Breaking the Tozaburo Yanagiya Method into seventeen acts, and checking each against the literature
Rakugo has been reported as a tool for language teaching since 1993 in the Japanese-language-education literature. Almost all of that work, however, consists of practice reports. Why it works — the structure underneath it — has rarely been analysed. This paper breaks the teaching procedure of the author's own online Rakugo classes into seventeen distinct acts, and checks each one against the existing research.
The result is that rehearsing a single Rakugo piece turns out to contain, simultaneously, elements whose effects have each been verified independently: reading aloud, memorising, repetition, acting, swapping roles, gesture, performing to an audience, and laughter. On acting in particular, a study that compared acting training directly against training in visual art and in music found that gains in empathy and theory of mind appeared in the acting group alone. That is the existing literature answering, directly, the question of why this art form rather than another.
At the same time, no measurement data exist for this method itself. What follows is structural support drawn from existing research. It is not evidence that this programme works. That distinction is stated explicitly in sections 6 and 8.
Keywords Rakugo ／ language education ／ heritage language ／ Readers Theatre ／ drama-based pedagogy ／ role reversal ／ Tozaburo Yanagiya Method
Using Rakugo to teach language is not a novelty. In Japanese state education, Zoro-zoro by Ensō Sanyūtei has appeared in the fourth-grade national-language textbook since the 1999 school year, and Rakugo has been carried as teaching material ever since. In class, children follow how the scene develops, decide where the story peaks, read it aloud, and act it.
In Japanese-as-a-foreign-language teaching, practice reports have accumulated as well. Kawasaki (2014) sorts them into four types: the linguistic-analysis type, which studies vocabulary and sound in fine detail; the appreciation type, whose aim is enjoyment; the performance type, which sets performing as the final goal; and the composition-and-performance type, in which students perform a short comic story and then write one of their own.
Kawasaki further argues that the reason Rakugo has been so little used is not the difficulty of the material but that teachers try to approach it as linguistic analysis. Yonemoto and Sogabe (2012) attribute its near-absence at beginner level to "the teacher's assumption that learners will find it too hard." In practice, Sakai (2001) needed only a single seventy-five-minute preparation session, after which international students reported that they had understood three pieces and found them funny.
This paper concerns the last two of the four types — performance and composition-and-performance. That is: not a lesson about understanding Rakugo, but a lesson in which Rakugo is the means of thinking about how to reach another person.
The aim of this paper is to show, structurally and against the existing literature, that the author's programme is sound as a means of language education.
The method was as follows. From chapter two of the author's book A New World of Rakugo ("A New Teaching Method") and from the procedure currently in use, the acts that a learner actually performs during rehearsal were extracted. Seventeen were identified. For each, prior research was then searched across education, second language acquisition, cognitive psychology and medicine, and the presence and size of any effect was recorded.
Strength of evidence was graded in three levels. Strong — confirmed by meta-analysis or a large cohort study. Moderate — consistent results across several individual studies. Weak — no direct evidence, theoretical coherence only.
Rehearsal in this method takes the form of a six-stage cycle, shown in Figure 1.
Two features stand out. First, memorising comes before repetition. In ordinary foreign-language teaching, memory is what repetition eventually produces. Rakugo rehearsal finishes the memorising first — "to the level where it comes out automatically, without thinking" (Hatasa & Kubota, 2009) — and only then repeats, in order to work on feeling. The order is reversed.
Second, once a piece is finished, the roles are swapped and it is performed again. The same script, from the other character's side. This is the core of the method, and — as section 5 sets out — the part that carries the greatest weight in the literature.
Stage 3 preceding stage 4 is the reverse of ordinary foreign-language teaching, where memory follows from repetition.
Broken down finely, the stages of rehearsal resolve into seventeen acts. Table 1 gives each act, the corresponding prior research, and the strength of the evidence.
| What the learner does in rehearsal | Corresponding research, and what it found | Evidence | |
|---|---|---|---|
| 1 | Read the script with correct reading and pitch accent | Systematic review of twenty-three Readers Theatre studies. Gains in reading skill, plus attitude and motivation towards reading | Strong |
| 2 | Say it aloud | The production effect. Words spoken aloud are remembered 10–20% better than words read silently (MacLeod & Bodner 2017) | Strong |
| 3 | Memorise until it comes without thinking | Hatasa & Kubota (2009). Research on actors' memory — active experiencing (Noice & Noice) | Moderate |
| 4 | Repeat the same lines | Meta-analysis of eleven Readers Theatre studies, large effect size (Mastrothanasis et al. 2023). Spaced retrieval aids vocabulary retention in children | Strong |
| 5 | Play the characters | Meta-analysis of forty-seven studies of drama-based pedagogy: significant positive effect on achievement (Lee et al. 2015) | Strong |
| 6 | Swap roles and play it from the other side | Goldstein & Winner (2012). Empathy and theory of mind rose significantly in the acting group only. No gain in visual art or music | Strong |
| 7 | Gesture with the fan and hand towel | Children who gestured themselves retained the learning. Watching someone else gesture did not produce it (Goldin-Meadow) | Strong |
| 8 | Manage breath and pause, project the voice | Slow breathing raises vagally-mediated heart rate variability (meta-analysis) | Moderate |
| 9 | Take making the listener laugh as the goal | Hatasa & Kubota (2009): "if it is done well the audience really does laugh" — an unambiguous goal. Humour lowers anxiety and raises motivation | Moderate |
| 10 | Perform in front of people | After ten weeks of improvisational theatre, 43% of students who screened positive for social anxiety in week one screened negative by the final week (Felsman et al., ~270 adolescents) | Strong |
| 11 | Watch the recording and correct oneself | Video self-modelling, twenty-three studies, effect size 0.83. Watching one's own successful attempts is the more effective form | Strong |
| 12 | Having learned it correctly, then change it freely | A two-stage structure: form first, individuality permitted after. Coherent with Hatasa's "memorise until automatic", but not directly tested | Weak |
| 13 | Keep going, one hour a week | Drama-based pedagogy shows stronger effects beyond five lessons (Lee et al. 2015). In Rakugo-based health classes, groups attending more often improved more (Ohira et al.) | Strong |
| 14 | Share the roles among several people (Minna de Rakugo) | Composing and rehearsing as a group contributes to autonomous and collaborative learning (Hatasa & Kubota 2009) | Moderate |
| 15 | Never alter the cue line handed to the other player | The discipline of ensemble playing. Consistent with turn-taking and joint attention, but not directly tested | Weak |
| 16 | Learn the culture from words that appear in the story | Maintaining a heritage language supports parent–child communication and the formation of belonging and identity (Japanese Society for Mother Tongue, Heritage Language and Bilingual Education, and others) | Moderate |
| 17 | Receive a stage name and belong to a school | Relatedness, in self-determination theory. Belonging sustains intrinsic motivation | Moderate |
Of the seventeen acts, eight were graded strong, seven moderate, and two weak.
The most important finding in this paper is that the existing literature answers this question directly.
Goldstein & Winner (2012) compared children and adolescents trained in acting against children and adolescents trained in visual art and in music. What they measured was empathy and theory of mind — the capacity to infer another person's mental state.
The result was unambiguous. Empathy scores rose significantly in the acting group only. Neither art nor music produced a rise. In a second study with adolescents, the acting group alone also improved on a naturalistic measure of theory of mind (the Empathic Accuracy Paradigm). The authors take this as evidence that these capacities are not fixed by age three or four, but can still be developed later, through playing a role.
The significant gain in theory of mind was found in the second study, with adolescents. The design is quasi-experimental; participants were not randomly assigned.
This result is decisive for the present programme because Rakugo is, among forms of acting, the one specialised in changing roles. A single performer plays several characters by turning the body left and right — kamishimo. And in this method, once a piece is done, the roles are swapped and it is played again (Figure 1). The same conversation is necessarily experienced from both sides.
One caution is required. No study was found, within the range of this search, that tested the practice of swapping roles and performing again. That is not a weakness of the method; it is a gap in the research. Section 8 returns to it.
The studies introduced here are not evidence of any effect of this programme. They are offered as context for the art form itself. This is an education programme. It is not intended to diagnose, treat or prevent any disease.
Large-scale epidemiological work on laughter and health has accumulated in Japan. In a six-year follow-up from JAGES (the Japan Gerontological Evaluation Study), 1,240 new cases of dementia were observed among 12,165 community-dwelling adults aged 65 and over. After adjustment for age, sex, education, income, depressive tendency, medical history, smoking and social participation, the group with the most occasions for laughter had a 26% lower risk of dementia than the group with the fewest (Wang et al. 2022). The response options for "occasions for laughter" in that study explicitly include going to see Rakugo or theatre.
In a survey by Ohira and colleagues of 2,471 residents of Osaka Prefecture (mean age 59), the relation between frequency of laughter and symptoms of cognitive decline was as shown in Figure 3.
The odds ratio for those with almost no occasion to laugh, against those who laugh almost daily, was 2.15 (95% CI 1.18–3.92, p = 0.01). Adding smoking and depressive symptoms to the adjustment attenuates it to 1.84 (95% CI 1.00–3.48, p = 0.05), with the same direction. Source: Tetsuya Ohira et al., research funded by the Nihon Seimei Foundation. This is an observational study and does not establish causation.
In the intervention arm of the same research, a six-month "laughter and health" class including listening to Rakugo was run with forty-six older residents. Among the forty analysed, the "general health" score of health-related quality of life rose significantly (p < 0.05), and the group attending fortnightly improved more on depression score and quality of life than the group attending monthly. Scores on the cognitive screening test (MMSE) did not change — 28.1 both before and after — because participants were within the normal range to begin with.
The Osaka International Cancer Institute has also run a randomised controlled trial of the effect of live Rakugo and manzai on immune function and quality of life in cancer patients (the WAROTEMAE 2018 study, sixty participants). The preceding iOSACA study reported that fortnightly laughter was associated with increased IL-12B production capacity, improvement in quality of life — particularly pain symptoms — and suggested gains in cognitive function.
What these show is that Rakugo has reached the stage of being used as an intervention in medical research. To repeat: this is not an effect of the present programme.
What Table 1 shows is that a single act — rehearsing one Rakugo piece — contains at the same time elements that would ordinarily be sold as separate lessons. Table 2 sets this out.
| What is gained | Where it is usually found | In Rakugo rehearsal |
|---|---|---|
| Reading aloud; retention in memory | Recitation classes | Stage 2, every time |
| Nerve to speak in front of people | Speech or drama classes | Stage 6 — the performance |
| Reading other people's feelings | Acting training (art and music will not substitute) | Stages 5 and 6, and the role swap |
| Japanese sound, accent and phrasing | Language classes | Stage 1 — it begins from correct reading |
| Japanese culture and the year's customs | Culture classes, weekend schools | Whenever a word in the story calls for it |
| A reason to continue; belonging | — | The weekly performance, the ranking board, the stage name, the school |
There is a second finding that matters. The meta-analysis of forty-seven studies by Lee et al. (2015) identifies two conditions under which drama-based teaching produces stronger effects: when more than five lessons are given, and when the classroom teacher (or the researcher) leads, rather than a visiting teaching artist.
The intervention study by Ohira et al. points the same way: those who attended more often improved more. The conclusion is that a continuing class has the advantage over a one-off performance or guest session. That this programme is built as a weekly online lesson, with every lesson recorded and archived, is consistent with that finding.
In heritage-language learning, sustaining motivation is often the hardest part. Rakugo has the property Hatasa & Kubota (2009) identified: there is an unmistakable purpose — to make the listener laugh — and if it is done well the audience really does laugh. The assessment is not a score. It is whether the room laughed. A child can read that judgement without help.
And because making a listener understand demands linguistic accuracy, the need to rehearse arises from inside the task. Repetition stops being homework and becomes the means of getting the laugh.
The limitations are stated plainly. They are not weaknesses in the sources; they are the nature of the sources.
That last point is, read the other way round, a gap in the research. This programme holds 243 folders of lesson recordings, roughly 26 gigabytes, going back to around 2021. Recording the same short comic story at enrolment and again three months later, and logging three measures — the proportion of words produced, the number of stumbles, and the time taken — would be enough to yield the first data against that gap. That is the next task.
Rehearsing one Rakugo piece contains, simultaneously, elements whose effects have each been verified independently: reading aloud, memorising, repetition, acting, swapping roles, gesture, performing, and laughter. Of the seventeen acts identified, eight are supported by meta-analysis or large cohort study, and seven by consistent results across several individual studies.
Above all, the finding that acting training produces gains in empathy and theory of mind that art and music training do not answers directly the question of why this art form rather than another. Rakugo is, among forms of acting, the one specialised in changing roles — and this method builds swapping the roles and playing it again into the structure of rehearsal itself.
That continuing beyond five sessions has been identified as a condition for the effect supports the design of this programme as weekly continuing practice rather than a single experience.
This paper has shown that using Rakugo for language education is not an ungrounded idea but a practice structurally supported by research across several fields. What remains is to begin measuring the programme itself.