we described in the last section. There were 10 boys and 10 girls at each grade and all spoke Cantonese. The average IQ of the fifth graders was significantly higher than that of the first graders (p
Materials and Procedures. An IQ test (the Raven’s Standard Progressive Matrices), two reading tests, and five tasks of semantic radical knowledge (see Appendix A for sample items) were administered to the children. The IQ test was included because IQ was found to correlate significantly with Chinese word reading in previous studies (e.g., Ho & Bryant, 1997b; Huang & Hanley, 1997). We developed all the test materials except for the Raven’s test and the Chinese word reading test, which were standardized test materials. Pilot studies were conducted to ensure that the test materials and procedures were appropriate. Apart from the Chinese word reading test that was administered individually, all others were group tests.
The Raven’s Standard Progressive Matrices (with local norms developed by the Hong Kong Education Department in 1986) was adopted to measure the children’s nonverbal reasoning ability and hence to estimate their intelligence. The first graders were given the short form (Sets A to C) and the third and fifth graders were given the full version (Sets A to E). In each item of the test, there was a target visual matrix with one missing part. The children were required to select, from six or eight alternatives, the one that best completed the matrix.
The Chinese Word Reading Test (developed and standardized by the Hong Kong Education Department in 1988) was used to measure the children’s word reading skills. Materials consisted of 65 Chinese two-character words of primary school levels arranged in ascending order of difficulty. The children were asked to read the words aloud one by one. The task was discontinued when the child failed to read 10 consecutive words.
The radical position judgment task measured the children’s explicit knowledge of the position of semantic radicals. Forty semantic radicals, 20 of high frequency and 20 of low frequency, were chosen as stimuli. Of these, half were left-radicals, 20% were right-radicals, 20% were top-radicals, and 10% were bottom-radicals, which approximated the distribution frequency of radical positions. In each item, the children were visually presented with two square boxes, of which one was divided into two halves horizontally and the other vertically (see Appendix A). These boxes represented left-right character structure and top-bottom character structure respectively. The children were asked to indicate the legal position of each semantic radical by pointing to the appropriate half in the boxes (left, right, top, or bottom).
The Chinese pseudocharacter meaning judgment task measured the children’s overall knowledge of the position, function, and semantic category of semantic radicals. Forty pseudocharacters with lexical or nonlexical semantic radicals were presented to the children. They needed to select, from four alternatives, a picture that might represent the meaning of these “funny” characters.
Results and Discussion
Table 1 presents the mean scores and standard deviations of the tasks given in Study 1. All the tasks had good internal reliability (all reliability coefficients > .80). Figure 1 shows the percentages of character-like responses for the four types of characters in the Character decision task. A two-way (grade χ type of character) analysis of variance (ANOVA) was conducted. The two main effects and the interaction effect were all significant (all Fs > 2.6, all ps
Apart from the first-graders’ score on semantic-relatedness judgment, scores on all other tasks that involved force-choice responses were significantly different from chance (all ts > 5.2, all ps
In both the semantic category judgment task and the pseudocharacter meaning judgment task the children at all the grade levels judged the semantic category of the lexical radicals better than the nonlexical ones (all ts > 3.23; all ps
Regarding the semantic transparency of the choice characters’ semantic radicals in the Chinese sentence comprehension task, first graders (transparent: 10.5, opaque: 10.8) and third graders (transparent: 14.2, opaque: 13.8) performed similarly on the transparent radical items and the opaque radical items. However, the fifth graders performed significantly better on the transparent items than the opaque ones [transparent: 14.7, opaque: 13.9, t (19) = 3.56,p
Correlation coefficients were calculated to examine the importance of various types of semantic radical knowledge for Chinese word reading and sentence comprehension. Table 2 presents a matrix of partial correlation coefficients among the various measures after controlling for the effects of age and IQ. It was found that character decision, semantic category judgment, and pseudocharacter meaning judgment correlated significantly with both Chinese word reading and sentence comprehension (all rs > .25, all ps
Pseudo-character meaning judgment was found to correlate significantly only with Semantic category judgment (r = .42, p
The semantic-relatedness judgement failed to correlate significantly with any of the reading or meaning judgment measures. This finding might be partly due to the fact that the first graders did not score significantly above chance level in this task. In addition, the finding may also suggest that knowledge of the function of semantic radicals may be less important than knowledge of their semantic categories in reading Chinese words and sentences.
Study 2 was designed to examine Chinese children’s development of phonetic radical knowledge. Although information about pronunciation is represented less systematically in Chinese than in other orthographies, various studies have shown that children do rely on phonetic radicals for sound cues, even sometimes when phonetic radicals only provide partial information to pronunciation (e.g., Anderson et al., 2003). As in Study 1, we looked at when children acquired three aspects of the phonetic radical knowledge (the position, function and sound value of phonetic radicals) and the relationship between phonetic radical knowledge and children’s reading development.
Participants. A total of 60 Chinese children, 20 each in Grade 1 (mean age = 7 years, 2 months, mean IQ = 106), Grade 3 (mean age = 9 years, mean IQ = 114), and Grade 5 (mean age = 11 years, 2 months, mean IQ = 111), were recruited from another government-subsidized ordinary primary school in Hong Kong. Again parents of these children consented to allow their children to participate in the study. The kind of reading instruction they received was similar to that which we described in the introduction of this paper. There were 10 boys and 10 girls at each grade and all were Cantonese-speaking. They were not required to have participated in Study 1 because it would have taken them too long to go through both studies.
Materials and Procedures. An IQ test (Raven’s Standard Progressive Matrices), a Chinese Word Reading Test, and five phonetic radical knowledge tasks (see Appendix B for some sample items) were administered to the children. Apart from the IQ test which was a group test, all other tests were administered to the children individually. The IQ test and the word reading test were the same as those used in Study 1. Again all the test materials were developed by the authors except for the Raven’s test and the Chinese Word Reading Test. Pilot studies were conducted to ensure that the test materials and procedures were appropriately developed and designed.
The character decision task was similar to the character decision task in Study 1. It measured the children’s knowledge of character structure. Test materials included 24 familiar characters, 24 rare characters, 24 pseudocharacters, and 24 noncharacters. Half of the characters had phonetic radicals with high frequencies and the other half had radicals with low frequencies. Again, the children were asked to judge whether a stimulus looked like a real Chinese character.
The radical position judgment task, as in Study 1, measured the children’s explicit knowledge of the position of phonetic radicals. Twenty phonetic radicals, 10 of high frequency and the other 10 of low frequency, were chosen as stimuli. For each item, the children were visually presented with two square boxes, of which one was divided into two halves horizontally and the other vertically. These boxes represented left-right character structure and top-bottom character structure respectively. The children were asked to indicate the legal position of each phonetic radical by pointing to the appropriate half in the boxes (left, right, top, or bottom).
The phonological-relatedness judgment task measured the children’s knowledge of the function of phonetic radicals and the use of analogy strategy. Twenty pseudocharacters were constructed as targets to avoid the children knowing the sound of the characters. Again, each pseudocharacter was created by combining a semantic and a phonetic radical in their legal positions. Like the procedures of semanticrelatedness judgment task in Study 1, in each item the child was asked to select from three alternatives a character that might have the same pronunciation as the target “funny” character. The three choice characters were common phonetic radical character, common semantic radical character, and control character. The common phonetic radical character was the correct answer.
The Chinese pseudocharacter naming task measured the children’s overall knowledge of the position, function, and sound value of phonetic radicals, and the use of both direct derivation and analogy strategies. Twenty-five pseudocharacters were presented to the children. They were asked to guess the pronunciation of these “funny” characters. The phonetic radicals of these pseudo-characters were selected from some phonologically regular and consistent characters that were within the children’s reading vocabulary. A pseudocharacter was considered to be correctly read if it was pronounced by the name of its phonetic radical or a character having the same phonetic radical as the pseudocharacter.
Results and Discussion
Table 3 presents the mean scores and standard deviations of the tasks given in Study 2. The internal reliability of the various tasks ranged from satisfactory (Chinese phonetic radical naming, reliability coefficient = .69) to very good (Chinese word reading, reliability coefficient = .99). The first and third graders in this study performed somewhat less well than those in Study 1 in the same Chinese word reading test. This difference in performance may be due to the fact that the first graders in the study had lower average IQ than those in Study 1 (106 vs. 112), and the third graders in this study were on average younger than those in Study 1 (9 years vs. 9 years 2 months).
Scores on all tasks that involved force-choice responses (i.e., character decision, radical position judgment, and phonological-relatedness judgment) were significantly different from chance (all is > 2.5, all ps 4.4; all ps
A two-way (grade x type of character) ANOVA was conducted on the character decision task. The two main effects and the interaction effect were all significant (all Fs > 4.2, all ps
Correlation coefficients were calculated to examine the importance of various types of phonetic radical knowledge for Chinese word reading. Table 4 presents a matrix of partial correlation coefficients among the various measures after controlling for the effects of age and IQ. Character decision, phonological-relatedness judgment, and Chinese phonetic radical naming (lexical items) correlated significantly with both Chinese word reading and pseudocharacter naming (all rs > .31; all ps
To the best of our knowledge, these studies are the first to examine systematically and comprehensively Chinese children’s development of radical knowledge. Findings show that radical knowledge is important for children learning to read Chinese. Various measures of semantic radical knowledge, including the position and semantic category of semantic radicals, correlated significantly with Chinese word reading and sentence comprehension in Study 1. Similarly, various measures of phonetic radical knowledge, including the function and sound value of phonetic radicals, correlated significantly with Chinese word reading in Study 2.
Since radical knowledge is important for children learning to read Chinese, a comprehensive single measure of radical knowledge is essential. To meet this end, the authors devised two original tasks, the pseudocharacter meaning judgment task in Study 1 and the Chinese pseudocharacter naming task in Study 2, which were demonstrated empirically as good composite measures of children’s knowledge of semantic radicals and phonetic radicals respectively. The two tasks had good internal reliability (all reliability coefficients > .92), and they correlated significantly with reading measures (all rs > .31; all ps
The Role of Radicals in Learning to Read Chinese
The Positional Regularity of Radicals. Many researchers have suggested that the knowledge of orthographic regularity is important in reading (e.g., Barker et al., 1992; Berninger, 1994; Corcos & Willows, 1993; Venezky & Massaro, 1979). Although past studies (e.g., Chan & Nunes, 1998; Cheng & Huang, 1995; Shu & Anderson, 1999) have shown that Chinese children possess some knowledge of character structure and positional regularity, these studies are the first to report a reliable association between the knowledge of positional regularity and measured reading performance in Chinese children. These findings show that knowledge of character structure and position of semantic radicals (but not of the position of phonetic radicals) is important for children learning to read Chinese words. As compared with phonetic radicals, semantic radicals have much greater positional regularities. The position of a semantic radical often provides more information than that of a phonetic radical about its identity (i.e., whether it is a semantic or a phonetic radical); this helps a reader to retrieve the information of radicals necessary for character recognition. For instance, a better knowledge of positional regularity may increase a reader’s awareness that most left radicals serve the semantic function and right radicals serve the phonological function in characters of left-right structure.
Overall, the knowledge of the positional regularity of semantic radicals seems to be more important for reading at the character/word level than at the sentential level (radical position judgment correlated much more strongly with word reading than with sentence comprehension, as shown in Table 2). Reading and comprehending sentences require some general and linguistic knowledge other than characterlevel knowledge.
Shu and Andersen (1999) have suggested that morphological awareness (e.g., knowing the function of semantic radicals) is important for learning to read in Chinese. Although knowledge of the function of semantic radicals did not correlate significantly with Chinese word reading or sentence comprehension in Study 1, there were signs showing that the older children did make use of the semantic cues of radicals in understanding Chinese text. As mentioned in the introduction of this paper, there are different degrees of transparency for the semantic implication of different semantic radicals. Some semantic radicals are relatively transparent and some are relatively opaque. In the Chinese sentence comprehension task in Study 1, only the fifth graders performed significantly better on items with transparent semantic radicals than those with opaque ones. This result suggests that from about Grade 5 children tend to rely on the semantic radicals of Chinese characters for meaning cues in processing a sentence. The use of semantic radicals for meaning cues at the character level may appear earlier. We suggest that the utilization of semantic cues of radicals at the sentential level may require the understanding and interpretation of relatively complicated semantic and syntactic context and how these relate to individual semantic radicals. The fulfillment of these requirements may be beyond the ability of lower primary school children.
Knowledge of the Specific Information of Radicals. After knowing which is the semantic or phonetic radical in a Chinese character and its corresponding function, a reader needs to know the semantic category or sound value conveyed by the radical in order to help him or her in reading. There have been very few studies on the impact of the knowledge about the specific information of radicals on reading.
In Study 1, knowledge of the semantic category of radicals was found to correlate significantly with Chinese word reading and sentence comprehension, and knowledge of the sound value of phonetic radicals was found to correlate significantly with Chinese word reading in Study 2. In both the semantic category judgment task in Study 1 and the Chinese phonetic radical naming task in Study 2, the children at all the grade levels did better on items with lexical radicals than those with nonlexical ones. It is not surprising that by referring to its information as a stand alone character, the lexicality of a radical affects how well a reader retrieves the sound or meaning of a radical.
As mentioned above, children may directly derive the sound of a Chinese character from the sound of its phonetic radical. That is why knowing the sound value of phonetic radicals helps children in naming Chinese characters and words. The semantic cues provided by semantic radicals may activate semantically compatible candidates which may also help in the naming process.
Single Measures of the Knowledge of Radicals
From the preceding discussion, we can conclude that radical is an important processing unit for children in the recognition of Chinese characters. The next question we ask is what would be a good measure for testing this knowledge. The pseudocharacter meaning judgment task in Study 1 was intended to be a composite measure of different semantic radical knowledge including that of position, function, and specific semantic category. Since this score correlated most strongly with Chinese word reading and highly with sentence comprehension among various semantic radical knowledge measures, it might be considered as one of the best single measures of children’s knowledge of semantic radicals.
Similarly, the Chinese pseudocharacter naming task was intended to be a composite measure of different phonetic radical knowledge including position, function, and sound value. The score of this task was found to correlate most strongly with Chinese word reading in Study 2. Thus, this test may be a good single measure of children’s knowledge of phonetic radicals.
The Development of Radical Knowledge
Chinese children seem to acquire some rudimentary knowledge of character structure quite early. Most of the first graders judged noncharacters as illegal and frequent characters as legal. However, our finding that only about half of the first graders judged pseudocharacters as legal suggest that their knowledge about positional regularity and character structure was not well developed.
Based on the results of Study 2, it seems that even first graders have acquired some knowledge of the position, function, and sound value of phonetic radicals, and this knowledge improves as they advance in grades. Similarly, Chinese first graders have acquired some knowledge about the position and semantic category of semantic radicals, but they have little idea that semantic radicals serve semantic function. It was not until Grade 3 that the children understand that they can rely on the semantic radical for meaning cues. Moreover, children’s semantic radical knowledge was better for semantically transparent, lexical radicals than opaque, non lexical ones. The need for children taking a number of years to understand and use the phonological and semantic regularities in Chinese may be partially accounted for by Shu et al.’s (2003) findings that characters introduced in lower grades are less regular or transparent than characters in higher grades.
Shu et al. (2003) have also suggested that it takes a long time for children to develop phonetic consistency awareness. However, our findings show that even the first graders have a fair level of phonetic consistency awareness (with 63% correct in the phonological-relatedness judgment task). This consistency awareness also develops gradually when children advance in grades (with 73% correct in the task for the fifth graders).
It follows from the above discussion that developmentally, Chinese children seem to learn to read Chinese characters from a visual and arbitrary to an analytic and rule-based pattern, similar to the trend of reading acquisition in alphabetic languages (e.g., Frith, 1985; Gough, Juel, & Griffith, 1992). Chinese children first develop knowledge in character structure and radical positions. This visually related knowledge does not help children to read novel characters; they may rely on rote memory in learning characters. When they find these strategies inefficient, they start to use some rule-based strategies, like the grapheme-phoneme conversion rules in English and the phonetic principle in Chinese. Learning the function and regularities of phonetic and semantic radicals help children read and understand novel characters and remember more words efficiently. In line with these suggestions, Ho, Yau, and Au (in press) have developed a model of radical knowledge development in Chinese.
Consistent with some recent research findings (e.g., Ho & Bryant, 1997b; Ho, Wong & Chan, 1999; Shu & Anderson, 1997), our findings confirm that the radical is an important orthographic unit in processing Chinese characters for children. Although instruction was not studied directly, these findings may have implications for Chinese reading instruction. Traditionally, Chinese teachers in Hong Kong have emphasized drilling and rote memorization in teaching children to learn to read Chinese. They normally employ a “whole-character approach,” often paying little attention to intracharacter orthographic units, especially the phonetic radical, in enhancing reading. Semantic radicals are mainly taught as a reference unit for looking up characters in a Chinese dictionary. This “whole-character approach” is useful in teaching integrated Chinese characters and irregular/inconsistent characters. However, teaching children the orthographic, phonological and semantic regularities of Chinese characters at an early stage may enhance their character decoding skills (especially those of compound characters) and reading comprehension. Based on the present findings, we suggest that positional regularities of semantic and phonetic radicals be taught explicitly to children in the first grade. Functional regularities of radicals should be introduced later, around second and third grade. The sound of individual phonetic radicals and the semantic category of individual semantic radicals should also be taught gradually from first grade.
Limitations and Suggestions for Future Research
Since the present studies are cross-sectional and the sample sizes are relatively small, we cannot provide a precise answer to the question regarding the development of various types of radical knowledge. This will await more thorough examination through longitudinal studies with larger sample sizes. A follow-up study of instruction of radicals would also be educationally meaningful.
In conclusion, our findings show that developmentally children start acquiring the knowledge of character structure, the position, semantic category, and sound value of Chinese radicals from about Grade 1. However, they do not understand the function of semantic radicals until Grade 3. These aspects of radical knowledge also associate strongly with Chinese word reading and sentence comprehension. On the basis of these findings, we suggest that radicals are an important orthographic processing unit in reading development in Chinese, and their positional and functional regularities should be taught explicitly in school to enhance children’s character decoding skills.
1 All pronunciation notes for Chinese characters in this paper are Cantonese pronunciations. For instance, in the syllable [dang]!, /d/ is the onset, /ang/ is the rime, and “1” means that the syllable is in the first tone, i.e., a high level tone.
Anderson, R. C., Li, W., Ku, Y.-M., Shu, H., & Wu, N. (2003). Use of partial information in learning to read Chinese characters.Journal of Educational Psychology, 95, 52-57.
Anderson, R. C., & Nagy, W. E. (1992, Winter). The vocabulary conundrum. American Educator, 14-18,44-46.
Anshen, F., & Aronoff, M. (1988). Producing morphologically complex words. Linguistics, 26, 641-655.
Barker, T. A., Torgesen, J. K., & Wagner, R. K. (1992). The role of orthographic processing skills on five different reading tasks. Reading Research Quarterly, 27, 335-345.
Berninger, V. W. (1987). Global, component, and serial procedures for printed words in beginning reading. Journal of Experimental Child Psychology, 43, 387-418.
Berninger, V. W. (1990). Multiple orthographic codes: Key to instructional interventions for developing orthographic-phonological connections underlying word identification. School Psychology Review, /9, 518-533.
Berninger, V. W. (1994). Introduction to the varieties of orthographic knowledge I: Theoretical and developmental issues. In V. W. Berninger (Ed.), The varieties of orthographic knowledge I: Theoretical and developmental issues (pp. 1-25). Dordrecht, The Netherlands: Kluwer.
Chan, C. K., & Siegel, L. S. (2001). Phonological processing in reading Chinese among normally achieving and poor readers. Journal of Experimental Child Psychology, 80, 23-43.
Chan, L., & Nunes, T. (1998). Children’s understanding of the formal and functional characteristics of written Chinese. Applied Psycholinguistics, 19, 115-131.
Chen, Y. P., Allport, D. ?., & Marshall, J. C. (1996). What are the functional orthographic units in Chinese word recognition: The stroke or the stroke pattern? The Quarterly Journal of Experimental Psychology, 49A, 1024-1043.
Chen, M. J., & Weekes, B. S. (1997, August). Semantic consistency and character recognition. Paper presented at the International Symposium on Cognitive Processes of the Chinese Language, Hong Kong.
Cheng, C.-M., & Huang, H.-M. (1995, December). The acquisition of general lexical knowledge of Chinese characters in school children. Paper presented at the Seventh International Conference on the Cognitive Processing of Chinese and other Asian Languages, Hong Kong.
Corcos, E., & Willows, D. M. (1993). The processing of orthographic information. In D. M. Willows, R. S. Kruk, & E. Corcos (Eds.), Visual processes in reading and reading disabilities (pp. 163-190). Hillsdale, NJ: Erlbaum.
Ehri, L. C. (1991). Development of the ability to read words. In R. Barr, M. L. Kamil, P. B. Mosenthal, & P. D. Pearson (Eds.), Handbook of reading research, (Vol. II, pp. 323-358). New York: Longman.
Ehri, L. C. (1992). Reconceptualizing the development of sight word reading and its relationship to recoding. In P. B. Cough, L. C. Ehri, & R. Treiman (Eds.), Reading acquisition (pp. 107-143). Mahwah, NJ: Erlbaum..
Ehri, L. C., & Robbins, C. (1992). Beginners need some decoding skill to read words by analogy. Reading Research Quarterly, 27, 13-26.
Feldman, L. B., & Siok, W. W. T. (1997). The role of component function in visual recognition of Chinese characters. Journal of Experimental Psychology: Learning, Memory, and Cognition, 23, 776-781.
Feldman, L. B., & Siok, W. W. T. (1999a). Semantic radicals in phonetic compounds: Implications for visual character recognition in Chinese. In J. Wang, A. W. lnhoff, & H. C. Chen (Eds.), Reading Chinese script: A cognitive analysis (pp. 19-35). Mahwah, NJ: Erlbaum.
Feldman, L. B., & Siok, W. W. T. (1999b). Semantic radicals contribute to the visual identification of Chinese characters. Journal of Memory and Language, 40, 559-576.
Foreign Languages Press Beijing. (1989). Chinese characters. Beijing: Foreign Language Press.
Freyd, P., & Baron, J. (1982). Individual differences in acquisition of derivational morphology. Journal of Verbal Learning and Verbal Behavior, 21, 282-295.
Frith, U. (1985). Beneath the surface of developmental dyslexia. In K. Patterson, M. CoItheart, &J. Marshall (Eds.), Surface dyslexia (pp. 301-330). London: Erlbaum.
Goswami, U. (1986). Children’s use of analogy in learning to read: A developmental study. Journal of Experimental Child Psychology, 42, 73-83.
Gough, P. B., Juel, C., & Griffith, P. L. (1992). Reading, spelling, and the orthographic cipher. In P. B. Cough, L. C. Ehri, & R. Treiman (Eds.), Reading acquisition (pp. 35-48). Hillsdale, NJ: Erlbaum.
Ho, C. S.-H., & Bryant, P. (1997a). Learning to read Chinese beyond the logographic phase. Reading Research Quarterly, 32, 276-289.
Ho, C. S.-H., & Bryant, P. (1997b). Phonological skills are important in learning to read Chinese. Developmental Psychology, 33, 946-951.
Ho, C. S.-H., Wong, W.-L, & Chan, W.-S. (1999). The use of orthographic analogies in learning to read Chinese. Journal of Child Psychology and Psychiatry, 40, 393-403.
Ho, C. S.-H., Yau, P. W.-Y., & Au, A. (in press). Development of orthographic knowledge and its relationship with reading and spelling among Chinese kindergarten and primary school children. In C. McBride-Chang & H.-C. Chen (Eds.), Reading development in Chinese children. Westport, CT: Praeger.
Hoosain, R. (1991). Psycholinguistic implications for linguistic relativity: A case study of Chinese. Hillsdale, NJ: Erlbaum.
Huang, H. S., & Hanley, J. R. (1997). A longitudinal study of phonological awareness, visual skills, and Chinese reading acquisition among first graders in Taiwan. International Journal of Behavioral Development, 20, 269-286.
Jackson, N. E., & Coltheart, M. (2001). Routes to reading success and failure. New York: Psychology Press.
Kang, J. S. (1993). Analysis of semantics of semantic-phonetic compound characters in modern Chinese. In Y. Chen (Ed.), Information analysis of usage of characters in modern Chinese (pp. 68-83). Shanghai: Shanghai Education Publisher (in Chinese).
Leck, K.J., Weekes, B. S. & Chen, M.J. (1995). Visual and phonological pathways to the lexicon: Evidence from Chinese readers. Memory and Cognition, 23, 468-476.
Li, H. (1997). Recognition of Chinese characters: A radical-based approach. Unpublished doctoral dissertation: The Chinese University of Hong Kong.
Li, H., & Chen, H.-C. (1997). Processing of radicals in Chinese character recognition. In H.-C. Chen (Ed.), Cognitive processing of Chinese and related Asian languages (pp. 141160). Hong Kong: Chinese University Press.
Liu, I. M., Chuang, C. J., & Wang, S. C. (1975). Frequency count of 40,000 Chinese words. Taipei: Lucky Books.
Miao, X. C., & Sang, X. (1991). A further study on semantic memory for Chinese words. Psychological Science, 1, 6-9.
Nagy, W. E., & Anderson, R. C. (1999). Metalinguistic awareness and literacy acquisition in different languages. In D. Wagner, R. Venezky, & B. Street (Eds.), Literacy: An international handbook (pp. 155-161). New York: Garland.
Nagy, W. E., Anderson, R. C., Schommer, M., Scott, J. A., & Stallman, A. C. (1989). Morphological families and word recognition. Reading Research Quarterly, 24, 262-282.
Peng, D. L., Li, Y.-P., & Yang, H. (1997). Orthographic processing in the identification of Chinese characters. In H.-C. Chen (Ed.), Cognitive processing of Chinese and related Asian languages (pp. 85-108). Hong Kong: Chinese University Press.
Perfetti, C. A. (1985). Reading ability. New York: Oxford University Press.
Shankweiler, D., Lundquist, E., Katz, L., Stuebing, K., Fletcher, J. M., Brady, S., Fowler, A., Dreyer, L. G., Marchione, K. E., Shaywitz, S., & Shaywitz, B. A. (1999). Comprehension and decoding: Patterns of association in children with reading difficulties. Scientific Studies of Reading, 3,69-94.
Shu, H., & Anderson, R. C. (1997). Role of radical awareness in the character and word acquisition of Chinese children. Reading Research Quarterly, 32(1), 78-89.
Shu, H., & Anderson, R. C. (1999). Learning to read Chinese: The development of metalinguistic awareness. InJ. Wang, A. W. Inhoff, & H. C. Chen (Eds.) Reading Chinese script: A cognitive analysis (pp. 1-18). Mahwah, NJ: Erlbaum.
Shu, H., Chen, X., Anderson, R. C., Wu, N., & Xuan, Y. (2003). Properties of school Chinese: Implications for learning to read. Child Development, 74, 27-47.
Taft, M., & Zhu, X.-P. (1997). Submorphemic processing in reading Chinese. Journal of Experimental Psychology: Learning, Memory, and Cognition, 23, 761-775.
Tian, B.-N. (1987, December). The concentrated character learning method: A primary school reading instruction reform project. The proceedings of the third international conference of the language education institute. Hong Kong. (In Chinese).
Tyler, A., & Nagy, W. E. (1989). The acquisition of English derivational morphology.Journal of Memory and Language, 28, 649-667.
Venezky, R., & Massaro, D. (1979). The role of orthographic regularity in word recognition. In L. Resnick, & P. Weaver (Eds.), Theory and practice of early reading (Vol. 1, pp. 85-108). Hillsdale, NJ: Erlbaum.
Wysocki, K., & Jenkins, J. R. (1987). Deriving word meanings through morphological generalization. Reading Research Quarterly, 22, 66-81.
Yang, H., & Peng, D. (1997). The learning and naming of Chinese characters of elementary school children. In H.-C. Chen (Ed.), Cognitive processing of Chinese and related Asian languages (pp.323-346). Hong Kong: The Chinese University Press.
Zhang, J. J., Zhang, H. C., & Peng, D. L. (1990). The recovery of the meaning of Chinese characters in the classifying process. Acta Psychologica Sinica, 22(2), 139-144.
Zhou, Y.K. (1980). Precise guide to pronunciation with Chinese phonological roots. Juin, China: People’s Publishing Co. (In Chinese).
Zhu, Y.P. (1987). Analysis of cuing functions of the phonetic in modern China. Unpublished manuscript, East China Normal University. (In Chinese).
Connie Suk-Han Ho
University of Hong Kong
Chinese University of Hong Kong
Chinese University of Hong Kong
Copyright National Reading Conference Fall 2003
Provided by ProQuest Information and Learning Company. All rights Reserved