As an empiricist, David Hume rejected Descartes’ rationalist philosophy; Hume was not a dualist, for him the concept of the mind is no more than a sum of particular impressions or perceptions. Moreover dualism is incoherent: the separation of mind and matter leaves no room for any interaction: how can an event of the mind result in a material action and vice versa?
Descartes doubted that the veracity of the senses are necessarily justified, but having found that he cannot doubt that he is doubting is obliged to invoke a deity to validate the external world, without which the Cartesian world is entirely subjective. It is solipsism: meaningful only to itself.
Like Descartes, Hume employs scepticism to try to find firm foundations for knowledge, yet it is unclear, for all his strict empiricism, that Hume does any better.
Descartes did not recognise a problem with causation. Why should he have done? Causal relationships are the basis of his rationalism part of the rational understanding from which knowledge derives. Hume however requires experience for his foundations, but in causation he finds no experience whatsoever, only a contiguous sequence of experienced events. Events themselves, Hume claims contain no implication of prior causes. Repeated experience of sequences of events, colliding objects bouncing off each other, disintegrating or deforming for example, leads to future expectations. However there is no logical certainty that a general rule can be inferred from a finite number of facts and the imagined causes for the sequences of events have no justification in logic nor in experience. Thus Hume concludes that our expectation of causes is a psychological habit of the mind.
It is curious to point out that by questioning the status of causation, Hume is weakening the arguments against Cartesian dualism.
It is at this point that Hume seems to have stepped back from pursuing the logic of his empirical scepticism. From where does the habit of expectation arise and how can we conceive of sequences of events in time and space? Why should we assume that expectations that we will maintain expectations in the future simply on the basis that we have employed the expectations in the past? There seems no reason to give psychological causes greater validity than material causes, in fact to maintain a distinction would resort to dualism.
It is worse than this. Expectations rely on memory; yet memory implies a causal relationship with events of the past. This is also true of any temporal sequence: if event A precedes event B, we only know this because we have the memory of event A, yet to trust our memory relies on a trust in causation; a causal link between event A and our memory of the event.
In solipsism there is no time only a construction of the mind, the past is a product of the imagination as much as is the present. Without causation, Hume cannot justify time. I think it could also be argued that perceptions of space also involve causal assumptions of relationships between perceptions of objects. This is the catastrophe: inspired by the progress of Science, both Hume and Descartes sought secure foundations for knowledge, however Hume’s empiricist scepticism finds an end point that differs little to Descartes’ rationalist scepticism.