11. Fallacy of Composition
Wireless Philosophy, Fallacies, 3:59. Original title: CRITICAL THINKING - Fallacies: Fallacy of Composition
Summary
Paul Henne (Duke University) explains the fallacy of composition, an informal fallacy that leads people to conclude, without justification, that a whole must share the qualities of its parts.
Definition and general form
Henne recalls that a fallacy is a defect in reasoning, and that fallacies come in two kinds: formal fallacies, which contain an error in an argument's form, and informal fallacies, which contain an error in its content, a distinction covered in his video on formal and informal fallacies. The fallacy of composition is an informal fallacy: people commit it when they infer, without justification, that what is true of the parts of something must also be true of the whole those parts compose.
Premise 1: The parts of whole A have qualities X, Y and Z.
Conclusion: Therefore, whole A must have qualities X, Y and Z.
He compares the move to arguing that because Arizona has an arid climate, the whole nation must have an arid climate too.
Two examples
Three and seven are each odd numbers, and together they compose the number ten, but it does not follow, and it is not true, that ten is odd. Henne then imagines a friend arguing:
Premise 1: Atoms are colourless.
Premise 2: Cats are composed of atoms.
Conclusion: Therefore, cats are colourless.
Both premises are true, yet the conclusion is false: the parts' colourlessness does not transfer to the whole cat.
When the inference still works
The fallacy does not guarantee a false conclusion. If every part of a cat is composed of matter, then the cat itself is composed of matter, and that conclusion happens to be true. Henne stresses that the fallacy lies in the missing justification for the inference, not in the falsity of the conclusion, which can go either way.
Key takeaway
The fallacy of composition is inferring, without justification, that a whole has the same qualities as its parts; the inference can occasionally land on a true conclusion, but it is fallacious reasoning regardless, and should not be trusted to reveal, for instance, colourless cats.