Hume on induction

Deduction provides us with certainty if the premises are true:

But how to determine the truth of premises?

With non-Euclidean geometry doubt is even cast on the truths of abstract reason. If not even geometry can boast indisputably true premises, it seems unlikely that other forms of knowledge, and the natural sciences in particular, can.

Can we study nature by only using deductive arguments?

David Hume - Relation of Ideas, and Matters of Fact.

The objects of some sciences are ideas and their relations, which we establish by thinking, without observing, or carrying out experiments.

Deduction seems not to be applicable to the study of nature.

Non-deductive arguments in everyday life

Catching the train, the sun coming up, walking down stairs.

Forms of inductive argument

Inferences from observed past events to future events (past-to-the-future inferences)

Inferences from observed events to past events (present-to-the-past inferences)

Inferences from past events to remote past events (past-to-remote-past inferences)

Inferences from observed events to general conclusions

Inferences from observable phenomena to unobservable phenomena

David Hume and the problem of induction

On the one hand inductive inferences are both unavoidable and widely trusted, both in science and in everyday life. We make them whenever we need to acquire knowledge of the world around us. But on the other hand inductive inferences do not seem to be justified in the way that deductive inferences are justified.

What is the foundation of all conclusions from experience?

‘… our conclusions from that experience [i.e. experience of cause and effect] are not founded on reasoning, or any process of the understanding’

Hume argues that human beings can only have knowledge of the ‘sensible qualities’ of things: that is, those qualities that can be apprehended by our senses, such as colour and weight. However, we cannot observe the hidden properties of objects.

Bread/nourishment example: this inference is based on an unjustified assumption: that because we have seen some hidden properties (i.e. being nourishing) being conjoined with observable properties (i.e. the colour and consistency of bread), we expect that objects carrying the same observable properties will also have the same hidden properties.

Second, the fact that bread has nourished me in the past (premise) does not entail that it will nourish me in the future (conclusion).

• our reasoning regarding what we have experience of (‘matters of fact’) is based on inductive inferences
• inductive inferences are not rationally justified. In other words, there is no guarantee that the truth of the premises (e.g. bread has nourished me in the past) is transmitted to the conclusion (e.g. bread will nourish me in the future). The conclusion is not logically entailed by the premises.

“The contrary of every matter of fact is still possible; because it can never imply a contradiction, and is conceived by the mind with the same facility and distinctness, as if ever so conformable to reality. That the sun will not rise to-morrow is no less intelligible a proposition, and implies no more contradiction, than the affirmation, that it will rise”

If you say that a triangle has four sides, this is a contradiction. There is no contradiction in saying that trees will flourish in December and January.

Can induction be rationally justified?

There is a problem in justifying our inductive arguments by saying that nature is uniform. In Hume’s words, if we do this, we go ‘in a circle, and [take] for granted which is the very point in question’

Peter Lipton says that we are all ‘conservative inductivists’ – ‘more of the same’.  Lipton imagines a situation in which one of his students, after a tutorial, rather than leaving through the door, is about to throw herself from the window. She is a member of a group called ‘revolutionary inductivists’: their motto is ‘it is time for a change’.

Logically, the argument that nature has been working in a certain way, therefore it will keep working in the same way, is no different from her argument that nature has been working in a certain way, therefore nature will work differently in the future

Hume believes it is not reasoning that teaches us to expect similar properties in similar objects: the infant who has painfully experienced fire, will avoid touching it next time it sees it. However, its behaviour is not the result of any rational argument, first of all because it is not capable of it. Yet, it is able to learn from experience - It is their ‘habit’ or ‘custom’, and it is part of their nature.