Something is necessary when it is fundamentally true in all contexts, this is what I've understood from what I've read regarding Hume's problem of induction. However, uniformities are not true in their literal sense but rather an assumption, would you say that uniformities are hence unnecessary? But sure they are necessary to allow the formulation of principles since without this 'unnecessary' assumption one would begin a cycle of infinite regress, since we need some sort of a stand point. Is my line of thought correct?

  • 1
    Be cautious about using the concept of necessity in Hume. See Answer. You can, of course, consider your problem outside the context of Hume. But Hume and necessity do not sit well together. But you have done real service in raising the topic.
    – Geoffrey Thomas
    Dec 29, 2017 at 18:01

2 Answers 2


It is one of Hume's central claims that we have no experience of necessity and cannot appeal to it in the explanation of anything. When we consider, say, the determination of an effect by a cause, we infer the existence of a 'necessary connexion' between them. Given the cause, the effect cannot but occur. This is a common way of thinking. But it has a purely psychological explanation due, Hume says, to the 'constant conjunction' of two types of event. We say that A causes B because and only because in our experience A-type events have invariably been followed by B-type events.

Hume backs up this claim with his basic position that ideas or concepts, such as that of necessity, require experience at some point in their derivation. But we have no experience of necessity. We do not experience one event causing another; we experience one event following another under conditions that Hume spells out. If the experience is missing then either we cannot form the idea in the first place or, if we do as in the case of necessity, it is a fabrication or fiction unjustified by experience. It is the product in Hume's language of the association of ideas.

This is all easily to be found in Hume : Treatise of Human Nature, I, III, 14, 'Of the idea of necessary connexion'. Don't be misled by the earlier I, III, 2, 'Why a cause is always necessary'. The heading is ironic and does not vindicate the necessity in causation at all.


That is the correct reasoning. The epistemological approach to regression indeed states that without a univeral context, k-disintegration occurs.

There are three types of logic in epistemology.

The linear belief chain known as logical reasoning which is known as forward thinking.

2nd teir cyclic belief chain that usually works in conjunction with the 1st for a, just as important, backward thinking logic.

The 3rd tier chain is the circular thinking for no tangible reasoning.

I think this third chain is what Hume was addressing.

  • 1
    One question, how would you define k-disintegration? is it merely the system breaking down... if uniformity is not assumed? Dec 15, 2017 at 17:50

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.