Hume and the Fight Between Scepticism and Human Nature (26-Feb-2004)

A dictionary definition of scepticism is "an attitude of doubt or a disposition to incredulity either in general or toward a particular object". Hume describes two types of philosophical scepticism:

  1. antecedent scepticism regards all propositions as doubtful; doubt is only suspended from those propositions for which proof can be supplied
  2. consequent scepticism regards propositions as doubtful only if evidence has shown that there is reason to question them
Hume regards antecedent scepticism as a philosophical dead-end, since if everything which cannot be proved is taken away, he believes that it would be impossible to rebuild a structure of knowledge "but by the use of those very faculties of which one supposes to be already diffident". However, he does acknowledge the value of approaching investigations in a sceptical frame of mind.

The paradox that Hume highlights in this passage is that despite there being well-found reasons for questioning, and therefore being sceptical about, certain propositions, it is an essential aspect of human nature that these doubts are ignored.

Hume gives two examples of the way that we are happy to disregard reason in this way. Firstly, we regard ourselves as interacting directly with the external world, despite the fact that a moment's reflection reveals our sensory experience to be at best a distorted representation of it. Secondly, and more importantly, we rely on our past experience as a guide to the future: "all our evidence for any matter of fact, which lies beyond the testimony of sense or memory, is derived entirely from the relation of cause and effect".

Hume says that it is not logically defensible to make general predictions of cause and effect based on our past experience of seemingly connected events. Take the fact that I have observed that water boils when I raise its temperature to 100 degrees. My natural inclination is to assume a causal link, and to predict that the next time I heat water, it will boil at the same temperature.

But because I make this prediction based on the evidence of past observations, I should have some reason to show that past observations are a reliable source of information. This is where the problem arises: what makes me think that past observations are a reliable guide? Only the fact that I have observed in the past that they generally are. This is rather like trying to establish the reliability of a court witness by asking "are you a reliable witness?"

There is no logical reason, says Hume, for supposing any uniformity of nature; the paradox is that we cannot help but do so, even as we acknowledge the irrationality of our position.


Mind and World page