Hume acknowledges that the skepticism employed in the previous section could never undermine our reasoning from common life: nature always wins out against abstract reasoning. However, he does claim to have shown that there is a step in our reasoning from experience that is not supported by any argument or process of understanding. There is no solid reason why we should reason according to cause and effect, and yet we never fail to do so.
Hume notes that someone thrown into the world with no prior experience would have no understanding of the process of cause and effect. Life would be an unintelligible string of unconnected events. We cannot sense causation, nor (as Hume has argued in the previous section) is it present to our reason. Hume's answer is that our inductive reasoning regarding experience is derived from custom and not from the understanding. This is why we need to see a process recur many times before we can begin to see two events in the process as causally connected. I need only examine the diagram of one circle in order to derive through reason the properties that all circles share in common. However, we must see many collisions of billiard balls and other objects before custom can implant in us the inference that the movement of one object is causally connected to the movement of another.
Without custom, Hume remarks, reasoning that concerns matters of fact could not extend beyond memory and present sense experience. We could not speculate nor even act if custom had not implanted in us the ability to see certain actions as having certain consequences. Nonetheless, Hume points out, all reasoning from experience ultimately falls back upon simple impressions. What I know about past ages might come from reading a history book, or what I speculate about the future might ultimately fall back upon observations I am making in the present. Our speculations about unobserved matters of fact rest upon a constant conjunction with our present impressions.
Hume suggests that we make inferences by means of the imagination, but draws a careful distinction between fiction and belief. Fiction is the product of pure imagination by means of which we can conjure up all sorts of strange images derived from our simple impressions, such as unicorns, alien civilizations, and what have you. Belief is a combination of imagination and a certain sentiment that we cannot control that suggests to us that our imaginings correspond with reality. When some memory or sense impression is present to our minds, the force of custom will then carry the imagination to think of something to which that impression is constantly conjoined. This force of custom forms our beliefs, and creates a more vivid, forceful, and firm version of our pure imaginings.
Cause and effect, like the other two laws of association discussed in section III, allow the mind to move from one thought to another. When these laws of association are led by custom, they form very strong instinctive beliefs. Hume remarks that it is fitting that our knowledge of causation should be formed by instinct rather than by reason. It is very important that we see the world causally, since it is the source of all action and speculation, and reason is too unreliable a tool. Young children have still unformed reasoning abilities, and even intelligent adults commit countless errors in their reasoning. The instincts enforced by custom are far less susceptible to error, and are thus a far stronger means of securing our knowledge of cause and effect.
The terms "skepticism" and "naturalism" are frequently mentioned in discussions of Hume, and his relationship with each is hotly debated. Hume is termed a skeptic on account of the doubts he raises as to the capabilities of reason. The classic account of modern skepticism is found in Descartes' ##Meditations##, in which all knowledge based on sensory experience is cast into doubt. We might read Hume as going even farther, casting our ability to reason inductively into doubt. While Descartes ultimately squirms away from his doubts, Hume sticks to his, claiming that we have no rational justification for anything outside of immediate sensations and a priori reasoning.
Essentially, Hume doubts the rational foundation of everything that is useful and helps us get by in the world. All action and speculation is based upon suppositions of cause and effect. If I did not think my actions would have any consequences, I would not act. For instance, I go to work because I know I will get money if I do so, and I help my friend because I know my friend will be better off if I do so. If I had no reason to expect any consequences from my actions, I would have no reason to go to work, to help my friend, or anything else besides.
Hume's line of skepticism cuts to the very heart of our conception of ourselves as rational beings. He makes us question what we mean when we say we do things for a reason. In fact, his argument seems to imply, what we call our reasons are not reasons at all, or at least they are not rationally justified.
This brings us to the naturalist line in Hume's philosophy. While Hume denies that we have reasons for believing or acting as we do, he also explains the causes for our behavior and our actions. He argues that induction and causal reasoning are implanted in us by custom and constant conjunction. In replacing reason with custom, Hume reconceives the nature of human thought and action. Most philosophy, in particular the rationalist philosophy of Descartes, sees human beings as primarily rational animals, informed and guided by reason. Hume's reconception sees us more as creatures of custom and habit, much like the animals we so frequently try to set ourselves above.
While Hume's discussion of custom and constant conjunction may seem odd to us, it is in fact just another way of framing something that should be relatively clear. Both Hume and the traditional philosopher would agree that certain events invariably follow certain other events, and both Hume and the traditional philosopher would agree that our behavior is largely dictated by our knowledge of this sequence. The difference lies in the fact that the traditional philosopher would then argue that there is some principle of cause and effect that we know and can see in operation between two connected events. Hume denies that we know any such principle, suggesting instead that habit simply implants an expectation in us that events will fall out in a certain pattern. He uses the term "constant conjunction" to suggest that we cannot say that two events are causally related, but only that we constantly find one followed by the other.
The harsh limitations that Hume sets upon reason might lead us to question the validity of science and the scientific method that Hume holds so dear. An interesting and significant point is that only philosophy, and not science, ever makes any claim regarding the certainty of causal reasoning. All scientific knowledge comes from experience, but science is also careful never to assert the certainty of this knowledge. For instance, Newton's three laws all come from induction: he observes that certain events invariably follow upon one another, and draws up laws to explain this constant conjunction. However, all laws of physics are nothing more than hypotheses. A physical theory, we could argue, can never be proved, but only disproved. Every piece of evidence in its favor only serves to make it more likely, but no evidence could ever make it certain.
Only philosophy, in its yearning for certainty, has tried to suggest that there is such a thing as a law of cause and effect. Science rests content in making predictions based on experience without claiming any kind of certainty or privileged reasoning to back these predictions up. Hume might then also defend his own philosophy, saying that he proceeds according to a similar method.