Fallacies: I

Core idea

Letter I is unusual: alongside its everyday corrections it contains a genuine theory of error. Francis Bacon's four "idols" — the recurring traps of the human mind — turn this topic from a list of myths into a diagnosis of why myths happen. The other entries become case studies. Iridology and counting sheep illustrate Bacon's idols; the words "idiot" and "Indian ink" show how language misleads; the non-existent Aurora Islands show a false fact copied uncritically for over a century.

Why it matters

Most of this book debunks beliefs one at a time. Bacon's idols give you a general tool: instead of memorising corrections, you learn the four shapes error tends to take, and then recognise them anywhere. That is the difference between being handed fish and being handed a net.

Why the myths persist

Idols of the tribe — built into everyone

Bacon's first trap is shared by all humans: we assume more order in nature than exists, generalise from too few cases, and "support an assumption by quoting affirmative instances and omitting all negative instances." That last is exactly what we now call confirmation bias.

Idols of the marketplace — words pretend things exist

Some words "are names for non-existent things which are supposed to exist because they possess a name." Iridology has a name, a literature and a founder; none of that makes the iris-to-organ map real. The Aurora Islands had a name on every chart — and never existed.

Idols of the cave — the bias of the individual

Each person's "peculiar mental or bodily constitution" tilts their view. Bacon's rule: whatever your mind "seizes and dwells upon with particular satisfaction is to be held in suspicion."

Key takeaways

Mental model

Copy sourceZoom

Practical application

Bacon's idols work as a checklist. When testing a belief, ask which idol it might be:

1. Tribe — am I only counting the hits? Note the cases that failed. A gender-prediction trick or a folk remedy that "always works" usually relies on forgotten misses.
2. Cave — does this idea please me too much? If you are delighted by a conclusion, raise your guard. Satisfaction is not evidence.
3. Marketplace — does the word do the work? "Iridology", "sanpaku", "Aurora Islands" all sound solid. Ask whether the thing exists apart from its name.
4. Theatre — am I inheriting a system? Some errors come pre-packaged in a respected framework. Ask whether you tested the framework or just adopted it.

Example

A wellness coach offers an "iris reading" and points to a fleck in your eye as a sign of "liver stress." Run Bacon's checklist. Marketplace: "iridology" is a real word with books behind it — but the nerve filaments linking iris zones to organs were never found by anatomists, so the practice names a connection that does not exist. Tribe: the coach remembers clients whose readings "came true" and forgets the misses. Cave: a tidy diagnosis from a glance is satisfying — which, by Bacon's rule, is exactly when to be suspicious. Three idols, one reading. You thank the coach and book an actual blood test.

Related lessons