Many of my readers knew that I have dedicated myself to using emerging technology to enhance the Chinese language learning experience. Before reading this book, I consulted with several Chinese language teachers across various educational levels. During these consultations, I did not use well-designed interview questions. However, this book has allowed me to clearly identify several challenges in the systematic teaching of Chinese within K-12 education. By accurately framing these specific issues within a broader context, I have gained a clearer understanding of their scope. This has also enabled me to quickly understand the official and widely recognized learning materials employed in Chinese language teaching.
Wise, J. C., Sevcik, R. A., Morris, R. D., Lovett, M. W., & Wolf, M. (2007). The relationship among receptive and expressive vocabulary, listening com- prehension, pre-reading skills, word identification skills, and reading com- prehension by children with reading disabilities. Journal of Speech Language and Hearing Research, 50(4), 1093–1109.
What is also remarkable about the Chinese phonology is that as com- plex as the tonal system appears to be, there is a very limited number of syllables. One estimate is just 404 when tones are not taken into consid- eration (Duanmu, 2007, pp. 319–329), and about 1277 when tones are counted (DeFrancis, 1984, p. 42). By contrast, English has over 8000 syllables (DeFrancis, 1984, p. 42).
Partly due to the relatively small sound inventory, Standard Chinese has a great number of homophones, which refer to words with the same pronunciation but different meaning (therefore representing dif- ferent morphemes). For instance, the tonal syllable shì corresponds to more than ten different morphemes represented by different characters, examples include 事 “issue”, 视 “vision”, 试 “to try”, 室 “room”, 饰 “to decorate”, 适 “to adapt to”, 示 “to demonstrate”, 释 “to explain”,市 “city”, 柿 “persimmon”. On average, a tonal syllable is represented by about five characters/morphemes; in contrast, English syllable is represented by on average 1.4 word spellings (Duanmu, 2007, p. 94). However, in Standard Chinese, homophones are not evenly distributed across all syllables. Duanmu (2007, pp. 94–95) provided a corpus-based analysis and listed 15 most frequent syllables in terms of homophone density. The top one, yi, represents more than 100 words when tone is disregarded. However, about 20% of the 1277 tonal syllables do not have a homophone, and some of the most frequently used words are also unique in speech and are uniquely represented by a single character (DeFrancis, 1984, p. 184), such as 我 wǒ “I”, 牛 niú “ox”.
In summary, Standard Chinese is a tonal language and has a rela- tively simple syllable structure and a small sound inventory. But its syl- lable-morpheme/character mapping is complicated by the wide presence of homophones.
The above content reminds me of someone on Twitter who is dedicated to solving the problem of homophones/similar pronunciation.
English pseudoword naming. The Word Attack subset from the Woodcock-Johnson Diagnostic Reading Battery (Woodcock, 1997) was used to assess the participants’ ability to produce a plausible pronunci- ation of a pseudoword. The entire subset was used for this task and it contained 30 pseudowords with increasing complexity. The participants were asked to try their best to sound the words out.
I will pay attention to all pain points in teaching next time I read, as well as how to solve them with technology and better interactive design as I think of them.
How to design a more scientific assessment system, especially for evaluation methods that I am not familiar with, as well as my memory of the Duolingo exam. It is worth mentioning that within just a year, I can clearly perceive Duolingo continuously optimizing the assessment of the Duolingo English Test to reflect actual language proficiency (making it increasingly difficult to improve scores through intensive pre-exam review preparation).
Considering some experiments conducted by the author that originally required semester-long units, how to simulate them based on the LLM.
Some of the unfamiliar linguistic terms or research methods that I skipped over
Design for people with reading disabilities.
Gough, P. B., & Tunmer, W. E. (1986). Decoding, reading, and reading disa- bility. Remedial and Special Education, 7(1), 6–10.
Ouellette, G., & Beers, A. (2010). A not-so-simple view of reading: How oral vocabulary and visual-word recognition complicate the story. Reading and Writing: An Interdisciplinary Journal, 23(2), 189–208.
In Chapter 10, I strongly became aware of some similar issues discussed at the time, both in designing the CU Boulder Generative AI course and in actual teaching. It gave me a clear guide on how to present those from academic perspective.
Simplified and traditional script student autonomy choice, and how to integrate personalized choices and convenient unified teaching in teaching practice.