[Special Feature: Thinking About English Education] Ippei Inoue: Things to Consider for the Future of English Education

2025/05/08

The standard image of Japanese people's English proficiency is likely that they can read and write but are poor at conversation. The stereotype of a Japanese person was someone who would run away or freeze when faced with a situation where they had to speak English with a foreigner.

I say "was" because it now appears that a reversal is occurring. One rarely sees today's children or the younger generation becoming intimidated in front of foreigners. On the other hand, it has been pointed out from various quarters that reading comprehension—including Japanese and among adults—may be declining in general. In fact, in schools, while there is an increasing number of students who feel less resistance to so-called English conversation, there are also concerns about a decline in the ability to accurately understand the content of textbooks.

What do these changes mean? As society and the environment and forms of communication themselves change significantly, it seems that the way English education should be is also being forced into reconsideration.

The reason I received the request for this article is likely because my specialty is sociolinguistics, English linguistics, and communication studies, and I have had several opportunities to speak out about Japanese people and English. Furthermore, I have a background as the principal of Keio Chutobu Junior High School. Based on that position, I would like to share some of my thoughts.

Recently, I have had more opportunities to be asked, "Since we have AI, don't we no longer need to study English?" Translation apps on smartphones evolve daily, and many people say they hardly feel the language barrier even when traveling abroad. In fact, English proficiency that used to be demonstrated when asking for directions has been replaced by smartphones. I seem to emit an aura that makes it easy for people to ask me for directions, and I used to be asked frequently, but recently that has stopped completely. This is because every traveler has a smartphone.

Even when reading academic papers in specialized fields, it seems possible to obtain a lot of information in a short time by using AI translation and summarization functions. This technological evolution certainly lowers language barriers and contributes to the democratization of international information access. If smartphone compatibility is not an issue, then studying English might indeed be "unnecessary." Even in online meetings, the accuracy of transcription has improved, and even if the other party is speaking English, it is translated instantly. While AI cannot yet handle instantaneous interaction, one can at least be a passive participant.

However, AI translation technology also has its limits. In terms of understanding nuances, cultural backgrounds, and specialized contexts, there are still parts that do not reach human linguistic ability. Mistranslations and unnatural expressions are not uncommon, and relying solely on AI translation in important meetings or negotiations still carries significant risk. Furthermore, in communication mediated by AI, the natural rhythm and "pauses" of dialogue tend to be lost. Depending on the position or situation, one cannot yet say that English training is unnecessary. English proficiency remains a differentiating factor. There is English that AI can replace, and English that it cannot.

Some point out the cognitive changes brought about by dependence on AI technology. There are reports that constantly relying on translation tools may lead to the loss of secondary effects such as brain activation and improved cognitive abilities brought about by foreign language learning. Language learning is not just the acquisition of a communication tool, but is also considered a process of fostering frameworks of thought and sensitivity to different cultures. In the future, hybrid communication skills that utilize AI technology and human linguistic ability complementarily may be required. It will be important to demonstrate human-specific linguistic sense and intercultural understanding while effectively mastering AI tools.

Keio Chutobu Junior High School, where I served as principal for six years until last year, is a fun school characterized by a high density of communication between students and teachers, and among students themselves. Of course, being just fun is not enough, so coordination with the high school is one of the issues that is constantly kept in mind. While teachers have great discretion regarding lesson content, the awareness shared by all teachers is to "provide the basic skills necessary to withstand classes in high school."

Many Chutobu students are proactive about communication. As part of the school culture of Chutobu, the distance between teachers and students is close, and many conversations are exchanged daily. Teachers often spend a long time interacting with students. From the perspective of a university faculty member, it is even moving. I came to the thought that love is time, and I often spoke about this to parents at parent-teacher meetings. This is conveyed to the children, and they communicate well. Before graduation, teachers warn third-year students, "When you go to high school, they'll be surprised if you're that clingy with the teachers." That is how dense the daily communication between students and teachers is.

I see the proactiveness of Chutobu students toward communication as something cultivated not only in classroom activities, extracurricular activities, and school events, but also through standing chats in the hallways and small talk in between. Perhaps because of this influence, students show an attitude of trying to speak without hesitation even in English. While the relatively high number of native teachers is a factor, I believe that without this fundamental orientation toward communication, linguistic ability in general—not just English—will not grow. Doing debates and presentations is important, but that alone will not improve communication skills. What is important is the students' low psychological hurdle to using English and their orientation toward communication.

Student exchanges also play an important role that is slightly different from their outward appearance. Chutobu currently conducts exchanges with Iolani School in Hawaii and Pocklington School in the UK. The number of students sent is not large compared to the whole. While it is a valuable experience for the students who can go, looking at it in total, the significance of receiving students might be greater. When receiving them, the exchange students are in the classroom, but what is more important is recess and after school. Here, students communicate with English speakers who are not evaluators. Whether their English is good or bad, the communication skills of Chutobu students are in full swing, and they go and engage with them. Furthermore, on holidays, the Chutobu student from the host family invites them to go out somewhere, and without fail, that student's friends also tag along and go together.

The exchange family is one-on-one, but when friends are included, the ripple effect is large. Interaction with international students is not just an opportunity for language acquisition, but can be a valuable experience for broadening intercultural understanding and international perspectives. The significance of interacting with peers from different cultural backgrounds is also great. Conversely, it can be an opportunity for Chutobu students to objectively re-examine their own culture and values. With the development of online technology, it has become easy to continue interaction even after the international students return home. If a relationship of mutual learning across borders is to be built, I feel that the basic foundation must first be the meeting of flesh-and-blood human beings, rather than starting suddenly online.

This is a slightly painful topic from the perspective of Chutobu, but when students move up to high school, the difference in academic ability with the so-called "exam group" becomes an issue. The high schools belonging to the affiliated schools of Keio University have very high academic standards for students who enter through external entrance exams, and their deviation scores are at the top level nationwide. Many students feel they will lose in terms of academic ability, and that is the concern of the Chutobu teachers who send them off.

However, if you think about it for a moment, it is obvious that the students in the exam group did not enter these difficult high schools solely through learning during junior high school class hours. They have spent a lot of time outside of school learning on exam preparation, and attending Juku or home study plays a major role. Of course, there are students in the internal advancement group from Chutobu who attend Juku, but it is estimated that there is a difference in the ratio and the amount of time spent. Therefore, it is natural that a difference in academic ability is seen between students who went through exams and those who did not at the point they advance to the same high school. Simply put, there is a reality that the quantity and quality of learning within school class hours has limits to begin with, and based on that, we must consider what and what kind of opportunities the school can provide. I do not think exam study is bad. Basic academic ability is also important. On the other hand, there is meaning in the school being an entity that can provide some kind of good environment. Students' thoughts and types are also diverse.

Not limited to English, there are individual differences in how languages are learned. I feel that differences that could be called "cognitive styles," rather than age, influence the entry point. Some learners try to grasp language through understanding grammar and structure, while others are the type to acquire it through the joy of communication.

It has been a long time since English education was introduced in elementary schools, but a phenomenon that became apparent once it became a regular subject is that a small number of students appeared who are remarkably poor at English. It is a level of dislike where they hate even looking at the alphabet. Even if there are students who were poor at math, they can probably do simple calculations, and I doubt there are students who hate even looking at the + or - symbols. While it is a source of pain for the individual, looking at it as a researcher, the existence of students who have such an extreme dislike is interesting as a phenomenon.

Although it is a rare case, there are students who were extremely poor at English in that way but become able to do it if they start from grammar and are taught logically. If the entry point is different, they may be able to enter. This suggests the possibility that there are multiple cognitive types of learners in foreign language learning.

I once conducted a simple survey myself, and it seems this is not a matter of age, such as being a junior high school student. It is actually not rare to find people who have bitter past experiences, such as hating being made to sing songs or finding it agonizing to be made to play games during English class in elementary school. Whether they are elementary school students, junior high school students, middle-aged, or elderly, there are people who are poor at songs and games as an introduction to language. And such people often prefer to enter through grammar.

People who can enjoy communication itself should start from there. There are also people for whom it is better to start from grammar and vocabulary. Some people like reading literary works such as novels. If there is a motivation such as wanting to understand Major League Baseball broadcasts in English, the path for language learning becomes visible.

Looking at the educational field, it is ideal to provide such a diversity of approaches. What is needed is not a reform of the curriculum, but a diversification of the curriculum. I would like to expect AI to handle that troublesome processing of the system.

The phrase "Japanese people can read and write, but..." is a standard introductory remark used to criticize Japanese people's English proficiency, but in fact, this has been suspicious for a long time. Comparing through exams like TOEFL is not necessarily appropriate, but even looking at average scores by country, Japanese reading/writing scores have never been high compared to other countries (and speaking is indeed low as well). For many Japanese people, "being able to read and write" just means they can understand the meaning of simple things when they see the characters; when it comes to long or content-rich texts, there are not many people who can read properly at a certain speed.

Recently, many people are having generative AI summarize texts, but based on a rough impression from a user's perspective, I don't think AI is very good at summarizing. It can be used for administrative documents, simple documents, or minutes, but when it comes to content-rich texts that require deep reading, it still seems difficult at this point. I don't know how AI summarizes, but even if a passage is impactful and full of implications if read carefully, if it is short in quantity and not repeated enough to become a keyword, AI seems to sometimes overlook it.

At least for reading at this level, humans are still superior. As attempts targeting entrance exam questions, improving scores on reading comprehension problems remains a difficult challenge for both the former "Todai Robot Project" (a project by Noriko Arai and others at the National Institute of Informatics to give AI enough ability to pass the University of Tokyo entrance exam) and recent generative AI.

In that sense, thinking as a faculty member of the Faculty of Letters, improving reading comprehension at that level (in any language) is the mission of university education. And that will surely become an ability that can overwhelmingly differentiate one from the coming generations and groups who are concerned about over-relying on AI, serving as a weapon that will not lose to AI.

While my specialty, linguistics, might be an exception, there seem to be people in the world who hold a not-so-good image of AI regarding literary researchers and the humanities. However, the ability to read comprehensively and deeply, from between the lines to the background and surroundings of a text, will gain high value precisely in the AI era. AI development may reach that level someday, but literary researchers are generally modest, unlike me, and do not seem to verbalize such advanced reading abilities much. Since it is a skill that has not been verbalized, I think it will probably be difficult for AI researchers to get their hands on it for the time being. Conversely, that might be an area where literary researchers can make some contribution to the development of AI.

With the development of generative AI, the level of automatic translation has also improved significantly in recent years. For linguistics as well, translation is an old yet new theme. In 2024, what was shocking to me was the American Japanese period drama series "SHOGUN," which was broadcast and streamed that year. It was a huge hit and won numerous awards, including 18 Emmy Awards. That is not the shocking part. What was surprising to me was that the majority of Americans and people around the world were watching this drama in Japanese with subtitles in English or other languages.

A good object of comparison is "The Last Samurai" from 2003. This work was also a huge hit, but the big difference was that the samurai with topknots all spoke English.

Generally, it was said that Americans do not watch many foreign films. Furthermore, it was previously assumed that there were few admirable people who would try to watch them with subtitles. Perhaps triggered by the stay-at-home demand due to COVID-19, the style of watching movies at home via streaming has become established. The factors for the hit may be varied.

From the perspective of Japanese people and English, what this phenomenon signifies is the potential for Japanese content—not just movies, but also YouTube videos that anyone can enter in the future—to become global content by adding English subtitles.

That English is probably not natural English from the perspective of British or American people. In an English composition class, it might be said that such English, which looks like a literal translation of Japanese, is no good. However, if not only the English-speaking world but the whole world prioritizes enjoying the content itself and comes to tolerate such Japanese-English, English will enter a dimension of globalization different from before. It is a new globalization of English created by AI, different from the era of "World Englishes," where English spread across the globe and created unique versions of English in each region.

If that happens, what will be required is content power. I think such an approach to English is also acceptable.