🛠️The Philosopher’s Toolkit #1: Collective Action Traps

The first instalment of The Philosopher's Toolkit

This is the first instalment of The Philosopher’s Toolkit. For an introduction to the series, see here. It was inspired by two things: Deleuze’s definition of philosophy as “forming, inventing and fabricating concepts” and the old saying that “if all you have is a hammer, every problem begins to look like a nail”. The Philosopher’s Toolkit aims to give you more tools for interacting with the world.

I also want to provide a little context beforehand so you know a bit about where the concept originates, who it’s popular with and what it’s useful for.

---

• Category: Systemic

• Thinker of Origin: Mancur Olson

• School of Origin: Neoclassical Economics, Political Theory

• Where it’s popular: Game Theory, Systems Theory, Economics and LessWrong Rationalists¹

• Useful for:

  • seeing limits of individual agency

  • seeing how hidden Systems shape our day-to-day lives

---

The idea of The Philosopher’s Toolbox came to me after a coffee with my cousin one morning. We were talking about smartphones and schools, and I got to telling her about today’s conceptual tool.

A “Collective Action Trap” is a problem that can’t be solved by an individual. Sure, Gandhi’s “be the change you wish to see in the world” is nice and all, but reality is rarely so accommodating.

If there are twenty kids in a class and they are all on Snapchat and Instagram, you are not necessarily helping your kid by depriving them of a smartphone. You just might be handicapping them socially, as all the other kids continue unabated in their usage.

This is the Collective Action Trap². When the system incentivises the individuals within it from doing the right thing, you’ve got yourself a trap. Network effects punish any conscientious abstainers.

This doesn’t just apply to kids and their smartphones. Consider the social media companies themselves. Meta knows the impact social media is having on kids; they did a heap of research on it. But if they pull back from these demographics, then their revenue decreases, and after next quarter’s earnings call, the share price is going to take a dive. Meanwhile, a competitor or a disruptor shows no such scruples and happily races to the bottom, hoovering up all that extra attention. Fast forward a decade, and that generation is hooked into a different social media. Meta falls off a cliff. In a situation where the CEO had a lot of scruples around this and tried to change the course anyway, the board would boot him. They are all caught in a collective action trap.

Climate Change is another collective action trap. You might choose to cycle to work only to find your neighbour‘s new Hummer undermines any difference you may have made in the next decade.

Or consider AI safety. Companies might feel like AI is getting out of control, but if they slow down, then their less scrupulous competitors will only push further ahead. The same thing works internationally. America can’t slow down AI because what if China or Russia get to AGI or ASI (Artificial Superintelligence) first? They are caught in a collective action trap.

What to do about a collective action trap?

Escaping the trap requires collective action.

The problem of kids and social media can’t be solved by individual parents. Parents have to agree collectively to ban social media among the kids; schools have to ban smartphones.

On a bigger scale, publicly traded companies can’t self-police. This is where regulation comes in (sorry libertarians). You introduce laws and you enforce them. A collective decision takes the problem out of the hands of the individual companies.

At the level of AI safety and climate change, this doesn’t work well. The Kyoto and Paris Agreements were attempts to take collective action to circumvent the collective action trap of climate change. The trouble, and it’s the same with AI safety internationally, is that there is no higher authority that can levy meaningful punishment. Parents and schools can punish kids; regulators can punish companies but America can’t meaningfully punish China or Russia, nor vice versa. This is where tariffs, sanctions, or, as a last resort, war comes in. Lacking a higher authority, equals must wrestle with each other. Needless to say, the results at this scale have been…suboptimal.

1. Commonly known as Rationalists, but I like to reserve that term for Descartes and his peers who squabbled against the Empiricists back in the 17th and 18th centuries. I’ve mixed it with the website where these 21st-century Rationalists came out of LessWrong.com

2. or the Collective Action Problem, as its original name goes. It was originally formulated in 1965’s The Logic of Collective Action by Mancur Olson.