Rachel (R’el) Johnston and Dr. Masakazu Watabe, Asian and Near Eastern Languages
Japanese has two types of writing: kana, which is essentially phonetic, and kanji, which is primarily logographic. I hypothesized that although they are not phonetic and orthographically more complex, learning kanji actually improves vocabulary learning. In order to test this hypothesis, I created a study comparing the learning of basic vocabulary for beginning Japanese students using kana as opposed to kanji. The study involved a pre-test, a thirty minute study period, and a post-test. The pre-test and post-test were exactly identical for all students. They each, along with the study packets, contained 100 vocabulary items in English and Japanese, equally divided among different parts of speech. Eleven of the students were given a study packet that included only kana and English while the other eleven had a study packet including kana and kanji along with English for each vocabulary term. After the study period I administered the post-test. I compared the improvement of those who used the kana-only study packet to the improvement of those who used the kanji study packet.
My test group included 22 BYU students who were enrolled in a Japanese class during winter semester of 2003. 21 of them were enrolled in Japanese 102 (second semester) and one in both 202 and 221 (fourth semester). One participant had had extensive prior experience with Chinese characters. The study took place on Saturday April 12, 2003, which was at the very end of winter semester. This means that the 102 students had learned approximately 100 kanji and become well-versed in kana. The 202/221 student had learned approximately 500 kanji. None of the kanji on my list had been studied by the 102 students. As I did not advertise my study to 221 students and was not planning to include them, the test did include kanji studied by the 221 student.
I administered the tests as planned. My analysis of the resulting scores indicates that there is a strong correlation between using kanji to learn vocabulary and improved learning of vocabulary. Those who used the kanji study packets had an average improvement of 46.27 while those using the kana-only packets had an average improvement of only 19.37. This is quite a significant difference. Both the subject enrolled in 202 and 221 and the subject with extensive Chinese experience had the kanji study packet, so I also analyzed the data not including these two subjects in case their greater experience had skewed the results. In fact it did not, and the difference in improvement is statistically significant whether those two subjects are included or not.
This research was elementary, but it is the first step in an important area of study that could affect methods for teaching Japanese as well as increase our understanding of the connection between written cues and the linguistic memory. I originally planned to re-administer the test one week and one month following the preliminary testing time, but due to logistical difficulties this was impossible. Further research could include later testing, as well as more sophisticated testing of vocabulary learning and retention, such as oral tests and written texts in context rather than a list format. The most immediate application of my research would be to integrate kanji with the learning of vocabulary as early in the student’s Japanese learning career as possible, since this would facilitate the learning of vocabulary in general, not just for reading. My research also has interesting applications for orthography generally. An extension of my research is the hypothesis that a disconnect between phonology and the grapheme is actually beneficial for language learning. Japanese is the language in which this can be demonstrated most obviously, since it contains two essentially complete writing systems, one phonetic and one non-phonetic. Further research would involve languages that have primarily phonetic orthographies, such as Spanish, Italian, and Swahili; languages with semi-phonetic orthographies such as English and French; and languages with essentially logographic orthographies such as Chinese. Comparing vocabulary learning across languages would be more difficult than a comparison such as mine within a language, but it could demonstrate a great deal about how the human mind processes graphemes and retains vocabulary.