False Inference

An argument that draws false conclusions from observations.

The observation may be true, but the inference drawn from it is false. In some forms of the fallacy, the inference itself is never explicitly stated, but it turns out to be a necessary step to reach the conclusion. Example:

[This year’s] examination results published in your paper reveal that in A levels, girls outperform boys in 25 of 31 subjects at Grade A; in AS levels, girls outperform boys in 26 of 31 subjects at Grade A; and in GCSE girls outperform boys in 23 out of 27 subjects at Grade A*, and in 24 out of 27 subjects at Grade A (home economics and tied results excluded). Despite this incontrovertible proof of the intellectual superiority of girls at the highest levels . . .^F10

There are five inferences here:

1. That intellectual superiority is a quality which school exams measure accurately and

2. . . . at the highest levels.

3. That all other factors affecting performance in exams are equal or irrelevant: these factors include the degree of application which boys and girls bring to school work, their respective aptitude for conformist behaviour, and the age at which they reach intellectual maturity.

4. That the results for one year can be taken as an indicator of how girls perform relative to boys in exams in all years.

5. A fifth inference—though this is one that the reader is encouraged to draw for himself—is that intellectual ability as measured by exam results comprises intellectual qualities of all kinds, including (say) creativity, persistence, personality, intellectual daring and a talent for conciliation.

It may, at a pinch, be reasonable to argue either for or against any of these; what is not reasonable is to try to make an argument hang on inferences which are taken to be self-evident, but which can in fact be shown with some confidence to be fantasy. So, there are two problems here. The first is that such mistreatment of logic threatens ultimately to mature to the point of destroying argument itself as a means of thinking and persuasion: it can simply insist that a certain inference is true, and all sorts of execration and punishment can then be used to reduce the opposition to silence.

The second problem is that, since false inference is not rooted in any logic, it is unstable, and can flip, depending on current attitudes. The example above is a spirited advocacy of the abilities of women—clearly a good cause. However, the same form of false inference could be used to argue any other view—an opposite and outrageous one, for instance. Disconnected opinions can quickly recalibrate to their opposite. And it is not just a matter of the person sneakily breaking the rules of logic and hoping that no one will notice, because the false inference is used most of all to fill the gaps and inconsistencies in the speaker’s own opinion. The person who makes the false inference is the first to be persuaded by it.

Should their opinion shift, therefore, as the opinions of campaigners sometimes do, false inferences can be applied to their new task—from supporting the worthy to supporting the indefensible or the absurd: the economy is growing, so we must all be getting happier; the Government doesn’t believe in peak oil, so it can’t be a problem; the midges are biting me, so I must be delicious. The logical rootlessness that comes from the false inference is an incentive to grasp at anything that moves, and to drift downstream with it into the forgiving, logic-free, lethal comforts of the big idea.^F11

Related entries:

False Premise, Ideology, Calibration, Abstraction.