Alrighty then. Time for metaethical mayhem! G.E.Moore’s Open Question Argument (the OQA, also known as the Naturalistic Fallacy) is famous and widely considered to be the foundation of Metaethics. Moore was a British philosopher who outlined his theory in Principia Ethica. The fundamental question: What is ‘good’? Let’s dive right in with the Philosophy revision! 2009 style!
What is good?
Firstly, check out the Intro to Metaethics if you haven’t already. You’ll need that background knowledge. Now, you probably have all sorts of ideas about what you think ‘goodness’ is, but for now, let’s define it as ‘pleasure’. So we think goodness=pleasure. We might also think of goodness as being reducible to pleasure. For example: anything that is pleasurable is good. This is known as Analytical Naturalism. (There is also Definitional Naturalism). Put simply, we are naturalising goodness (reducing it) into pleasure (a natural property). Hence, we would be walking down the Cognitivist road and saying that there are objective moral facts. In this case, our objective moral fact would be pleasure.
It is also important to note that when we say goodness=pleasure, we are not saying that pleasure is identical to the property of goodness. We are saying that pleasure has the property of goodness. Pleasure belongs to the class of things which are good.
But hold it right there! says Moore. Goodness cannot be defined in natural terms!
You what now? Natural terms? What Moore means here is Natural properties, like the ones that we encounter in science and our day-to-day lives. Greeness is a natural property, as is the roughness of paper or the heaviness of a rock. These are all what most consider to be the natural properties of objects in the world around us. They are causal – they affect us and are detectable by our senses. It is also more controversial (and Moore may disagree), but psychology might also be thought of as ‘natural’. The inputs we receive from objects are processed in our mind, which consists of grey and white matter (among other gooey things). It all boils down to the sciences.
In contrast Non-natural properties might be those that are not detectable by the senses like physical objects are. They are not causal and are usually pictured as floaty, metaphysical things that are neither here nor there. One might also choose to lump non-natural properties in with Supernatural properties, such as having the property of being approved by God. But, if you read my introduction, you may, like me, find supernatural properties implausible.
So, why is Moore saying that Goodness cannot be defined in terms of natural properties? (In fact, he says Goodness cannot be defined at all (Moore  1993:91), even in terms of non-natural properties!) The reason why is because he claims it commits the naturalistic fallacy. But before going back to our goodness=pleasure proposition, consider the old bachelor example:
A Closed Question
A bachelor is an unmarried man. (Bachelor = Unmarried man)
To test it, ask the question:
Is an unmarried man a bachelor?
Answer: Yes (assuming you don’t nitpick and saying that divorced men, widowed men and single men are different types of bachelor). The reason why is because, analytically, the word bachelor simply means ‘unmarried man’. This is known as a closed question. Actually, even considering the question, ‘Is an unmarried man a bachelor?’ shows conceptual confusion, because it necessarily is the case that all bachelors are unmarried men. We can’t even ask such questions without looking stupid!
Okay, now consider Goodness:
An Open Question
Anything which is pleasurable is good (Goodness = Pleasure)
But, is anything which is good, pleasurable? More exactly:
That thing is good, but is it pleasurable?
If you want to think of it in terms of classes:
That thing belongs to the class of things that are good (have the property of being good), but does it belong to the class of things that are pleasurable (have the property of being pleasurable)?
Hmmm. Now this is not so clear-cut. Not everything that we consider good is pleasurable. (For example, sacrificing a kidney to save a friend or enduring a painful marathon to raise money for charity). Hence, for Moore, it is an open question as to whether good=pleasure.
Perhaps you were thinking all along that pleasure could never equal goodness? Fine. Try substituting something else:
Anything which is charitable is good.
That action is charitable, but is it good?
The Open Question Argument still holds, albeit less clearly. Moore’s point is that one can always ask the question ‘Is X good?’ There is never a definite answer in the way that ‘All bachelors are unmarried men’ can provide us with. This is very similar to the line of reasoning we saw in the Introduction to Metaethics, with Euthypro and Socrates debating whether holiness was loved by the Gods because it was holy or holy because it was loved by the Gods.
But, asking such questions is significant. In contrast to looking stupid when asking if our bachelor is an unmarried man, we can ask ‘is anything which is charitable a good thing?’ quite reasonably. We aren’t stupid for asking questions like that.
Hence, goodness is not conceptually identical to pleasure (or anything else). We can quite well conceive of situations where something is pleasurable or charitable without it necessarily true that it is good. So, goodness seems like it is indefinable in terms of natural properties like pleasure or the feeling of being charitable. It remains an open question as to what goodness is, argues Moore.
So what on earth is goodness??
Objections to the OQA
Moore would say that goodness is a simple and indefinable property, in the same way that yellow cannot be defined in any other way than simply as ‘yellow’. However, several objections have been raised in response to Moore’s definition of goodness. Let’s take a look:
Begging the Question: Frankena argues that we do betray conceptual confusion by saying ‘X is good’ and then asking ‘but is it good that X’? Simply put, it doesn’t make sense to question ‘but is it good that X’ after first establishing that ‘X is good’.
But, for competent users of moral language, it seems that this is a valid question. A better objection then:
One might suppose that the apparent openness in the relevant questions we perceive is mistaken. That, ultimately, we are wrong in thinking the predicates (good, bad – etc) refer to moral properties. In fact, they don’t refer to properties at all!
We could admit that something is pleasant, but not go as far as saying that it is good. We could think as defining something as good as a value statement. Hence, goodness does not refer to natural properties nor to non-natural properties – it points to our value judgements. Taking this view would lead one down the path of Non-Cognitivism. (The idea that moral judgements do not express beliefs or truth-apt statements, they are just expressions of our approval or disapproval. For example, Emotivists would argue that ‘goodness’ refers to the ‘Hooray’ and ‘badness’ to the ‘Boo’ in the Boo-Hooray theory.
This seems plausible, but it is sacrificing a lot to get there (depending on whether or not you are sympathetic to Non-Cogntivism).
This is a more sophisticated response to the OQA put forward by Frege. He famously distinguished between Sense and Reference when making moral judgements. The basic idea is that two objects can both have different senses, but actually be referring to the same thing.
Take the ‘water’ and ‘H2O’ example:
Water and H2O have the same reference. Water is water and H2O is the chemical symbol for water. In other words, they both point to the same thing.
However, they have different senses. People prior to the 18th
century did not have the concept of H2O. Therefore, when they learned the chemical makeup of water, they learned something new a posteriori. They had the old sense of water – the blue stuff in the sea – and the new sense of water – two parts Hydrogen and one part Oxygen. Yet both refer to the same thing.
The same analogy can be made with Hesperus and Phosphorus – the evening star and morning star. We assigned different names – different senses – for what we later discovered is actually the same star. But before learning this fact, we truly thought that there were two different stars – essentially two different references.
With this in mind, could we not now say that the link between goodness and pleasure, or goodness and charity (etc) has not yet been found. They could just be different senses but have the same reference and it is possible we can find that reference in the world a posteriori. The sense-reference objection, then, provides a strong objection to both Analytical and Definitional Naturalism, but leaves another counterpart – Synthetic Naturalism – to play with.
But let’s consider this OQA in greater detail. What are the Semantic facts that Moore is stating in his OQA? And what is the Ontological conclusion? Here’s is my lecturer’s (Dr. Pekka) clearly laid-out example:
(1) Two properties are identical only if the predicates that ascribe them are synonymous (the semantic test of property identity).
e.g. The property of being a vixen is the same as the property of being a female fox, because ‘vixen’ and ‘female fox’ are synonymous or analytically equivalent.
(2) So, goodness is identical with a natural property N only if ‘good’ and ‘N’ are synonymous.
(3) If two terms are synonymous, then substituting one for the other does not change the meaning of a sentence (the substitution test of synonymy).
e.g. ‘Vixens run fast’ means the same as ‘Female foxes run fast’.
(4) Substituting any ‘N’ for ‘good’ changes any sentence’s meaning.
e.g. ‘Pleasure is good’ says more than ‘Pleasure is pleasure’; asking “This is pleasant, but is it good?” need not betray any conceptual confusion.
e.g. “Bachelors are unmarried men” is not saying more than “Bachelors are bachelors.”
(5) So, ‘good’ is not synonymous with any naturalistic term ‘N’. [3, 4]
(6) So, being good is not identical with any natural property N. [2, 5]
Got that? Let’s have a look. The semantic facts mentioned are 1-4 and the conclusions are 5 and 6. If we are to criticise Moore on the Sense and Reference point, I think we would be looking at pulling apart point 1. I might re-jig it like so:
(1) Two properties are identical only if the predicates that ascribe them are synonymous *by nature of their reference* where ‘reference’ is that as understood in Frege’s account of sense and reference).
With this in place, 4 would cease to function, because we do not know ontologically that goodness and, say, a natural property like pleasure don’t refer to the same thing. (Ontology is our ‘picture’ of the world). In other words, goodness and pleasure have the possibility of referring to the same thing but, quite naturally, we have very different senses for them.
This is known as Synthetic Naturalism or ‘metaphysical naturalism’. Essentially, we agree that goodness and some property N are not analytically identical, but we think that they stand for the same property. Not only that, but this fact is discoverable a posteriori. This is outcome of the inclusion of Frege’s sense-reference idea in the original equation. For example, ‘goodness’ and ‘pleasure’ are not analytically identical in the same way that ‘bachelor’ and ‘unmarried man’ are, but they can refer to the same property, if only we could find the reference.
Moore could reply to this by saying that, in contrast to our knowing certain facts like yellowness a posteriori (through experience), goodness is known a priori (without the need for experience). This certainly seems to follow our own common sense – we seem to know intuitively what good things and acts are. But the moral naturalist might argue that goodness is actually known a posteriori. I will come on naturalist accounts like Railton’s over the next few days.
Okay. That’s quite enough of Moore for now. Must move on! Much to do! ^^
‘An Introduction to Contemporary Metaethics’ by Alexander Miller (2003)
The lecture notes of Dr. Pekka Vayrynen at the University of Leeds