Every good ideology has a Pascal's Mugging at its core
Belief in an unlikely but transformative future vision is the most powerful motivating force for groups of people to come together and work extremely hard. This belief both unifies and motivates its followers. In this essay, I argue that every effective ideology has such a belief at its core.
Every ideology promises its followers either a wonderful outcome or an evasion of a tragic one. Frequently, such promises provide tantalising rewards and eternity to enjoy them. Reward’s recipients would be proven right in the grand history of time.
Even rational people, such as Pascal, can be attracted to this promise. They accept it by calculating that the wonderful outcomes even when multiplied by a tiny probability are still worthwhile to pursue. Accepting such a belief is called Pascal’s Mugging.
Pascal’s Mugging is also an effective instrument for building ideological communities. The outside world is likely to misunderstand or even ridicule people who believe in these unlikely visions. Otherwise the outside world would be pursuing such visions. The outer world is cold and misunderstanding. In contrast, the insiders provide emotional support. They understand your deepest and most meaningful belief.
What’s Pascal’s Mugging and why do we find it so absurd?
Pascal defended the belief in god on probabilistic basis. If one believes in god and she doesn’t exist, nothing’s gained and nothing’s lost. And if god does exist, then the eternal life in heaven is nearly infinitely great. Consequently, Pascal claimed that as long as that there’s a non-zero probability of god’s existence, the rationalist should believe in her.
Pascal’s Mugging is the inversion of this argument by Nick Bostrom. Nick argues that Pascal’s logic exposes him to a simple mugging. All that the mugger has to do is claim that they can provide Pascal with some extremely large reward in the future. As long as we admit that there is some, no matter how small, chance the mugger is telling the truth, it makes sense to hand the money over to him now. That is, because the small probability multiplied by the huge reward are assumed to be larger than whatever amount Pascal might be carrying in his wallet.
Mugger: Hey, may I have your wallet?
Pascal: No, I’d prefer to keep it myself.
Mugger: What if I told you that I’d bring you a bottle with a genie tomorrow?
Pascal: Then I’d tell you that you’re a damn liar!
Mugger: How confident are you that I can’t do that?
Pascal: As confident as one can be. I’d say nearly 100%.
Mugger: Nearly?
Pascal: Well, yes. One can never be certain of anything.
Mugger: Would you say that there’s at least one in a trillion chance that I could be telling the truth?
Pascal: Hmmm, maybe just about.
Mugger: And you could surely use the Genie to make turn Earth into a space-roaming civilization. That would be worth at least a quadrillion dollars!
Pascal: The expected value of this bet is a formidable thousand dollars and I’m only carrying a measly fifty in my wallet. You are doing me a favour by accepting this wallet from me.
Pascal hands over his wallet to the Mugger. He’s still waiting for both the Mugger and the Genie to this day.
The mugging feels paradoxical. At the core of the apparent paradox is a simple mathematical error. The total probability of any wonderful event is fixed. Therefore, given a promise of a single, wonderfully impactful event, it is extremely unlikely for it to dominate the probability of all such events in the past. The probability sum of the category of events is useful as our true internal probability assessment can be witnessed by our actions. If we say something is both likely and impactful our actions should reflect that. If they don’t, we are either lying our deluding ourselves.
Assuming you weren’t chasing people with Genie lamps in the past, you should not accept someone’s promise of being a Genie master. Victims of Pascal’s Mugging tend to overly focus on a single catastrophic event, at the expense of infinite other events in the same category. This creates a Zeno’s paradox of probability assessment.
In the general case mugging victims don’t just shift the focus of the calculus towards the unlikely event, their whole lives become focused on it.
Ideologies as Pascal’s Mugging Incidents
Any effective ideology has Pascal’s Mugging at its core.
I define ideology as a group of related ideas that tend to be believed together rather than individually. There’s no moral judgement attached to the word in this essay. Agile software engineering has been responsible for improved productivity and developer happiness in many teams. Scientology has been responsible for harassment of its opponents and alienation of its members from their previous social ties. Both Agile software engineering and Scientology are ideologies by this essay’s definition.
For an ideology to be effective, its members need to be: persuasive and productive. Persuasiveness is measured by ability to convince new members to join the ideology. Ideology is productive if its members consistently work towards the goals of such ideology.
Effective Altruism (EA) is an example of an effective ideology. EA has motivated many young ambitious people to change their career paths in ways that enable them to donate more to EA-supported charities.
Effective ideologies use a form of Pascal’s Mugging by convincing their members that by joining them, they will experience an extremely large reward as a result.
At the core of every effective ideology is a belief in an unlikely but powerful future vision. Making this vision a reality is the core driving factor of the ideology.
Follower’s of the ideology experience a similar cognitive illusion as mathematicians who try to reason about expected value of unlikely events. By focusing their attention on a particular version of the future, other outcomes seem less important. This makes the powerful future vision dominate the imagination at the expense of probabilistic calculus.
Belief in an unlikely future vision also serves a social purpose. Ideology’s community by its definition has accepted the future vision. People outside of this community will think of you as crazy if you speak about your beliefs openly. Therefore fellow people from the in-group become the only people with whom you can discuss the most interesting vision for the future without being ridiculed.
Rewards don’t have to be experienced directly, even in the best case scenario. Yes, many ideologies promise infinite rewards in the afterlife. But one can follow an ideology where they believe that they’ll land on the right side of history long after their death. In this case, even if the promise materializes, they won’t be there to witness its rewards directly. Nevertheless, the future vision is worth sufficiently for a lot of people.
Case studies:
Salafi Islam
Salafism is a branch within Sunni Islam that emphasizes the emulation of the earliest Muslims. Known as the Salaf, these are the companions of Prophet Muhammad and two subsequent generations. Salafism rejects how Islam has evolved past Salaf generations and advocates a return to the this period.
Al-Qaeda, ISIS, Talban, and nearly every other organization of Islamic extremism would consider itself Salafi. Salafism does not justify violence. However, two aspects of Salafism lend themselves well to justification of extremism within those who are searching for such.
First, is the conservative interpretation of Islam within Salafism. This aspect is most frequently used to explain Salafism’s appeal to extremist organizations. Salafism is known for literalist interpretations of Islam’s holy texts and rejection of later innovation within Islam. This can be used to justify extremism by, for example, redefining the concept of Takfir (declaring another Muslim as a non-Muslim). Takfir is a serious offence within most interpretations of Islam, including Salafism. However, Al-Qaeda and other extremist organizations are reinterpreting this concept and using it to justify violence towards other Muslims.
Second, being right in the grand history of time justifies departure from current societal norms. This is the main lure, used to conduct Pascal’s Mugging here. Being shunned by contemporaries is a tiny price to pay for being right in the long run. Every visionary of the past, from Jesus and Copernicus to J.C.R. Licklider and James Clark were ostracised or at least seen as weird by their contemporaries. Any society is likely to criticise ideas just because they are different. Most people tend to be wrong about most new ideas most of the time. People working on radically different ideas must apply a strong filter to the cynicism of others if they want to succeed. Unfortunately, it is difficult to distinguish between knee-jerk cynicism of newness and constructive criticism. So it’s easy to dismiss all criticism. The absence of a good filter leaves the person working on the new thing without an external sanity check.
TODO: what about non-extremist Salafi followers, do they generally show stronger devotion and depth of study than other schools? e.g. is complete memorisation of Quran more common within Salafi’s than in other schools of Islam.
Effective Altruism
Effective Altruism (EA) is an agnostic ideology. The core idea is doing good and being able to measure it quantitatively. Within EA probabilistic assessment of situations is also very common. EA followers focus on variety of projects that aim to deliver the highest amount of value. Different projects focus on a different timelines in the spectrum from immediate (< 1 year) to existentially long (> 1K years).
EA exemplifies Pascal’s Mugging in two ways:
- Doing good quantitatively as a reward in its own right,
- Longtermism: being able to overcome catastrophic risks to human survival (e.g. AI safety).
It is fascinating to see how appealing EA is to ambitious young people. In Going Infinite Michel Lewis describes the confusion in the eyes of 2 Sigma managers as they were losing top talent to FTX. It was confusing because the salary was lower and normally the benefits of a top paying hedge fund are sufficient to retain nearly everyone.
Being able to outcompete top companies for talent and drive capital towards objectively good charities definitely justifies the effective in Effective Altruism.
Veganism
I’m neither a Salafi Muslim nor an Effective Altruist, albeit, I do support the latter ideology. The analysis of the previous two groups was an observation from the outside rather than the inside. It is easy to criticise beliefs of others, but my theory would be invalid if I’d consider myself exempt from it. Now I turn to analysis of two groups that I do consider myself part of: vegans and startup founders. Both of these groups exemplify minorities who were able to leverage asymmetric societal and economic impact to their size. I believe that Pascal’s Mugging plays a central role in why followers of these groups have managed to collaborate so well and work as hard as they have.
- Veganism as Pascal’s Mugging. What does it mean to be right in the grand scheme of time for a vegan?
- What impact have vegans had?
- What has been the role of cooperation between vegans in leveraging this impact?