The Ambit of Reason According to Hume

It is no coincidence that in his “attempt to introduce the experimental method of reasoning into moral subjects”¹ David Hume sets out by examining human nature, in particular the understanding and the passions.

Hume distinguishes two types of reason. What he does is to divide the understanding, or reason, into two branches: the branch which judges from demonstration the abstract relation of ideas and the branch which judges from probability those relations of objects of which experience only gives information.

There is the purely demonstrative or deductive type of reasoning, concerning the relations of ideas and yielding mathematical or formal-logical type knowledge (‘pure knowledge’). This type of reason I will call narrow-sense reason or demonstrative reason. There is also the type of reasoning dealing with matters of fact, the reason of day-to-day life, and yielding cause-and-effect knowledge (or what Hume says should strictly be called belief). This much broader type of reason I will call either broad-sense reason or causal reason as it tells us about means and likelihoods: how to wash my car; where to have my book published; how to grow three rice crops a year and the chances of success.

Narrow-sense reasoning is a comparison of ideas.² Hume first deals with the connection of narrow-sense reason to the relation of causation and later treats the issue of being sceptical of even this narrow-sense reason.³ It is the first issue though, the argument from experience,⁴ where Hume makes a famous contribution. The question that he raises is whether this narrow-sense reason gives grounds for a demonstrative belief in the relation of causation. As Hume’s discussion of the belief in necessary connection has been thoroughly canvassed by many others, in this narrative I present only an outline.

In comparing two objects, how does the mind conclude that there is a cause-and-effect relation between them? What Hume wants to test is whether induction, the process of moving from observed past events to a belief in unobserved future events, can be known through demonstrative, a priori reasoning. Other ways of expressing this might be, Is there a deductive proof for the conclusions of induction? or Is there any deductive or logical reason to believe, of any constant conjunction, that it will continue into the future? or Is it ever reasonable (narrow-sense) to infer that one’s past experiences entail a similar outcome in future? The role of narrow-sense reason in the inference from observed to unobserved is thus the butt of the first prong of Hume’s attack on reason in Book I.

In the move from one hundred, one thousand or even ten thousand experienced sunrises to the conclusion that the sun will rise tomorrow, can there be demonstrative knowledge? Does narrow-sense reason determine our minds to make the inference? In Book I, Part iii Hume argues that narrow-sense reason can never show the causal connection between objects in the empirical world.⁵ There simply are no deductive grounds for moving from particular instances, however many, to a conclusion of nature’s uniformity.

And yet all humans do believe in causation, in “the supposition that the future resembles the past."* We are so constructed that we simply cannot help but believe in the uniformity of nature. Hume himself attributes this supposition to the imagination and his principle of the association of ideas. But it is not my intention here to examine this aspect of Hume’s explanation. It suffices to note that Hume demonstrated that the nature of humans’ belief in causation is not based on reason (narrow-sense) and to notice his response to that demonstration.⁹

The picture of Hume as the arch sceptic, the destroyer, stops here, in a land where there can be no knowledge of even the most basic things such as causation and physical existence. But this image of Hume as an extreme sceptic, a Pyrrhonist, a nihilist is a wrong representation of Hume. For “none but a fool or madman will ever pretend to dispute the authority of experience, or to reject that great guide of human life.”¹⁰ Having discredited traditional views of the influence and role of narrow-sense reason — for it cannot even lead to knowledge of the everyday world of facts, consequences and external realities — Hume shifts ground; he does not despair.

Of course we believe in causation; all humans do, he gaily admits. Total scepticism is impossible.

I should reply, that this question is entirely superfluous, and that neither I, nor any other person was ever sincerely and constantly of that opinion. Nature, by an absolute and uncontrollable necessity has determin’d us to judge as well as to breathe and feel.... Whoever has taken the pains to refute the cavils of this total scepticism, has really disputed without an antagonist, and endeavour’d by arguments to establish a faculty, which nature has antecedently implanted in the mind, and rendered unavoidable.¹¹

Reason, in the narrow-sense, may not justify empirical generalisations and the natural sciences but humans are so constructed that they naturally believe such generalisations anyway. No one can really be a total sceptic. All of us rely on a certain degree of uniformity in nature, whether that reliance is demonstratively well- founded or not. Belief in such fundamentals as causation, physical existence or personal identify is not rational (narrow-sense); it is natural and universal, a contingent but real characteristic of human nature.

As mentioned earlier in the introduction, Hume’s main concern was to bring a scientific attitude to bear on questions of human nature such as believing, knowing, feeling and evaluating. A main theme of that undertaking, in all Hume’s works, is the denigration of reason and the concomitant uplifting of what Kemp Smith refers to as this ‘natural belief. Nowhere is this more starkly exemplified than after his attack on narrow-sense reason. In matters of causation and the existence of an external world, narrow reason is impotent and cannot lead to knowledge or demonstrative belief. Yet there is belief — an areasonable¹² (narrow-sense) natural belief.

Nature has not left this to choice, and has doubtless esteem’d it an affair of too great importance to be trusted to our uncertain reasonings and speculations.¹³

So the gap is bridged, not by demonstration but by a natural human belief, and the important task of calculating likely means and results, based on empirical evidence, proceeds. I judge the future in terms of my experience of the past however areasonable (narrow­sense) this may be. The strict limitations of the traditional conception of demonstrative reason are in this way overcome. Of course one can still ask questions of justification: How many observations are ‘safe’? Are there intervening causes? And so on. But I postpone my discussion of justification and evaluation until later in this chapter and turn now to causal reason, the other type Hume distinguishes.

Hume does not think that all reasoning must be deductive. Quite the opposite is true and failing to appreciate this can lead critics to avoid, rather than confront, a main strength of his position. Reasoning, for Hume, has to do with either demonstration by comparing ideas or with inferring to matters of fact. Reasoning produces a belief or judgement or conclusion from some logical and linguistic relationship or from empirical evidence. The latter form of reasoning, arriving at beliefs from pieces of evidence, includes all of science and factual information and is the vastly more significant of the two. Only thanks to causal reason do I know my way downtown, what will happen if I pull this trigger, how to make friends or even how to feed myself, read or move this pen.

But it is essential to emphasise again just how Hume resolves this first difficulty, of how to support causal reasoning, for it is a technique he returns to again when discussing morals. He firstly examines the claim that knowledge (strict sense) can be obtained of the external world — in this case the claim that some timeless relations of ideas can illustrate beyond question the necessity of necessary connection — and shows it to be wanting. Nothing out in the fabric of the universe could ever demonstratively convince us of causation. Then Hume looks inside the human mind¹⁴ and finds his answer. We humans happen to be such that our mental characteristics include a natural belief in nature’s uniformity. The general approach is to question reason's ability to lead us to knowledge by discovering some sort of quality out in the world, and then to offer an alternative in terms of qualities of human psychology. This is how Hume attacks the pretensions of narrow reason and it is exactly the method he uses to attack the much more important target of broad-sense reason when he turns to discuss morality. It is not that broad-sense or causal reason does not count as reasoning for Hume; it is not that causal reason is insignificant in human affairs; it is simply that cause-and-effect reasoning cannot make distinctions between what is vicious and what is virtuous. No calculation of any likely outcome or matter of fact can tell us whether an action is good or bad. Moral qualities cannot be discovered from the causal relations (or deductive relations for that matter) of objects in the external world. Rather, says Hume, moral •evaluations are to be explained by looking inward at human qualities. Thus his general technique remains constant though the target shifts. Yes broad-sense reason is essential to human development, but can it motivate us to act? Can it distinguish good from bad? Can it give us a propensity we lacked before? These are the questions that arise when Hume turns to moral subjects in Book III of the Treatise.

b)