Hume's Theory of Ethics 1

click here for ordinary reading click here for a high contrast version

'The rules of morality ... are not conclusions of our reason' - David Hume

Treatise of Human Nature, Book III Part I Section I

The question

David Hume takes as his starting point the making of 'moral judgments'. We make moral judgments. 'On what basis' do we make them?

The question comes into focus when we know that an answer that Hume rejects is that it is our reason which dictates them.

Kant thought otherwise, pursuing the idea that moral judgements rested on a kind of avoidance of contradiction. Immoral acts were those which involved you willing two contradictory things. It was he thought your reason which enabled you to spot those kind of inconsistencies. Acting immorally was acting quasi-irrationally. So his was a version of the view that moral judgements were dictated by reason.

(Though because of their dates Hume couldn't actually have been reacting to Kant I'm mentioning the Kantian theory because it is one we have encountered.)

The moral philosophers Hume was actually thinking about he characterises as follows:

"Those who affirm that virtue is nothing but conformity to reason; that there are eternal fitnesses and unfitnesses of things, which are the same to every rational being that considers them; that the immutable measure of right and wrong impose an obligation, not only on human creatures, but also on the Deity himself..." Hume, Treatise, Book III, Part I Section 1, Everyman p.166.

Hume argues against this that reason cannot itself rule out any action - or any action in. On its own it doesn't give us any guidance at all.

Hume's conception of reason

Willaim Blake's 'The Ancient of Days'. Urizen, depicted here, is Blake's symbol for reason and law. He didn't think much of either I believe.

In explaining why he thinks that reason cannot be the source of our moral judgements Hume reminds us of what he thinks reason is, or at least what it does.

This is a terrific step to take. He doesn't leave reasoning, as we have done ourselves so far, as something we know all about, a thing too familiar to we human beings to require clarification or theorizing.

Twentieth Century philosophers could be accused of making that perhaps tempting assumption. We mature English speakers use the word 'reason' all the time, at some level therefore knowing the rules that givern its use. Nobody could be better acquainted with it.

Hume feels obliged to try and do better than that. He suggests it is a sort of mental operator. It does two things.

It can test for 'agreement' and 'disagreement' amongst ideas.

And it can test for agreement and disagreement between an idea and a state of affairs. (An idea in this case being not a concept but a proposition. )

How reason works

Hume thought our reason was what we would now call a computer. It accepts items as input and can compare items with each other, coming out with results of the form Yes and No.

What do I mean by 'items'? Hume called them 'ideas'. He thought your senses delivered 'ideas' to you, and that many of your ideas were given to you by your senses in that way. So an example of an idea would be the colour you were thinking of when you correctly reported that you were seeing something yellow.

But sometimes, he thought, ideas having originated in sense experience linked up or were linked up into composites, or were morphed into derivatives. My thought that 'brothers are always male'' would be a complex idea consisting of several ideas provided by sense being put together.

My reason-computer has in its memory both this complex idea and the simple ideas of which it is made up - both - 'brothers are male' and 'brother' and 'male' on their own.

The first thing the computer can do is to split complex ideas into their components. In our example:

In computer terms that doesn't sound too extraordinary. Somehow the computer has put the component ideas togther at some stage. All it is doing now is to retrace its steps and split the complex up again.

So, it has split up 'brothers-are-male' into components.

But, in this case, at least one of the components is itself complex. I'm thinking of 'brother'.

So our reason-computer can go on to split this idea into components too.

If this is so, it means that in having the idea

we are actually having the idea

Ie, the idea

is actually the idea

since the idea

is a complex idea made up of

and

So far we have established that when a person has the idea

the idea s/he has is in fact

What can your reason, according to Hume, do with this?

Well, he thinks the computer (reason) has another power. It has, he thinks, the power to compare any two ideas it has for equality - ie it can test to see if two ideas are 'identical'.

But in our idea

our computer can see that

on the left is the same as

on the right.

This is how our reason can in some cases affirm the truth of an idea of ours. It is true in virtue of the fact that our reason-computer can compare ideas. And it sometimes comes up with the result: idea A is the same as idea B.

So: when you have the complex idea that A is B you can sometimes just be saying that two ideas are one and the same. Ie 'male siblings are male' is true in virtue of 'male' occurring on both sides of the 'equation'.

Reason then, according to Hume's conception of it, can compare one idea with another.

I'm not going into his account, partly because I'm not sure I understand it in detail, but he also thinks reason can compare ideas with states of affairs. So if we have the idea

our reason can compare this idea with the actual spatial relationship of cat to mat.

The bottom line is that our reason is a tool for comparison'. It compares one idea with another, or it compares an idea with a state of affairs. But says Hume the judgement that I ought to do such and such - a moral judgment - is not based on a comparison. And that is why moral judgements are not made by reason.

Let's think the outcome through again.

The two things reason can do are:

1. Spotting identities ('agreements') between ideas.

Brothers are male

For example, 'Brothers are male'. This proposition is a way of saying 'male siblings are male', Hume thinks. What reason does is to tell us that 'male' occurs twice here - once on each side of the 'equation', as it were. In other words, it is spotting that there are two instances of the one idea, maleness. It is this sameness that makes the proposition as a whole true.

2. Spotting agreements between ideas and the facts they purport to be 'copies' of.

Eg whether there is agreement between the idea the cat is on the mat and the fact of the matter (as revealed ultimately through sense experience) - the cat's being or not being on the mat.

What I say here should perhaps be spelled out in a little more detail.

With the cat-on-the-mat example, we are talking about the second way in which Hume supposes reason works - its mode of working when it takes into account 'those relations of objects of which experience only gives us information.' Hume, A Treatise of Human Nature, Book II, Part III Section 3 Everyman p.125.

What does Hume mean here?

He thinks that we are often guided in our action by the prospect of pleasure, or pain. Our reason can help when this is indeed what is guiding our action by discovering for us cause and effect relations among objects or events. It does this by exercising its second function, comparing our ideas with states of affairs.

Ie reason can help with our decisions when it is pleasure we are after because it can tell us what is likely to bring us pleasure.

If we have a particular pleasure in view, our reason can often tell us how to get it. If pain threatens, our reason may be able to tell us how to avoid it. Reason's role here is to work out, basing itself on our previous experience and knowledge, what causes what. Reason, applying itself to past experience, may deliver the judgment that eating truffles is likely to be followed by toothache. It will therefore be useful to us in our effort to avoid discomfort.

The overall effect of what is happening here, according to Hume's theory, is that reason is telling us there is a relation of agreement between my idea - a proposition in this case, the proposition namely that eating truffles is likely to result in discomfort - and a fact, the fact that truffles do indeed cause discomfort.

Hume's theory of reasoning and moral judgements

Does Hume's view of reasoning rule out the theory that reason produces moral judgements?

Yes it does, he says. He argues that reason cannot generate moral judgments on its own because it is incapable of guiding action; and moral judgements are nothing if they are not guides to action.

So two points to argue for.

1. The thesis that we think of moral judgments as likely to influence behaviour.

Well, he thinks this is obvious. Lots of effort goes into promulgating moral judgements. (Treatise, Book III Part I Section I; Everyman p.166 para 4.) And the reason for this is surely that we think they have a deep influence on what a person does.

2. The thesis that reason doesn't on its own influence action.

Most of these are to be found in the Treatise, Book II Part III Section 3. Hume repeats in the present context just one of those arguments. It bases itself on the theory of reason I have tried to present above. Here it is:

Reason deals with truths and falsehoods. These consist in agreement or disagreement either between a fact and an idea that purports to represent it, or between one idea and another. A Treatise of Human Nature, Book III, Part I Section 1, Everyman, p.167.

Whatever can't be part of any such agreement or disagreement is incapable of being true or false.

But Hume thinks, 'passions, volitions, and actions,' are not capable of being party to such agreements or disagreements. They are, he insists 'independent'. If I have a passion, whatever it is, it doesn't 'agree' or 'disagree' with any other passion or any other thing. It is an independent existence. A Treatise of Human Nature, Book III, Part I Section 1, Everyman, p.167.

Magritte's The Human Condition Some perceptions are copies

Another way of putting this: A 'passion' has no 'representative quality' (Hume, A Treatise of Human Nature, Book II, Part III Section 3 Everyman p.127)

Hume's point is that anger (for example) is not like the perception of a physical object like a mountain, say, which on his account involves a relationship between the object and a mental entity. The perception (Hume's terminology) that is the anger (the feeling) is no kind of copy of an independent entity. It is 'an original existence' (Hume, A Treatise of Human Nature, Book II, Part III Section 3, Everyman p.127).

You therefore can't have any basis for saying that anger is contrary to reason (or 'contradictory') since a contradiction has to involve two things, one a purported copy of the other. 'It is impossible... that this passion be opposed by, or be contradictory to truth and reason; since this contradiction consists in the disagreement of ideas, considered as copies, with those objects which they represent.' (Hume, A Treatise of Human Nature, Book II, Part III Section 3 Everyman p.127)

Can't resist this second way of putting things by Magritte (his The Two Mysteries)

So the argument is:

1. Moral judgments, if anything does, influence behaviour.

2. Reason doesn't on its own influence action.

Subargument:

2.1 Passions are not copies of anything.

2.2 Reason always has to have two things to compare - either idea with object or idea with idea

2.3 Therefore reason cannot evaluate passions

3. Therefore, reason can't on its own generate moral judgments.

Credits

Cat mat pic thanks to Honeysuckle Studio

Magritte's The Human Condition I and The Two Mysteries thanks to 【自然˙啟蒙˙現代性】研讀會(and Magritte)