At the end of Section 2 (Book 1, Part III) of the Treatise of Human Nature, David Hume set down two questions that he intends to answer. Firstly, why do we think that it is necessary for every object to have a cause? Section 3 will attempt to answer this question. Section 6 then leads on to the second of Hume’s questions: from which part of the mind do we draw such an inference?
Section 3: Why a Cause is Always Necessary
Section 3 considers the first question proposed by Hume. Is it true that everything must necessarily have a cause? Although common experience (and philosophy in general) would suggest that the answer is ‘yes’, Hume believes that this cannot be categorically proven. There is no demonstrative proof that can be used to prove that a cause is necessary.
Furthermore, because every idea is distinct from every other idea (the ideas of cause and effect are also distinct), we can easily conceive of an individual idea being existent at one moment in time but non-existent at another moment. The separation is plainly possible for the imagination, which implies there is no absurdity in us thinking of a simple idea separated forever from every other idea.
To prove that an idea is linked to another, we would have to show that a relation exists based on one of Hume’s four key criteria: resemblance, proportions in quantity and number, degrees of qualities, and contrariety. Since ‘resemblance’ and ‘contrariety’ are at opposite ends of the spectrum, this would seem to cover all possible means of defining a relationship. However, none of these criteria can be proven demonstratively.
Despite Hume’s effort to demonstrate that necessary connection cannot be proven, it’s difficult to think what Hume has in mind when he says that it’s possible that some object in existence doesn’t have a cause. This goes against all our common sensations and rational thought processes, so it is hard to picture what that object would be. Hume is not suggesting that we should visualise such an object, rather that the proposition of necessary connection can’t strictly be proven.
Hume’s basic argument in Section 3 can be summarised as follows: because we can visualise any item existing on its own and distinct from other items, this is enough to show that the item does not necessarily have to have a cause, a relationship with any other object. This argument does not seem persuasive. It seems to be based on the premise that because something is possible in theory, it is also possible in fact. You could apply the same argument to any number of ideas, for example, because God is possible in theory, it would be wrong to adopt an atheist point-of-view.
Section 6: Of the Inference from the Impression to the Idea
In Section 6, Hume re-considers his notion of ‘constant conjunction’. By looking at objects as distinct and separable ideas, we can find no property in the object itself that necessarily leads us to another object, such as a property that suggests a relationship of cause and effect. No object on its own implies the existence of any other object.
Therefore, we must look elsewhere to discover where that relationship is established. If reason cannot provide the answer, then we must turn to experience. We remember the fact that we have often experienced event A along with even B, so we draw the conclusion that the two are necessarily connected. Although we only ever experience one original impression – the very first encounter with the conjunction, as all subsequent occurrences are merely copies of existing ideas – the repeated experience does have an impact, helping to reinforce ideas of cause and effect. Eventually, if the constant conjunction is strong enough and the other two requirements of causality are present – contiguity and priority in time – we form the idea of necessary connection.