What is the ⿻ theory of change?

The book *Plurality*'s view that instead of overturning the world all at once, new ways of doing things should first be tried in communities of around 100,000 people with many different kinds of members, and that many such small successes should be built and spread while being loosely linked together.

Even if you have a good idea, trying it out on a whole country at once is dangerous. If it fails, the damage is too great. But testing it only among a few close friends will not tell you whether it works for anyone else. The "theory of change" in the book ⿻ Plurality is a view about what size of group to try new ways of doing things in, and how to spread them.

In short

It does not try to overturn the world all at once. New ways of doing things are first tried in groups of around 100,000 people, with many different kinds of people inside them. Many such small successes are built, and spread while being loosely linked to one another [1][2].

The book likens the groups where things are first tried to the seeds of a plant, calling them "seeds".

For example

Why 100,000? (the book's calculation, not a made-up example). Taking the world's population as roughly 10 billion, the book gives its square root, about 100,000, as a rule of thumb [2]. Since 100,000 × 100,000 = 10 billion, 100,000 groups of 100,000 people would add up to the whole world. At this scale, the diversity within a group and the diversity across groups become roughly equal.

As examples on this scale, the book names middle-sized municipalities, the employees of a large corporation, a Catholic diocese, a large university and the fans of a sports club, among others. But it also writes, "There is, of course, nothing magic about 100,000" [2].

Shibuya is just the right size (Tokyo, January 2024). At the event "Glen in Japan", on hearing that the population of Shibuya ward is a little over 200,000, Glen Weyl was delighted that it was exactly right. He explained the book's square-root argument and said that cities are a very good scale for new ⿻ experiments. At the same time, he added that 100,000 is a good scale for experiments, but things also have to spread both above and below it [3][4].

Making the book's own promotion an experiment (2024). The authors also applied this approach to promoting the book: "choosing endorsements to pursue, media to seek coverage in, events to hold, etc." They divided their whole audience along major lines of division: for culture, religious and secular; for politics, the West and the rest; for business, tech and non-tech; and so on. They chose endorsers respected on both sides of each line, and repeated this for smaller groups. The authors call this "⿻ Marketing" [5].

Taiwan and web3 were the first "seeds". The book treats Taiwan (including g0v) and web3 (the world of internet services built on blockchains) as the two most prominent sites of experimentation so far. The book says that both are roughly the same size: in 2021, web3 applications (dApps) had about 1.5 million monthly active users, and the services built by the g0v community reached similar numbers. Both figures count users, not the population of Taiwan. Only a fraction of the web3 users actively took part in the most ⿻-adjacent services, such as GitCoin (a donation platform). Even so, the book judges that the kinds of diversity within the two are radically different [6].

Taiwan's Presidential Hackathon (a competition to solve social problems with technology) uses citizen participation through Quadratic Voting to choose the top 20 teams (as chapter 7-0 puts it; chapter 2-2 of the same book says that participants and citizens were asked to help select the winners using it [7]), and each year promises five teams support for the following year. The book presents this as a mechanism for raising local experiments into national projects. In one case, an environmental group's work monitoring water and air pollution became a nationwide project [8].

In more detail

Why not revolution?

The book cites research finding that no violent revolution of the twentieth century led to a durable democratic government [9].

It then uses the philosopher Hannah Arendt's comparison of the American and French Revolutions. The American Revolution lasted because it grew out of local democratic experiments and built up legitimacy (rightfulness that people accept) first. The French Revolution tried to solve discontent over living conditions by seizing power quickly, and the building of legitimacy could not keep up [9].

What kind of connection breeds new things?

Summing up research from evolutionary biology to linguistics, the book writes that "dramatic innovation thrives in environments where a diversity of 'groups' (e.g. linguistic, economic, or biological) that are internally tightly connected and externally loosely connected interact." More connected or more centralized structures stifle innovation or make it dangerous, and more disconnected ones do not let it spread [10].

Neither testing from above nor scaling fast and leaving it be

The book compares two ways of experimenting [11]:

Experimenting on communities. Experts design a method, test it on a community, and roll out what works everywhere at once.
Experimenting with communities. As in the early computing communities, the people who use the technology also make it.

The book holds that both have strengths and weaknesses. But it writes that problems arise when the latter is combined with "blitzscaling", spreading all at once with investors' money. It cites ChatGPT reaching 100 million users within a few months. This invites harm to those outside the community, use by early adopters as a tool for dominating others, and intensified competition, the book says. So technology should grow in a roughly balanced way across society's major divides, according to the book [11].

Five conditions for a seed

The book lists five conditions for a good "seed" [12]:

Large enough to encompass the diversity the technology aims to bridge, yet small enough to be one of a very large number of experiments.
Early adopters who gain tangible value and want to contribute to building the technology, and who are not so vulnerable that failures would deeply harm them.
Having prestige within some network, or being able to gain it.
Being a robust community with mechanisms for dealing with harm.
Seeds that are diverse among themselves and connected by a loose network of communication.

The book writes that achieving all of these at once is impossible, but that they provide a rough "north star" to steer toward [12].

On to Parts 6 and 7

From this idea, Part 6 of the book chooses sectors outside government in which to "plant seeds": the workplace, health, media, the environment and learning. The Japanese and Traditional Chinese editions list all five sectors, including learning, at this point. The English edition lists five sectors, including learning, at the start of the chapter, but only four, leaving out learning, at this point [13].

Chapter 7-0 confirms that new ⿻ experiments start not from the policy of a single government but spread outward from diverse, mostly medium-sized organizations, and then sets out the role of government [14]. The concluding chapter 7-1 lays out levels of involvement from a thousand deeply engaged people to a billion sympathizers, and sketches a flow in which culture, politics, business and research reinforce one another [15].

What is debated

"100,000" is not an exact number. The book itself writes that there is nothing magic about 100,000, and Weyl said in Tokyo that things need to spread above and below it too [2][3].
Can it really scale up? A review in dialogue form, by Hilma Church-Turing and Stephen Reid in their newsletter, presents the ⿻ vision as a path: realize it on a small scale in places such as Taiwan, California, Japan and Finland; show its effects; grow a base of supporters who have experienced good governance; and eventually reach a stage where the question is no longer whether to do it but how fast it can spread. Yet the review's narrator also points out a tension: because deliberation takes time, the depth of deliberation that makes ⿻ meaningful may be fundamentally incompatible with the scale it aims for. The "reader" in the dialogue is also left doubting whether it can scale [16].
Reservations about their own assessment. The authors themselves write that readers will be better placed than they are, at the time of writing, to judge whether ⿻ Marketing worked [5].

Where it fits in Plurality

This idea sits at the center of chapter 6-0 of the book, "From ⿻ to Reality". It explains why each chapter of Part 6 chose its sector [13].

In Part 7, chapter 7-0 ("Policy") and chapter 7-1 ("Conclusion") carry it to the scale of policy and social movements [14][15].

Its historical backdrop is "The Lost Dao", which describes the US research agency and the early internet community. The first examples of seeds are Taiwan's digital democracy and g0v. The core of the idea lies in collaboration across social difference and supermodularity. What it stands against are libertarianism and technocracy among the three paths.

Sources

Plurality, 6-0, section "The graph structure of social revolutions" — https://www.plurality.net/v/chapters/6-0/eng/ ［CC0］
Plurality, 6-0, section "Fertile ground" and Figure 6-0-C — https://www.plurality.net/v/chapters/6-0/eng/ ［CC0］
Glen in Japan Transcript (2024-01-03, English automatic captions) — https://scrapbox.io/plurality-japanese/Glen%20in%20Japan%20Transcript
WIRED.jp, "デジタル・デモクラシーは多元的テクノロジーによって実現する" (2024, in Japanese) — https://wired.jp/article/glen-weyl-talk-2024/
Plurality, 6-0, Figure 6-0-B and the paragraph after it — https://www.plurality.net/v/chapters/6-0/eng/ ［CC0］
Plurality, 6-0, section "Surveyor's map" — https://www.plurality.net/v/chapters/6-0/eng/ ［CC0］
Plurality, 2-2, the paragraph on the Presidential Hackathon — https://www.plurality.net/v/chapters/2-2/eng/ ［CC0］
Plurality, 7-0, section "A road less traveled" — https://www.plurality.net/v/chapters/7-0/eng/ ［CC0］
Plurality, 6-0, section "The graph structure of social revolutions", paragraphs 1–3 — https://www.plurality.net/v/chapters/6-0/eng/ ［CC0］
Same (the paragraph on patterns of connection) — https://www.plurality.net/v/chapters/6-0/eng/ ［CC0］
Same (the paragraphs on ways of experimenting and rapid scaling) — https://www.plurality.net/v/chapters/6-0/eng/ ［CC0］
Same (the numbered list of conditions) — https://www.plurality.net/v/chapters/6-0/eng/ ［CC0］
Plurality, end of the section "Surveyor's map" in 6-0 (English, Japanese and Traditional Chinese editions) — https://www.plurality.net/v/chapters/6-0/eng/ ［CC0］
Plurality, 7-0 "Policy", opening — https://www.plurality.net/v/chapters/7-0/eng/ ［CC0］
Plurality, 7-1, sections "Transformative horizon" and "Mobilization" — https://www.plurality.net/v/chapters/7-1/eng/ ［CC0］
Technological Metamodernism, "Plurality" (2026-01-04) — https://technologicalmetamodernism.substack.com/p/plurality

We welcome error reports and questions ("What is…?"). Submissions become public issues.

Report an error in this article ・ Ask a question