The Setouchi Triennale 2025 in numbers

Hello everyone,
I hope this message finds you well. While the islands are enjoying some well-deserved rest, the Setouchi Triennale Executive Committee has released official numbers regarding the 2025 Setouchi Triennale. (Source)

I'm sharing a few of them here with clarifications and personal comments.

Artworks

First, the total number of artworks was 256, and there were 25 events.

I don't remember the exact numbers from previous years, but 256 seems like a good number. If my count is correct, I saw 148 of them this year, not counting the permanent art that I had seen before but not in 2025.

Visitors

The Triennale is proud to announce that 1,084,128 people attended this year.

This number is much higher than in 2022, when attendance was much lower due to the pandemic. However, it is lower than in 2019, when attendance was 1,178,484. I was surprised because almost everywhere was much more crowded than usual. However, it's true that the lines to enter some art sites were smaller than I expected.

One reason the number of visitors is slightly lower than in 2019, despite the increase in foreign visitors (more on that in a few paragraphs), is probably the Osaka Expo, which attracted more people, especially casual Japanese visitors who had already been to the Setouchi Triennale and didn't feel the need to return this year.

Another reason is that the economy is so bad right now that many Japanese people may have decided to save their money this year. As you know, that's definitely one of the reasons I visited less often than in previous years.

Major caveat about these numbers

All right, it's time for a major caveat about these numbers. I've noticed that Japanese official reports often include numbers with little detail about how they were calculated.

This is definitely the case with the Setouchi Triennale. Let me explain. It'll make sense if I show you the attendance numbers for each island and location.

• Naoshima: 338,459
• Teshima: 145,364
• Megijima: 61,327
• Ogijima: 56,620
• Shodoshima: 140,121
• Oshima: 8,204
• Inujima: 38,407
• Takamatsu: 47,444
• Uno Port: 21,944
• Seijima/Shamijima: 31,551
• Tsuda: 32,090
• Hiketa: 34,690
• Utazu: 42,783
• Honjima: 25,781
• Takamijima: 20,666
• Awashima: 21,660
• Ibukijima: 17,017

Add all those numbers together and you'll get 1,084,128!

Do you see where I'm going with this? To get this number of roughly one million visitors, they simply add up the number of visitors from all the locations. But that assumes each visitor only goes to one place, which is obviously not the case.

Also, I'm not sure how they came up with the numbers for each location. Previous reports mention that these numbers are estimates based on data collected at various collection sites. The sites' locations are determined according to factors such as the distribution of artworks in a given area, as well as viewing routes.

Based on what I saw and what I know about islands such as Megijima, Ogijima, or Takamijima, I'd say the numbers for each island seem accurate.

But this "million visitors" thing is definitely not accurate. It'll be more proper to call it "visits" rather than "visitors."

One way to make the visitor count more accurate would be to see how many Art Passports were sold. This number has been published: 107,098 Art Passports.

Does that mean only 107,098 people actually attended the Triennale? No, it doesn't. While most visitors to the "small" locations had an Art Passport, I suspect many tourists who only spent one or two days on Naoshima or Teshima didn't buy one.

What is the actual number of visitors, then?

Your guess is as good as mine. I'd say 107,000 is a reasonable estimate for the smaller islands, but 338,459 visitors to Naoshima is also probably accurate. The question is how many of those visitors had an Art Passport. I'm not sure if it's possible to know the answer.

Overall, the way the festival is designed—with multiple locations and entry points, and the option to purchase an art passport—makes it nearly impossible to accurately count how many people came to visit at least one artwork.

Survey

The results of a visitor survey were also published.

With only 12,700 respondents, I'm unsure of its accuracy, and I doubt it sampled people the way official polls do. Still, I find a few of the results interesting, and they're probably close to reality.

Gender

Of the visitors to the Triennale, 63.7% were female and 35.4% were male. That seems accurate. While the ratio is probably closer to 50-50 for foreign visitors, there have always been more female Japanese attendees at the Triennale (and at any cultural event, really), for both cultural and practical reasons.

Age groups

The survey indicates that people in their 30s, 40s, and 50s comprise the largest groups, each accounting for around 20% of the total. At first glance, this seems accurate. However, I also feel that the numbers for children and teenagers (3.7%) and older people (12.7% in their 60s and 4.2% aged 70 and over) are too low. It makes sense for children; they are much less likely to respond to a survey. I'm not sure about older people; maybe they care less about responding to surveys. In any case, these results suggest that the survey sample may not be representative of the actual population that visited the Triennale.

Geographical origin

• Kagawa Prefecture: 27.4%
• Okayama Prefecture: 9.4%
• Other Domestic Areas: 40.2%
• Overseas: 23.0%

It feels about right, except for one thing. The number of foreign visitors should be much higher. It was already around 20% in 2019, and there were many more foreign visitors this year than six years ago. I would guess that it was close to 80-90% on Naoshima, but even on the smaller islands, I remember some days when about half of the visitors were from overseas. I wouldn't be surprised if the actual proportion of foreign visitors were closer to 40%.

I don't have any data yet on country representation, but based on my observations, visitors from the "Three Chinas" (my term, not intended to cause a diplomatic incident) — mainland China, Hong Kong, and Taiwan — make up the vast majority of foreign visitors (well beyond 50%). This has always been the case, but even more so this year.

As a reference, in 2019, 34.9% of visitors were from Taiwan, 26.7% from mainland China and 11% were from Hong Kong.

First time?

Yes: 54.1%
No: 45.9%

It has been around that 50/50 divide for a few Triennales now.

Online

• Official website visits over the past 12 months: 2,431,434 (20% of which were from abroad).
• Meanwhile, my website, Setouchi Explorer, received approximately 60,000 visits. I believe this is the highest number of visits in a 12 months period to date; however, I think 2019 had similar numbers. I wish I had kept track of them.
• Japan Art Islands, this newsletter, was started in February 2025 and currently has 397 subscribers. I didn't start from scratch, as approximately half of you were "imported" from my Setouchi Explorer subscriber list. In any case, thank you all for following me, whether you've been with me since the beginning or just arrived this year.

More numbers

Guidebook

About 52,000 copies of the official guidebook were sold. That number feels low to me, but I suspect that few foreigners bought it. Not to mention, there are more and more online resources and an app. Still, that's an average of one guidebook for every two Art Passports, which seems realistic.

Volunteer Staff

The Koebi-tai, the volunteers who essentially run the festival, had 6,466 members this year. Some members have been involved since 2010; others are members for just a day or two. This is fewer than in 2019, when there were 7,165 members.

The report also mentions 2,560 corporate and organizational volunteers from 142 companies and organizations, but I'm not sure what that means. If sponsors sent staff, I'm not sure where they were. Perhaps with the Koebi?

There were 688 residents. I assume this refers to island residents. Indeed, on some islands, especially the western islands, most volunteers are locals.

Official tours

A total of 2,300 people participated in 88 official tours, which were 70.4% full (that's higher than in 2022).

Art Passports

As I mentioned, they sold 107,098 art passports. However, the report states that 44% of these were digital, which surprised me greatly. At most art venues, I felt like one of the very few people with a paper passport. Well, this number has to be more accurate than my perception.

Conclusion

I've always found these numbers interesting. Perhaps it's because statistics was one of the few areas of mathematics I enjoyed as a student.

If you like numbers and statistics, I hope you enjoyed this post. If not, I doubt you're still reading.

In any case, that's all for today!

The next post will probably be similar to previous ones, so stay tuned. If you haven't subscribed yet, now is the perfect time!

Thanks for reading and see you soon.