Curriculum design in teaching Chinese characters to American students: when and what?

1 Introduction

Orthography is a standardized system of writing. It is the collection of rules that allow us to visually represent a language. There are three major types of orthography. Languages that use symbols for syllabic sounds have a syllabic orthography, such as Japanese; languages, like English, that have symbols for individual sounds use an alphabetic orthography; and the Chinese language, in which symbols are used for entire morphemes, uses a logographic orthography (DeFrancis 1984; Perfetti and Dunlap 2008; Shen 2013; Xu and Jen 2005; Zhou and Marslen-Wilson 1999). Even though there are many characters in Chinese and 81.2 % of the characters are semantic-phonetic compounds compared to 3.9 % as pictographic, 1.3 % as ideographic, 12.3 % as compound indicatives, 1.2 % as borrowing, and 0.07 % as explanatory (Sun 2006), only about 30 % of the total amount of characters, i.e., only 37 % of the total amount of the semantic-phonetic compounds, have some internal features truly corresponding to sounds in the syllable (Zhou 2009). The majority of the semantic-phonetic compound characters are not phonetic because “the so-called phonetic compounds represented sounds fairly closely when they were made, but often are no longer appropriate for modern pronunciations” (Chao 1976: 92). In other words, the phonetic part of the semantic-phonetic characters only reflects the category of sound because of the changes of pronunciation in thousands of years, even less possible to entirely reflect the real way of the pronunciation of the character (Huang 2005).

As a logographic orthography, the Chinese language inherently values both the written form and the spoken form, which is expressed by the conventional Chinese word for “language”, that is, “yǔwén (語文).” “Yǔwén” is composed of two syllables representing two different characters, each with its own meaning: yǔ (語) represents yǔyán (語言), the vocal aspect of the language; and wén represents wénzì (文字), the image or symbol of the language. Therefore, “language” in Chinese refers to two different symbolic systems: yǔ and wén, spoken and written, sound and script, or phonology and orthography. Their existence indicates that the sound and the orthography of the language are both important. Wén (文字) is not only an integral part of the Chinese language, but it also carries cultural meanings. Traditionally in China, knowing characters, or shízì, indicated and continues to indicate one’s status and intelligence. The historian Zheng Qiao wrote in Tongzhi in 1161 AD, “In the empire, those who know characters are regarded as wise and learned and those who do not as stupid and mediocre” (cited in Wilkinson 2022).

There is not an accurate account of the total number of characters in the Chinese language, but we can find a rough number by looking at the most representative dictionaries. For example, Zhōnghuá Zìhǎi, the largest Chinese character dictionary available in print and published in 1994 contains 85,568 characters. The second edition of Hànyǔ Dàzìdiǎn (the Great Compendium of Chinese Characters), the international standard reference for Chinese characters published in 2010, includes 60,370 characters. The best-selling and most popular Chinese dictionary Xīnhuá Zìdiǎn, in its 12th edition published in 2020, has over 13,000 characters.

Even though Chinese has tens of thousands of characters, “adult literacy is defined as knowing at least 3,000 characters” (Wilkinson 2022: 38) in addition to other influencing factors involved in literacy development such as knowledge of words, grammar, punctuation, discourse and so on. There has been an official listing of 7,000 commonly used characters in Chinese since 1988 (Wilkinson 2022). Out of the 7,000 characters, knowledge of the most frequently used 2,500 characters is enough for a person to comprehend 97.97 % of everyday Chinese writing. Knowing the next 1,000 most commonly used characters would enable comprehension of 99.48 % of everyday Chinese writing and knowing an additional 1,700 (for a total of 5,200 characters) would enable 99.99 % comprehension (Wilkinson 2022).

2 Literature review

Two previous review studies of character teaching have approached the literature from different perspectives. Li’s (2020) systematic review used a thematic analysis method to analyze articles on Chinese character teaching and learning published in and outside China between 2005 and 2019. This study found a large gap in the use of research methodologies between the studies published in China and those published elsewhere. Li also discovered that the use of computer-assisted language learning in character teaching was a key research topic outside China, while exploring new pedagogical ideas and strategies was of interest among scholars in China. This review study provided valuable information on research themes and methodology trends, but it did not examine the issues of teaching and learning characters in second language curriculum.

Zhang and Ke (2018) reviewed research findings that focused on the impact of learners’ knowledge about radicals on character learning, the effects of L1 background on L2 Chinese character acquisition, and the learning strategies adopted by L2 learners. (Note: A radical is the base graphical component of a compound character under which the character is traditionally listed in a Chinese dictionary). They also discussed two curricular issues regarding character instruction: the timing of introducing Chinese orthography; and whether we should teach handwriting or computer-assisted writing. They focused on the reasons for, benefits of, and problems of delaying or not delaying character introduction and how to balance handwriting with computer-assisted writing. A systematic review of literature that gives a more detailed and precise understanding of the issues of situating Chinese character instruction within the curriculum design is still needed.

Another two review studies focused on character teaching strategies. Zhang (in press) systematically reviewed literature relevant to strategies used in character teaching in a classroom/lab setting online and offline to L1 English CFL learners. She identified 22 empirical studies published from 1952 to the present. These studies were cross analyzed in terms of research methods, learners’ proficiency level, learning settings, methodological approaches, and theoretical bases. Out of the 39 methods utilized in these studies, she found six categories of character instructional methods. The analysis of the data points out the strengths and weaknesses of the current research on character teaching. One of the weaknesses is the lack of research on using different approaches at the curriculum level and so she called for research on character teaching from the perspective of curriculum design.

Zhang and Da (in press) completed a synthesis study on how technology has been integrated to assist character teaching. This study exhaustively examined the literature related to technology use in teaching characters and found 24 relevant empirical studies from the beginning of this century. These studies used different types of technology to improve various aspects of character learning. This synthesis study shed light on how to use specific technologies to maximize students’ character learning, but it does not provide suggestions on issues such as the timing of character introduction and how to integrate character learning into the curriculum.

Other studies review how CFL learners cognitively process characters and what factors affect the cognitive development of students’ character learning. For example, Shen (2013) systematically examined CFL learners’ literacy development in terms of character learning, including the cognitive characteristics of learners’ development of metalinguistic awareness in Chinese, the strategies they employed when learning to read and write Chinese characters, and the optimum pedagogical practices to facilitate character learning. Ésik (2020) discussed general questions such as whether CFL learners should learn to read and write Chinese characters and whether simplified or traditional characters should be taught, but she did not reference previous scholarly works. A historical view, however, is necessary because debates over how character teaching appears in the curriculum continue since the 1930s when Chinese began to be taught in America (e.g., Allen 2008; Knell and West 2015). These studies have no doubt contributed to expanding our understanding of character teaching, but a clear comprehensive review of what scholars and practitioners have proposed is needed.

The purpose of this synthesis study is to identify peer-reviewed research on teaching Chinese characters, to systematically analyze this body of research, and to offer a tentative model for curriculum integration of Chinese character instruction. Due to the absence of a transparent boundary for Chinese words and the fact that “Chinese orthography requires no space between characters, regardless of their morphological status” (Sun 2006: 46), the distinction between the notion of “word” (cí) and the notion of “character” (zì) often becomes obscured in speakers’ mindset. This synthesis study primarily focuses on the instruction of zì (characters) – a more intuitive concept according to Sun (2006) – each of which represents a morpheme-syllable irrespective of its morphological classification. It is also important to note that while the debate between monism and dualism approaches in teaching Chinese exists (Bellassen 2018), its discussion is more appropriately reserved for a separate article.

The research questions for this study are:

1. What does nine decades of literature say about the curriculum for Chinese character teaching in the United States?

2. What can be drawn from the literature to help design a model that integrates character teaching into the current CFL curriculum?

3 Methodology

To answer the research questions, a systematic review method was adopted to synthesize current knowledge on various topics. The increased demand for using research evidence to inform policy and to practice decision-making in evidence-informed education (Nelson and Campbell 2017) has called people’s attention to the “contextual and methodological limitations of research evidence provided by single studies” (Newman and Gough 2020: 4). Synthetic reviews of research carried out in a systematic, rigorous, and transparent manner can help address those limitations (Newman and Gough 2020).

A few steps are necessary for conducting a systematic review (Newman and Gough 2020; Tawfik et al. 2019). These steps include constructing selection criteria for a literature search based on the research questions, selecting studies using selection criteria, developing a search strategy, and synthesizing results of individual studies to answer the research questions. These steps were followed in this study.

Four criteria were used in the literature search. First, only peer-reviewed journal articles were included. Second, each study had to focus on the curriculum design for Chinese writing system teaching, such as whether characters should be taught, when characters should be introduced, and what aspect of the written language should be taught. Studies done in the area of teaching strategies that resulted in helping teachers design specific methods in teaching characters but that are not directly related to how to integrate teaching written Chinese in the curriculum design were not included in this study (see Section 1 for a comprehensive rationale behind the exclusion of studies focusing on teaching strategies. For a comprehensive examination of specific character teaching methods, please refer to the forthcoming systematic review by Zhang [in press] and Zhang and Da [in press]). Third, each study must inspect teaching Chinese to American students in the United States. Fourth, the study had to be written either in English or in Chinese. The search included studies published between 1937 and 2022. Given the inherent difficulty in conducting empirical studies pertaining to curriculum design issues and the possibility that there might be a limited number of such studies available, it is pertinent to note that the search scope is not restricted solely to empirical research. Concept papers authored by esteemed experts in the field hold considerable value, as they offer valuable insights, vision, and guidance for the advancement of the discipline.

Three search methods were used for the purpose of identifying all available studies that met the four criteria.

1. Database search. Four databases were used: Linguistics and Language Behavior Abstracts (LLBA), ERIC, PsychInfo, and Education Index Retrospective (EIR). The databases complement each other to cover all the important journals in a wide range of fields, including linguistics and languages, education, and educational psychology. When searching these databases, “Chinese character”, “hanzi”, “written Chinese”, and “Chinese writing system” were used separately as search terms to include all possible articles. A manual screening was then done with the search results to gather relevant articles.

2. Focal journal archive search. The two most important CFL journals in the United States were closely inspected: Chinese as a Second Language – the flagship journal of the Chinese Language Teachers Association – and the Journal of Technology and Chinese Language Teaching. The recent indexing of these two journals indicates that some issues might not have been indexed in the four databases. The search of the archives of these two journals ensured the exhaustiveness of the search.

3. A manual search. A manual search through the references of the selected articles from the previous two searches was conducted. The manual search helped find any relevant articles that the other searches might have missed.

These three searches resulted in a set of 98 articles that were related to Chinese character learning and teaching. These articles were further screened manually based on the selection criteria. Sixteen articles satisfied all the criteria. Information about each of the 16 articles was recorded and analyzed. See Appendix for the 16 articles.

4 Findings and discussion

The chronological distribution of the 16 studies shows an increase in studies published to address character teaching curricular issues in the past two decades (see Figure 1). However, compared to other aspects of character teaching, such as teaching methods and character learning development and strategies, there are far fewer publications on curricular issues, such as whether and when to teach characters and what skills or aspects of the characters to teach (Loh et al. 2018; Shen 2013; S. Zhang 2021; Zhang in press). The two earliest published articles studied teaching Chinese characters. “A minimum vocabulary in modern Chinese” by Kennedy (1937) was published in The Modern Language Journal, and “Learning the Chinese script can be easy” by Shen (1958) was published in Language Learning. This confirms that scholars have endeavored from the beginning to address the challenging issue of teaching Chinese in the United States.

Figure 1:

Chronological distribution of 16 articles published from 1937 to 2022.

As expected, a majority of the 16 studies were conceptual (see Figure 2). Out of the five empirical studies, three were experimental studies, one was a survey study, and one was a mixed methods study. The small number of empirical studies might be due to the fact that researching the effectiveness of curriculum design needs more resources, time, and support than studies on teaching strategies.

Figure 2:

Research methodologies adopted in the 16 studies.

Kennedy’s (1937) concept paper focusing on character teaching was the first to point out the main reasons for the challenges American CFL learners face. First, “its ideographs present in themselves no correlation between sound and form”, and second, “[the ideographs] have almost entirely lost what might have been a compensatory advantage, namely, the correlation between form and concept that would characterize a more primitive system of pictorial writing” (ibid.: 587). He stated that the learning of ideographs (or characters) was the fundamental problem in the study of written Chinese. He also addressed the issue about how many characters are appropriate for beginner learners to master. In addition, Kennedy extended his argument to literacy development by calling attention to two other difficult aspects in learning written Chinese: words and “connected prose” (ibid.).

Thus, Kennedy’s (1937) article laid the foundation for character teaching. First, it pointed out the distinct feature of the Chinese language from other alphabetic languages that characters are not equivalent to words; therefore, the teaching method for alphabetic languages should not apply to character teaching. Second, it revealed that students needed to make extra effort in learning characters because “combinations of ideographs require a further effort of learning” (ibid.: 590). Third, it reasoned that learning characters did not necessarily start with learning to write spoken Chinese because the most frequently spoken language is not necessarily frequently written. To economize learners’ efforts, this article proposed the need for “a more scientific standard of selection” of characters to be mastered (ibid.: 591). Fourth, it suggested that word frequency should be a factor in selecting characters that are taught. This suggestion corresponds to the more recent argument by Huang (2005).

This 1937 article heralded the major themes of debate regarding curriculum design of character learning in later research. Two themes were identified among the 16 articles: when characters should be introduced and what about the characters should be taught. These two themes were interwoven in the 16 articles owing to the interdependent relationship between the timing and content of teaching characters.

5 A tentative model for teaching written Chinese to beginners

While acknowledging that the number of empirical studies included in this analysis is limited to five and that some of these studies may have inherent limitations, it is important to highlight the significant progress made in advancing our understanding of character teaching within curriculum design. The collective insights presented in the 16 articles have played a pivotal role in fostering a comprehensive understanding about how we can introduce characters and their closely related elements in a CFL program or in course design. In addition to the 16 articles, there are two curriculum design models proposed by scholars that can serve as additional references for ensuring comprehensive coverage of various aspects of the Chinese characters. It is important to note that these models concentrate on specific issues or target particular audiences, thereby providing valuable insights into some special scenarios of character teaching. Jiang (2017) targeted international CFL learners when she proposed sequencing spoken and written learning, emphasizing character recognition over production, and learning handwriting before typing as a strategy for improving learning. Chu (2020) proposed a model emphasizing typing over handwriting and called for a systematic study on how to integrate the Chinese writing system into the curriculum. A synthesis of these two relevant models and the findings and discussion of the 16 articles in this project helped form the tentative model proposed in this section.

This tentative model is for beginner students to gain knowledge of the Chinese characters, hone character recognition and production skills while improving listening and speaking skills, gain confidence to function in the Chinese language community, and become lifelong Chinese language learners. The basic concepts used to design this model are: (a) learning spoken Chinese that is not slowed down by the progress of character learning; (b) learning useful typing tools to communicate in social media with native Chinese speakers; (c) laying a solid foundation for character learning that can help learners follow the inner logic of the characters and apply previous learning to future tasks; (d) using character learning to give students an overview and preliminary understanding of the writing system and related culture; and (e) serving as the first step in learning shūmiànyǔ and developing reading literacy early in their learning. See Figure 3 for the framework of the proposed model. This tentative model for Teaching Written Chinese to Beginners has two features that reflect the design principles. Principle #1: Start speaking and writing at the same time but with two different tracks; Principle #2: The two tracks are designed differently to maximize learning.

Figure 3:

A tentative model for teaching written Chinese to beginners.

6 Conclusions

This study synthesized research related to curriculum design for teaching Chinese characters published from the 1930s to the present. Based on the synthesis, a tentative model for integrating teaching characters into the curriculum was created. This model is preliminary, and it still needs an enormous amount of work to design and develop. The design and development of Track A might be easier compared to the design and development of Track B as current textbooks for speaking might work, but the typing requirements and practices need to be carefully created to make students’ typing practice effective and engaging, as Xu and Jen (2005) and P. Zhang (2021) indicated. Track B, with its distinct approach, necessitates the development of specialized teaching materials, which can draw upon existing resources previously published, such as Allanic (2009), Bellassen (1989), DeFrancis (1964), Kubler and Kubler (2018), and Yin and Sun (2018). However, due to the unique nature of Track B, designing and developing materials for this track requires an in-depth understanding of Chinese characters and literacy development. Moreover, the pedagogical strategies employed in teaching Track B should be informed by comprehensive studies conducted on character teaching approaches. These studies should also serve as valuable resources in informing the development of materials for Track B (Zhang in press).

Many practical questions are waiting to be explored. The major ones include, how much content should be developed for Track B? If not all of the proposed contents fit in Track B, what should be prioritized? Is it necessary to merge Track A and Track B at some point, or should they remain two separate tracks? Or should Track B eventually be developed into a reading and writing track focusing on reading and writing shūmiànyǔ? If Track A and Track B should be merged, when should that occur?

Indeed, the proposed model raises several important research questions that warrant investigation. For instance, it is crucial to determine the effectiveness of implementing this model in enhancing students’ speaking proficiency, character recognition, and production skills. Additionally, it is important to explore the model’s efficacy for different student populations, such as middle or high school students, heritage learners, and those in immersion programs. An investigation into whether Track B aligns with learners’ expectations and preferences is also valuable. Furthermore, determining the most effective instructional approaches for both Track A and Track B is essential for optimizing student learning outcomes. Certainly, the feasibility and effectiveness of teaching Track B online are important aspects that should be explored. With the increasing prevalence of online learning platforms and the advancements in technology, investigating the viability of implementing Track B in an online setting is crucial.

These research questions highlight the need for rigorous empirical studies in this domain. By conducting robust research, both researchers and practitioners in the CFL field can gain a deeper understanding of how to proceed with confidence. Addressing these open questions will contribute to the ongoing improvement and refinement of CFL instruction.