▶ CLASSICAL LOGIC·5–10 MIN

MODUS TOLLENS PRACTICE

The classical denying-the-consequent move. The fastest way to spot a broken conditional.

What this is

Modus tollens: if P then Q. Not-Q. Therefore not-P. The move is the engine of most theory-falsification.

The practice is small and useful: take a claim of the form 'if [theory], then [prediction].' Look at what's actually happening. If the prediction isn't happening, the theory takes a real hit — and you've just used modus tollens.

Steps

1.Pick a theoretical claim with a testable prediction. E.g.: 'If raising minimum wage to $15 causes unemployment, we'd see job losses in Seattle after the 2017 raise.'
2.State the conditional clearly: 'If P (theory), then Q (prediction).'
3.Check Q. What did actually happen?
4.If Q didn't happen: the theory is in trouble. (Not necessarily refuted — maybe the conditional was wrong, or other factors interfered — but it owes you an explanation.)
5.Try this with three of your own beliefs. What did you predict? What actually happened?

▸ AFTER

Which of your beliefs have you stopped checking against their predictions?

Reflections you write below are saved to your trajectory — Claude reads the prose and adds a small dimensional shift to your map, the same way it does for daily dilemmas and diary entries.

Create an account Sign in

More on this practice

Modus tollens — 'the mode that denies' — is one of the oldest validated argument forms in the Western tradition. The Stoic logician Chrysippus listed it among his five 'indemonstrables,' the basic inference patterns from which all others could be built, in the third century BCE. Its shape is simple: if P then Q; not Q; therefore not P. If the theory predicts rain and the ground is dry, the theory has a problem.

The form became the backbone of a whole philosophy of science when Karl Popper made it the engine of falsification. A scientific theory, Popper argued, earns its standing not by being confirmed but by surviving attempts to refute it: you derive a risky prediction, check it, and if the prediction fails, modus tollens forces a retreat. A single black swan refutes 'all swans are white' in a way that ten thousand white swans can never confirm it.

There's a famous catch worth holding onto — the Duhem-Quine thesis. When a prediction fails, modus tollens tells you something in your reasoning is false, but not which thing. Maybe the theory is wrong — or maybe an auxiliary assumption, the instrument, or the background conditions are. The disconfirmed theory takes a hit, but it can sometimes legitimately pass the blame to a neighbor. This is why a failed prediction owes you an explanation rather than an automatic surrender.

Common pitfalls

Confusing it with its invalid mirror, denying the antecedent: 'if P then Q; not P; therefore not Q' does not follow.
Treating a single failed prediction as a knockout. Modus tollens shows something is wrong, not that the headline theory specifically is — check the auxiliary assumptions first.
Only running it on other people's theories. Aim a risky prediction at one of your own beliefs and actually check it.

A worked example

You believe a colleague dislikes you. Stated as a conditional: if they dislike me (P), then they'll avoid working with me (Q). You watch. In fact they volunteered to pair with you twice this month (not-Q). Modus tollens does its quiet work: not-Q gives you not-P — the avoidance you predicted didn't happen, so the dislike you assumed takes a real hit. You pause before concluding outright that they like you (maybe your conditional was too crude), but the belief that felt like fact this morning is now a hypothesis that failed its first test.

Thinkers in this lineage

Chrysippus — The Stoic logician who catalogued modus tollens among his five basic indemonstrable inferences.
Karl Popper — Built falsificationism on modus tollens — theories advance by surviving attempted refutation.
Duhem & Quine — Showed that a failed prediction indicts the whole web of assumptions, not one theory in isolation.

Where to read further

The Logic of Scientific Discovery
Karl Popper · 1959
The case that science progresses by falsification — modus tollens as method.
Two Dogmas of Empiricism
W.V.O. Quine · 1951
The essay behind the holism that complicates any clean falsification.

Pairs well with

Kindred practices

Falsification test — Before holding a belief, ask: what observation would prove it wrong? If nothing could, modus tollens has nothing to grip.
Debugging by elimination — The programmer's version — a failing test denies the consequent and sends you hunting the false premise.

▶ What to do next