QuestionsCan we be certain of anything?The chicken and the principle of induction
Can we be certain of anything?Yes, through reason

The chicken and the principle of induction

Bertrand Russell 1872 · 1970Clarify thought with logic, doubt dogmas.
See profile
What it claims

The principle of induction is known by its own evidence, without proof from experience: it is a priori knowledge, and probability, never certainty, is all we can seek about the future.

Why

Experience cannot show that the future will resemble the past without already assuming it: this principle must be accepted as self-evident.

In short

"The chicken and the principle of induction" is an argument by Bertrand Russell (The Problems of Philosophy, chapters 6 and 7) on the question "Can we be certain of anything?" (answer: yes, through reason). Its claim: The principle of induction is known by its own evidence, without proof from experience: it is a priori knowledge, and probability, never certainty, is all we can seek about the future. According to AskPhilosophy's logic check (Z3 solver), the argument is valid only if one grants a premise the author leaves unstated (P4). AskPhilosophy presents 3 objections, 2 of them with a documented reply, and the original text behind each premise, checked word for word.

The Problems of Philosophy, chapters 6 and 7
Valid with an implicit premise

A chicken is fed every day by the same man, until the day that man wrings its neck. Are we better placed than the chicken when we believe the Sun will rise tomorrow? Russell takes up Hume's problem, but does not reach the same conclusion. Hume concluded: we believe out of habit, not reason. Russell shows that everything rests on a single principle, which experience can neither prove nor disprove. So we must choose: accept it because it is evident in itself, or give up any reason to believe in the future. He takes the first path.

P1All our expectations about what we have not observed (tomorrow's Sun, bread that nourishes) are justified only if the principle of induction is: the more often two things have been found together, the more probable it is that they will be again.
P2Experience cannot prove this principle: every argument from experience already assumes it, so it would be circular. (Nor can experience disprove it: the deceived chicken or the black swan show that a probable prediction can fail, not that it was not probable.)
P3So there are only two possibilities: accept the principle because it is evident in itself, or give up all justification of our expectations about the future.
P4 · implicit premiseOur most ordinary expectations are indeed justified, at least as probable: it is not unreasonable to think bread will nourish better than a stone. Russell does not prove this: he notes that everyone believes it without hesitation.
Therefore the principle of induction is known by its own evidence, without proof from experience: it is a priori knowledge, and probability, never certainty, is all we can seek about the future.

click a premise to reject it

If you accept every premise, logic forces you to accept the conclusion. So the debate is about the premises.
Checked by a logic solver: the premises cannot all be true while the conclusion is false.
Without the implicit premise: Without P4, the computer finds a world where the first three premises are true, but the principle is not evident and none of our expectations is justified: the sceptic's world, where we are no better placed than the chicken.
Valid with P4. Without it, only the dilemma of P3 remains, and one can take the other branch: give up all justification. That is the sceptic's choice, and roughly Hume's (we believe by custom, not reason). The heart of the difference between Hume and Russell is therefore P4: Russell refuses to put himself in the chicken's position and grants that some knowledge does not come from experience. He thus sides with the rationalists on one point, while remaining an empiricist about everything that exists.
Analogy. You cannot check a rule with the rule itself, just as you cannot test a scale's accuracy by weighing the scale on itself. You must either trust the scale or give up weighing anything. Russell chooses trust, but a reasoned trust: the scale gives probabilities, never certainties.

Objections

3

Each objection is a conversation: click to see who replies and where the debate stands.

Objection anticipated by Russell attacks premise 2

But we do have experience of the future! What was the future has constantly become the past, and has always been found to resemble the past.

Russell replies
Russell
We have experience of "past futures", not of "future futures". And the question is precisely: will future futures resemble past futures? An argument starting only from past futures cannot answer it.
Hume 1748 · attacks premise 4

Our expectations are not justified by reason, and that does not matter: they come from custom, not from reasoning. We cannot help believing, and that is enough to live by. Speaking of "intrinsic evidence" is giving a learned name to a habit.

Russell replies
Russell
Habit explains why we believe, not whether we are right to believe. Animals too expect the same thing to happen again, and the chicken dies of it. We must distinguish the fact that the past causes our expectations from the question whether there is a reasonable ground for giving them weight.
The debate goes on (our summary) Hume would reply that Russell merely names this reasonable ground without showing it: "evident in itself" does not explain why the principle would be true rather than merely irresistible.
Karl Popper 1934 · attacks premise 1

Science does not need the principle of induction. It never justifies its theories by favourable cases: it puts forward bold hypotheses and tries to refute them. A theory is never proved, only "not yet refuted". Russell's problem disappears if we give up justifying our beliefs.

no reply
summary: their text is not yet in our corpus
We have not yet found a direct reply by Russell in the texts.
The debate goes on (our summary) Many of Popper's critics answer that, to prefer tomorrow the theory that survived yesterday, one must assume the future will resemble the past: induction comes back through the back door.