Paper 007: The Learners’ Development of Using Polysemous Words: The Case of ‘Over’
AZEMOTO, Risako (Kyushu University, Japan)
Keywords: second language acquisition, polysemous words, learner corpus
Abstract
Acquiring multiple senses of a polysemous word perfectly is difficult for language learners. Although some researchers have tried to find the effective way of teaching a polysemous word, few studies have examined how learners develop their knowledge of it. Therefore, this presentation aims to show insights of how Japanese learners of English develop their use of a polysemous word “over”. To achieve this purpose, this study set two research questions;
RQ1: Do Japanese learners of English use “over” the same way as native speakers do?
RQ2: What is the difference between “easy” senses and “difficult” senses?
To answer these two research questions, both quantitative and qualitative research was undertaken. First, a learner corpus, Longman Learners’ Corpus, and a native corpus, British National Corpus, were used to collect sufficient examples of “over” from each proficiency level of the learners and English native speakers. After that, semantic tags of “over” from Tyler and Evans (2001) were used to annotate each example. Then, a correspondence analysis was generated based on the cross-tabulation which organizes the results of the annotation. Finally, the difference between easy senses and difficult senses for Japanese language learners was discussed based on the result of CA and the frequency of each sense.
Correspondence analysis shows clearly that Japanese language learners use “over” differently from English native speakers in terms of its senses. There are some senses which have a strong connection with the learners, such as covering, completion, and repetition senses, while control, focus-of-attention, transfer, preference, the-other-side-of senses have a strong connection with native speakers. In addition, by focusing on the senses which are no occurrence at the beginner level and which are produced little even at the advanced level, this study has succeeded in explaining which senses are difficult for language learners to produce. This study revealed that the protoscene and the senses which are in cluster 4, 5, and 6 in the polysemy network of over by Tyler and Evans (2001) are especially difficult for Japanese language learners. These findings might help teachers when they teach the word “over” to Japanese language learners of English, and help researchers when they assess the learners’ production of English.
Reference
Tyler, A., & Evans, V. (2001). Reconsidering prepositional polysemy networks: The case of over. Language, 724-765.
Presentation video
Supplementary Information
None
Q&A live (Zoom) session
No longer available.
This looks to be a very interesting paper. I hope you enjoy the conference! – Organizing committee
Thank you very much for your encouragement!
Thank you for this great presentation!
Have you or do you have any plans to examine Japanese learners’ unidiomatic use of OVER? It seems like quite a few occurrences of the word were excluded in your study because they did not correspond to native-like usage. Perhaps there are interesting patterns worth exploring further?
My PhD project is about the language of school EFL textbooks used at secondary school level in France, Germany and Spain (so not Japan sadly) so naturally I was wondering how many of these many senses of OVER are actually explicitly taught or even just featured in pedagogical materials. This would also make a very interesting follow-up study!
Thank you very much for your comment and questions!
“Have you or do you have any plans to examine Japanese learners’ unidiomatic use of OVER? It seems like quite a few occurrences of the word were excluded in your study because they did not correspond to native-like usage. Perhaps there are interesting patterns worth exploring further?”
I excluded the examples when the native speakers judged the word “over” was used ungrammatically, such as “I hope to study English over seas after ~~~” and “All the club talked over the concert.”
So basically, the excluded examples are the errors I think. As you said, exploring such kind of errors may be very interesting! Although there are some sentences which do not make sense at all and cannot understand what the learners wanted to express, the further exploring may reveal when JLEs use “over” in different meanings from native speakers.
“My PhD project is about the language of school EFL textbooks used at secondary school level in France, Germany and Spain (so not Japan sadly) so naturally I was wondering how many of these many senses of OVER are actually explicitly taught or even just featured in pedagogical materials. This would also make a very interesting follow-up study!”
Actually, I am building a corpus of English textbooks which are used in Japanese public high schools, so yeah! It must be interesting to examine how the word “over” is taught in those textbooks. Thank you very much for your comments!
Thank you for an interesting talk. A couple questions and comments. (1) Do you think JLE need to use “over” the same way as ENS do? (This is not criticism, but the topic I have thought for many years). (2) 12’35” Any comments on the interpretation of the Axis 2 (It explains 24% of the variance)? Interestingly, Beginner and Advanced are in the lower half, while Intermediate is in the upper half. (3) Maybe you say, if ENS use something more often than JLE, it is difficult. Otherwise, it is easier. However, is this always the case? (Shin Ishikawa Kobe U)
Ishikawa sensei, thank you for your comment and questions!
(1) Do you think JLE need to use “over” the same way as ENS do?
As I mentioned in the presentation, I must admit that it is pretty inappropriate to argue that language learners have to write/speak like native speakers. Then, I’m also wondering what the goal or the ideal language behaviour for language learners is, or what kind of language we can set for the role model? (does it make sense?) So, to deal with this anxiety, although I compared the frequency of each sense between proficiency level, I avoided comparing it with native speakers.
(As for the number of the senses of “over”, in my opinion, more senses language learners acquire, the more helpful for them to develop their English ability.)
(2) Any comments on the interpretation of the Axis 2 (It explains 24% of the variance)? Interestingly, Beginner and Advanced are in the lower half, while Intermediate is in the upper half.
When I saw the CA chart for the first time, I was also surprised with the result in terms of what you mentioned. I think the biggest reason why INTERMEDIATE is located on the upper side is the effect of 5A1. 5A1 is “over and above” sense, and there are 10 examples in INTERMEDIATE. Actually, 6 of them are used in the phrase “over population”. I think it might have effected too much on the CA result. I think this happens because of the theme of the writing test/activity.
(3) Maybe you say, if ENS use something more often than JLE, it is difficult. Otherwise, it is easier. However, is this always the case?
When I discuss which senses are “easy” or “difficult” for language learners, I only compared the frequency between the proficiency levels. Actually, language learners at all the groups use “over” one-third frequently as much as native speakers, so if I had compared the frequency of each sense with native speakers, most of the senses would be judged as “difficult” for language learners!
Thank you for your clarifying comment! A good work!
Japanese people seem to use ‘over’ less frequently and less correctly. Do Japanese use other prepositions more frequently? Do they use other prepositions incorrectly? Do Japanese tend to omit the use of ‘over’?
The usages of ‘over’ in fixed expressions are easier to use. In contrast, ‘over’ is also used in describing spatial relationship without involving the use of fixed expressions. According to your research, is it difficult to use this type of ‘over’? This type of ‘over’ cannot be used correctly without understanding the differences between various schematic meanings of other prepositions. Therefore, understanding the meaning of ‘over’ is difficult with the mere understanding of how it is translated into Japanese. Moreover, such factors seem to make it difficult for Japanese to use this word in the description of a proto-scene.
@Risako, thank you for the talk! I’m quite intrigued by the methodology you chose. Would you say that Linguist 3 is more qualified than Linguist 2? Also, if all the linguists disagree so much, would it not be better to remove that item from the study?
As for the correspondence analysis, is this method valid when the frequency counts are so very low?
Finally, the frequency table you show later seems to be simple a normalized count of the same frequency data that you used to create the correspondence table. Can you explain how the two tables differ except in terms of normalization?
Thank you again for your thought provoking talk!