Based on an appearance on the Pascal’s Wager podcast, November 2024. This is not a transcript but a reconstruction of the main ideas from the conversation.

Video: [YouTube](https://www.youtube.com/watch?v=LINK

Rationality as a Bludgeon

Start with an uncomfortable observation: the word “rationality” is today most often deployed irrationally. In media discourse it has become a weapon — a way to stigmatise opponents rather than describe a genuine quality of thought. The people who shout loudest about rationality typically make no effort to justify their use of the term or to build even a minimal conceptual framework.

This is especially visible in political discourse. In the context of American elections, rationality is regularly invoked to declare supporters of one candidate stupid and supporters of another enlightened. That is not rationality. It is political combat wearing the costume of epistemology.

So before asking what it means to be rational, it is worth setting aside this vulgar, socio-political version of the word — and speaking about rationality in a strict sense.

Weak and Broad Theories of Rationality

A useful starting point is Jon Elster’s Sour Grapes (Gaïdar Institute edition in Russian). He distinguishes between weak and broad theories of rationality.

Weak rationality is simply coherence: the absence of internal contradictions among beliefs, desires, and actions. Three corners of a triangle: beliefs (how I think the world is), desires (how I want the world to be), and actions (what I do as a result). If there is no logical explosion among these three — the person is rational in the weak sense.

An uncomfortable implication: by this criterion, a serial killer who consistently acts on his belief that violence is permissible is formally rational. This is inconvenient but honest. The weak theory does not evaluate the content of beliefs — only their internal consistency.

The broad theory adds a requirement on the beliefs themselves: they must be grounded in available evidence, and desires must be autonomous. This immediately raises the central question:

— How much evidence does one need to gather before a belief can be considered rationally justified?

Elster answers honestly: gathering no information about one’s environment is irrational, but gathering it indefinitely is also irrational. There is some optimal volume. Where exactly that boundary lies is a matter of ongoing dispute. And it is precisely here that different models of rationality diverge.

Rationality and Religion: An Uncomfortable Neighbourhood

— The New Atheists — Dawkins, Harris, Hitchens — argued that the central failing of religion is its irrationality. Is that fair?

It is a well-worn thesis and it requires qualification. Medieval Catholic philosophy — Aquinas, Anselm — was deeply engaged in the rationalisation of Christianity. “I understand in order to believe” precedes faith within that tradition. Rational theology is a serious, centuries-long intellectual undertaking that cannot simply be waved away.

But there is a more fundamental problem with the New Atheist attack. Most religious claims are not falsifiable — and that is not a bug but a feature. How would you refute the Four Noble Truths of Buddhism? How would you demonstrate the falsity of the claim that one should follow the Noble Eightfold Path? These are normative statements, and the instruments of empirical verification do not apply to them.

Astrology is more vulnerable than religion in this respect: it makes specific predictions about reality — and those can be tested. They have been tested. It does not work. Part of religion’s survival advantage is precisely that it is smarter about avoiding this trap.

— What about the Orthodox heresiologist who explains why Mormons believe nonsense, while himself starting from no less problematic foundations?

This is a classic case. A person applies a sceptical standard to others’ beliefs while failing to notice the beam in their own eye. This is not moral hypocrisy but epistemic inconsistency: applying a criterion selectively, only when it is convenient. If the standard operates only on behalf of one’s own interests, it is not rationality.

Rationality and Goals: A Tool, Not an End

The most important thing to say about rationality is this: it is an applied instrument for achieving goals. Nothing more.

If a Mormon wants to reach the Mormon heaven, his actions — a two-year self-funded mission, abstaining from tea and coffee, marrying without premarital sex — are perfectly rational relative to that goal. From the perspective of someone with different preferences they look absurd. Rationality does not determine goals; it helps achieve them.

The attempt to turn rationality into a sacred cow — a universal method that can determine not just means but ends, that can generate the meaning of life and settle moral questions — is the disease of radical scientism. I do not share it.

Rationality does not heal emotional pain. Rationality does not explain why one should live. It is a narrow but highly effective tool within its proper domain.

Cognitive Honesty Matters More Than Rationality

— Do we actually need to be rational? And what is the best way to go about it?

I dislike the word “mindfulness” — it has been colonised by personal development culture. But the idea behind it is sound: reflection on one’s own beliefs.

I try to track where a given belief comes from. Sometimes I realise it is simply dislike — personal antipathy towards someone that colours my assessment of their ideas. I say so directly. I have a video that opens with the words: “I simply cannot stand this person, so roughly seventy percent of what you are about to hear is bias.” That is the minimum of cognitive honesty.

Where beliefs are hard-won and tested, I act with confidence. Where I recognise their weakness, I am more cautious about the actions that follow from them.

One further point: rationality works well for evaluating actions, less well for evaluating beliefs, and poorly for evaluating desires. With desires, Sapolsky arrives with his argument against free will, and the whole structure starts to wobble. But that is a different conversation.

Sour Grapes

Elster’s central metaphor: the fox cannot reach the grapes and tells herself they are sour. This is rationalisation — adapting desires to what is achievable, then persuading oneself that this was what one wanted all along.

We all do this. Recognising it in oneself is already something.

Further reading:

  • Elster, J. Sour Grapes. Cambridge University Press.
  • Pinker, S. Rationality. Allen Lane.
  • James, W. The Will to Believe.
  • Van Fraassen, B. The Scientific Image.