Philosophy of Probability
You Don't Know What You Think You Know About Chance
Probability doesn't describe the world — it describes your ignorance of it.
The Idea
Most people treat probability as a feature of reality, something out there in the coin or the dice. But this assumption quietly breaks down under scrutiny, and the crack it leaves has been widening in philosophy for over a century. There are, broadly, two rival camps. The frequentist says probability is real and objective — it's the long-run frequency of an outcome across many repetitions. Flip a fair coin a million times and heads will hover around fifty percent; that regularity is what 'fifty percent' actually means. The Bayesian disagrees, fundamentally. For a Bayesian, probability is a degree of belief — a measure of how confident a rational agent should be given the evidence available. Probability lives in minds, not in things. This isn't just a technical squabble. It changes everything about how you interpret a weather forecast, a medical test, a courtroom verdict. When a doctor says there's a seventy percent chance your treatment will work, are they reporting a fact about your body, or summarising a state of knowledge? The Bayesian says the latter — and insists that as new evidence arrives, you are rationally obligated to update that belief in a precise, calculable way. What makes this philosophically vertiginous is that both interpretations are internally coherent. The disagreement isn't about the maths; it's about what the maths is pointing at. Probability might be the most practically useful concept humans have ever developed, and we still can't agree on what it fundamentally is.
In the World
In 1978, a statistician named Dennis Lindley wrote a short, provocative paper crystallising what he called 'the Bayesian position' — that all uncertainty should be expressed as probability, and that rational belief-updating is the only coherent foundation for science. Around the same time, the broader scientific community was overwhelmingly frequentist, allergic to the idea that subjective belief had any place in rigorous methodology. The tension wasn't academic; it played out in clinical trials, economic forecasting, even criminal investigations. One famous collision happened in the UK legal system during the 1990s and 2000s, where prosecutors repeatedly misapplied probabilistic reasoning in ways that sent innocent people to prison. The error — later dubbed 'the prosecutor's fallacy' — arose from confusing two very different questions: the probability of seeing this evidence if the defendant is innocent, versus the probability of the defendant being innocent given this evidence. These are not the same calculation, and conflating them is not a quirky mistake. It follows directly from failing to think in a Bayesian way. Sally Clark, wrongly convicted of murdering her two infant sons partly on the basis of deeply flawed statistical testimony, spent more than three years in prison before her conviction was overturned. The stakes of getting the philosophy of probability right turned out to be a mother's freedom and, ultimately, her life — she died a few years after release. Probability isn't just an abstract debate. It's the scaffolding under our judgements.
Why It Matters
There's a quieter, more personal dimension to all this. Most of us walk around with confident beliefs about risk — our health, our relationships, our plans — that we've never examined for coherence. We treat our gut sense of 'likely' or 'unlikely' as though it were reading off some objective dial in reality. The Bayesian insight invites something closer to epistemic humility: your probability estimate is your current best belief, not a fact, and it should shift when evidence shifts. That's harder to live by than it sounds. We tend to update downwards reluctantly (it's difficult to let go of optimistic forecasts) and upwards in panicked lurches (one bad news story restructures everything). Recognising that probability is a discipline of mind — not just a number — opens space for asking: what evidence am I actually weighting here, and am I doing so honestly? That's not a mathematical question. It's a practice of attention. On a Monday morning, as you form views about how your week might go, noticing the difference between 'this feels likely' and 'I have good reason to think this is likely' might be the most useful philosophical move you make all week.
A Question to Ponder
Is there a belief you're holding with high confidence right now that you've never actually tested against evidence — and what would it take to genuinely update it?
Get a new one of these every morning.
Start learning with Thinkable