Empiricism and Hume: the claim that all knowledge derives from sense experience, the distinction between impressions and ideas, the copy principle, the fork between relations of ideas and matters of fact, and Hume's sceptical conclusions about causation and the self.

What this dot point is asking

Hume is the empiricist set thinker for Knowledge and Doubt, the counterweight to Descartes. You must explain the empiricist claim that all knowledge comes from sense experience, Hume's distinction between impressions and ideas and the copy principle, his fork between relations of ideas and matters of fact, and the sceptical conclusions he reaches about causation and the self. Hume is the alternative answer to the problem of knowledge: not reason, but experience, and with strict limits.

Empiricism and impressions and ideas

The difference between an impression and an idea is one of force and vivacity: actually feeling pain is an impression, while remembering or imagining pain is the corresponding, fainter idea. This division is the foundation of Hume's empiricism, because it lets him ask, of any idea, where its source impression came from.

The copy principle

The copy principle does serious work. A person born blind has no impression of colour, so can form no idea of it. Complex ideas (a golden mountain) are assembled from simple ideas (gold, mountain) that each copy an impression. When a philosopher uses a term that answers to no impression (Hume suspects "substance" and "necessary connection" are like this), Hume challenges them to produce the impression it copies, on pain of the term being meaningless. This is empiricism turned into a weapon.

Hume's fork

The fork is a sorting test for any claim to knowledge. "All bachelors are unmarried" is a relation of ideas, true by reason and definition. "Water boils at 100 degrees Celsius at sea level" is a matter of fact, learned from experience and conceivably otherwise. A claim that fits neither branch, and cannot be traced to experience, Hume treats as empty: this is the basis of his scepticism about metaphysics and about ideas that overreach the evidence.

Hume's sceptical conclusions

Applying the copy principle and the fork, Hume reaches two famous sceptical conclusions:

• Causation. We never observe a necessary connection between cause and effect, only one event constantly followed by another (constant conjunction). Our idea of causal necessity comes from the mind's habit of expecting the effect, not from reason or observation. So we cannot rationally prove that the future will resemble the past (the problem of induction).
• The self. When Hume looks inward he finds only particular perceptions (a feeling, a thought), never an impression of a continuing self that has them. So the self is, for Hume, a bundle of perceptions, with no impression to ground the idea of an enduring "I".

These show empiricism's double edge: it explains knowledge through experience, but it also limits knowledge severely, undercutting claims, including some Descartes relied on, that go beyond experience.

Examples in context

Take the everyday belief that the cue ball makes the red ball move. Hume asks for the impression of the "making", the necessary connection. We see the cue ball move, touch the red ball, and the red ball move; we see one event followed by another, but no impression of a force compelling it. All we ever observe is constant conjunction. Our sense of necessity, Hume argues, is just the mind's habit of expecting the usual effect. He applies the copy principle (where is the impression of necessary connection?), finds none, and concludes our idea of causation is grounded in custom, not reason.

Try this

Q1. State the copy principle and give an example. [2 marks]

• Cue. Every simple idea copies a corresponding simple impression; for example, a person born blind has no impression of colour and so can form no idea of it.

Q2. What does Hume say is the real source of our idea of causal necessity? [2 marks]

• Cue. The mind's habit or custom of expecting the usual effect after constant conjunction, not reason or any observed necessary connection.