Hume’s Position Considered for the First Time

One obvious question arising from Hume’s position is why reason has been seen to conflict with emotion or the passions. Hume concedes that,

only in two senses [can] any affection be call’d unreasonable.

A fear of sailing off the edge of the earth is areasonable²⁶ because the planet is not flat. Attempting to satisfy my desire for wealth by growing money trees is contrary to reason because money does not grow on trees. But in truth in both these cases it is not really the passion but the accompanying belief which is contrary to reason.²⁷ Consequently, while it is contrary to reason to seek an immortality potion or eternal youth, it has nothing to do with reason — and so is not strictly contrary to reason — to want to be young while one remains alive or even forever. This is Hume’s point in the famous purple passage:

‘Tis not contrary to reason to prefer the destruction of the whole world to the scratching of my finger. ‘Tis not contrary to reason for me to chuse my total ruin, to prevent the least uneasiness of an Indian or person wholly unknown to me.²⁸

It is not their contrariness to reason (broad or narrow) which makes these deliberately provocative examples so shocking. Rather, it is a lack of proper or acceptable feelings, sentiments or preferences (which is Hume’s implicit message). Hume would certainly disapprove of the finger lover — but this just shows that disapproval is not solely rational, a point he takes up in Book III and I take up in the next two chapters.

Another reason people make the mistake of thinking reason can oppose the passions, according to Hume, is that what is properly termed reason tends to be associated with “certain calm desires and tendencies, which tho’ they be real passions, produce little emotion in the mind...”²⁹ People use ‘reason’ when they really refer to calm passions. Thus Humean passions are not limited to those that produce great strength of feeling, like jealousy or greed. Nor is strength of feeling a determinant of a passion’s motivational efficacy. Calm passions, like the desire for one’s long-term happiness, can oppose the more recognised, violent passions. And it is these calm passions that are often confused with reason in thought and language.

This leads me into an important aspect of any discussion of human reason. The popular usage of the term and its cognates must be distinguished from the term in a philosophical sense.³⁰ This demand serves to focus thought and to prevent waffle. (There is, perhaps, tendency enough to turn ‘reason’ into a vaguely metaphysical, catch­all concept.) For Hume, reason properly so-called is mathematical or logical knowledge as well as experience-based calculation and causal belief, or any combination of these, and perhaps the state or process that leads to this knowledge and belief. But it is this, only this, and no more. As we shall see, Hume recognises that the adjectives ‘reasonable’ and ‘rational’ in common usage often also serve, explicitly or implicitly, to endorse particular courses of action. While these and other usages of the term reason, somehow more inclusive, might well be conventional — even permissible — they must still be distinguished from what Hume calls and carefully delineates as reason.

I wish by this demand to drive into the open those, not least legal philosophers, who appeal to reason as a final arbiter in questions of morals and legal reasoning. Let us always settle, as a first step, what is being referred to when terms such as ‘reasonable’, ‘rational’ and ‘reason’ are employed.

It is arguable that Hume underestimated or underemphasised both the intimacy of the partnership as well as the variety and importance of the roles of reason. Those differences fall within the basic Humean framework of the desire-belief theory. To break out of (transcend) that framework, to construct a distinctively and purely cognitivist anti-

Humean view, requires more than just the amalgamation of reason and sentiment or the incorporation of sentiment into reason (via the back door). It requires the rejection of sentiment as inessential to moral motivation.³

In other words, a hard and fast dichotomy between wanting and believing may well be too great an abstraction for practical purposes.³² At the end of the day the two components may best be understood as inseparably amalgamated or fused. But if this be so it leaves, as Shaw argues, the Humean framework undamaged. Desires, wants, the appetitive are still a sine qua non of action, whether or not the causal reasonings that influence them and allow other desires at times to prevail, are impractical to separate out (save for heuristic and instructive purposes) or not. Either way, reason alone cannot move action. Sentiment remains essential. And that is enough to make the Humean case. Those who dislike or disavow the Humean conception of reason’s inertness need to say more than that reason and desire are inextricably linked.

A different tack would be to say that Hume’s conception of reason is unduly restrictive, too tied to the normal causal world. Yet if other conceptions of reason are to be used to refer to something more than Hume’s, to a disembodied perspective or to a capacity to intuit or to ‘see’ inalterable and eternal relations between objects in the external world (non-ordinary empirical ‘truths’) or to a priori categories necessarily part of the human mind or to anything else, then such other usages need to be defined and explained, not simply asserted as metaphysical fact or used without any definition but with all the usual associations of the term. These possibilities, in the context of moral theory, will occupy me in chapters to come. However as regards reason, any non-Humean position would need to do at least two things: Firstly, it would need to describe what the characteristics of reason are beyond Hume’s two-part model. Secondly, it would need to make the case that such a conception of reason is more illuminating about human conduct and human nature than the model propounded by Hume — I.e. of narrow, demonstrative and broad, causal reason being the only types of reason but existing in conjunction with, and perhaps inseparably tied to, feelings, desires and dispositions which motivate humans and themselves require explanation. (And the explanation given of desires and propensities need not be the one Hume gives, grounded as it is in the theory of ideas, but could be anything from an evolutionary explanation to a Freudian one.)

Hume in effect attempts to take the mystery away from reason. There is no divine element to the understanding. It is true that the concomitant emphasis on feeling and sentiment, natural inclinations and beliefs throws up something of a mysterious cloud around innate pre-dispositions and nature.

Nor can I omit here, in this first consideration of Hume’s position, to raise the very un-Humean question of justification or evaluation. Hume examines why and how humans come to perform any action A. His answer would follow a line such as this: I want to achieve goal G, to bring about state S catering to this desire. I believe that action A will be an effective method (given all known circumstances, past causal regularities, and future likelihoods) to achieve G. Thus action A is due to:

i) goal G which, contingently, I wanted or desired to occur; and

ii) my belief that A is an effective way of obtaining G.

Actions can therefore be ‘explained’ by having regard to my actual propensities and actual beliefs. Hume does not subscribe to the traditional faculty of practical reason, which tells us what to do or what actions to take, but divides it into a belief component and a disposition (or feeling) component.

There are two components of action A which conceivably need to be justified.

# 1) the belief (which recommends doing A to achieve G); and

#2) the desire (which provides the motive).³³

To ask whether an action is reasonable, whether in effect it is justifiable or to be approved, involves an evaluation of both these branches separately.

Branch #1) is rational — can be justified —■ in two possible senses. Firstly, if it is a true belief. If A, in the particular situation with all the variables, was in fact an effective way to achieve G. This involves causal reason making a judgement about means and likely consequences and being correct. There are indications in Hume’s reason-as-representational argument³⁴ that he limits ‘reasonable’ beliefs to these true beliefs. But, if so, Hume is surely too restrictive.³⁵ The label ‘reasonable’, when referring to beliefs, encompasses more than just true beliefs. The label has to do with the method of cause-and-effect generalising: with what procedures have been followed; how many instances or what sources have been relied upon; how careful the checking has been. And the ‘reasonableness’ of the method which leads to belief is solely a matter of experienced efficacy. In other words, the ‘reasonable’ method — justifying the tag ‘reasonable belief — is itself a function of contingent experience and what happens to produce the safest results. And this is the second sense in which branch #1) of an action A can be justified — where A was a course which, let us say, an impartial, knowledgeable person would have agreed at the time would lead to G. This second sense merely involves causal reason making a judgement about means and consequences which falls within some empirically established standard, regardless of whether the judgement proves to be correct.

But how can branch #2), regarding the desire component, be evaluated or said to be rational? On the Humean scheme there are two compatible possibilities. One is to say, simply, that desires cannot be rational. My desire or urge to want G is simply not rational; a desire cannot have the label ‘reasonable’ attached because

it is simply a fact, albeit a contingent fact. But this first possibility is somewhat misleading in its simplicity. The other Humean possibility, which has greater explanatory power, is to say that a desire for G can be evaluated as ‘reasonable’ where the belief that G is a good means to G₂ (a further goal) is reasonable and G₂ is desired. This second possibility involves a means-end chain which continues on through G₃, G₄, etc., until it reaches an underived or simple desire. (E.g. I want to study hard so I can do brilliantly well on this exam and eventually obtain a first class law degree which will ensure I find a job at a multi­national law firm, thereby earning a large income, saving lots of money and affording myself the chance to travel widely.) In the course of the chain any particular desire (say for G₄, getting a job at a big law firm) can be evaluated for whether it is a reasonable means (say for G₅, earning lots of money). But even here, on the way to reaching the end of the chain, it is really the surrounding beliefs that are being justified or evaluated as reasonable — not any such claims about the desire component in terms of its nature as a pure want, as an end in itself. And, a fortiori, this must be true of the final link, the underived or simple desire. It cannot, on the Humean scheme, be evaluated in terms of reason. Thus, on either of these two Humean interpretations, I think it is accurate to say that a desire as a pure end cannot be evaluated as rational, as being in accord with or contrary to reason. (Alternatively, for those who prefer amalgamating the desire and belief components on the grounds I have sketched above, the desire part of a desire-belief motivating force cannot be evaluated in terms of reason.) Only as a means to some further desire or as a term of endorsement, and then often hidden and masquerading as a justification or evaluation, can a desire be termed reasonable³⁶ once one accepts Hume’s strict conception of reason.

As a result, in responding to the query of whether a particular action is reasonable there are two possible options for a Humean. Either one can base one’s answer solely on branch #1) — the action is reasonable if the belief is justified — or one denies that actions are either reasonable or unreasonable because the pure desire component is not ultimately open to review in any way having to do with reason. (I.e. one cannot ‘justify’ desires in terms of what is ‘reasonable’ or ‘rational’ although one may still ‘evaluate’ desires in terms of normal tendencies and the subjective responses such desires usually evoke.³⁷) If what is to be justified is broadened from a single action to a whole, complex decision-making process then the same distinction holds. Again, ‘reasonable’ either signals that the package of beliefs is causally true (or would fall within some empirically established standard) or it is merely a term of endorsement of the process. Is there such a thing as a practical reason which tells me what to do? Only if the aspects of actions for which there can be ‘good reasons’ are limited to the whole package of surrounding beliefs. But reason has no role in vetting the worthwhileness, as opposed to the consequences and sufficiency, of wants. For Hume, a rational or reasonable desire would be analogous to a kind meteor or a green logarithm.

The important point for my purposes here is to take from Hume his emphasis of the difference between narrow/broad reason and all the rest of human responses which might go under the denomination of sentiment, feeling, disposition, propensity, preference, emotion or passion. On this Humean basis, to speak strictly and philosophically, the description of an action or course of conduct as ‘reasonable’ should refer only to the belief or believing component together with an implicit recognition of the existence of a motivational component not describable in terms of reason. But to speak in terms of common usage and common understanding (and Hume himself believed strongly that even the most philosophical of us cannot help thinking with the ‘vulgar’ most of the time), the description of an action or course of conduct as ‘reasonable’ admittedly involves more than a mere reference to the surrounding beliefs. This needs to be emphasised. As generally used the adjectives ‘reasonable’ and ‘rational’, either implicitly or explicitly, serve also to endorse the action being described. In this looser sense of ‘reasonable’ and ‘rational’, as terms of approval, an action that is generally regarded as appropriate can be called reasonable, as can one that the evaluator herself regards as appropriate. There is nothing wrong with or objectionable about this ‘vulgar’, endorsing usage; we all make use of it. However the fact that we commonly couch the act and state of approving in the language of reason, why ever we do this, is certainly not determinative of whether approval is in fact a rational, reason-based process. This can be forgotten when the two senses are intermingled or even when we are being strictly philosophical. Consciously or otherwise, describing an act as ‘reasonable’ or ‘rational’ lends a prestige to the act of approving³⁸ and tends to lead us to pre-judge why and how humans approve. For my part, therefore, I will henceforth take the precaution of following a rigid policy of using the term ‘reasonable’ and its cognates in the strict, philosophical sense and will demand that mere approval be expressed in other ways. The challenge, in turning to morality and law in later chapters, will then be for others to explain how wants, desires, sentiments and feelings, that is to say all the human capacities and responses lying outside Hume’s bipartite conception of reason, can be evaluated in terms of ‘reasonable’ and treasonable’ —■ a task which involves elucidating how reason, alone, can create underived wants. Or put another way, one task for the anti-Humean is to show how reason alone can move people to act.

d)