Wednesday, January 04, 2017

G.E. Moore and the Definition of Definitions

235 days until classes start.

Today, I started W.D. Ross, The Right and the Good. It is another classic text that I have seen mentioned on several class syllabi for the University of Colorado, so I thought I should get it read. This and Rawls Theory of Justice are both on my "before I start graduate school reading list".

And it is not just reading it. It is studying it - as I am trying to do with Moore and Sidgwick. I will be writing some things on these papers as well.

However, I am still harvesting G.E. Moore's Principia Ethica Chapter 5 for useful insights.

I have uploaded a newest (20170104) version of a short paper discussing Moore's naturalistic fallacy on the Desirism facebook group.

In working on this draft, I had some thoughts that fall into the area of philosophy of language.

I am a pragmatist about language. Language is a tool that we invent to help us fulfill our desires. Moore was talking about the definition of the term 'good' almost as if it is something that exists in nature for us to discover, and not as something that we invent to serve a purpose.

Moore's claim that 'good' is an undefinable term particularly bothered me. If it is an 'undefinable term', then how do we get it to be the case that others know what we are talking about when we use it? How can we build a useful, practical language out of undefinable terms? There must be some way to get together with other people and reach an agreement to the effect, "Let's agree, when we use the term 'good', to be talking about the following . . ."

And once we reach this agreement, there must be a way to pass this fact on to new users of the language - whether they are born into it or migrate in from other languages.

I added a section to the paper that said,

Language is a tool used largely for communication – to cause ideas to appear in the thoughts of the person who receives the communication. A ‘definition’ is anything that one can do to cause the desired idea to appear in the thoughts of the recipient. One way to define a term is by listing its parts and the relationships between them. However, that cannot be the only way to define a term.

In fact, when it comes to defining the term ‘good’, the question is: What can we do to get the same idea to show up in the minds of different people when they receive the term ‘good’ in a communication?

One of the ways which we can define a term – one that refers to a part of a whole – is by listing its relationships to the other parts. Using the concept of a horse – which is a complex concept made up of multiple parts – we can define those parts by describing their relationships to other parts. We can define a molar by describing its positon in the mouth, and each of the other bones by describing their location relative to the other bones.

In fact, some of our terms are strictly relational – they are terms we use to describe the relationships between one thing and another. Consider, for example, the term “sister”. You can know everything there is to know about the parts of a horse – its liver, its eyes, its tail, its blood – and none of this will help you to understand the idea that this horse is the sister of some other horse.

One might say that we can define ‘sister’ in that a sister is a part of a family. However, Moore does not consider the possibility that we can define parts according to what they are a part of. Complex things are defined in terms of their parts, but parts are not to be defined by their relationship to the whole. If terms could be defined according to what they are a part of, then every term would have a definition and there would be no “undefinable terms.”

One particular type of relational definition that is particularly useful is a functional definition – where we define something according to the role that it plays in a larger system. One can hand you all of the parts of a mousetrap. However, an understanding of all these parts and even an understanding of how the parts are put together does not allow a person to understand that this contraption has a function – to trap mice. In fact, any number of devices, having different structures and different parts, can all be properly called a ‘mousetrap,” having nothing in common but the fact that they are used to catch mice.

There are certainly cases in which it is impossible to define a term. I cannot think of anything I can do to cause my dog (assuming I had a dog) to have an idea of a disjunctive syllogism when I use the term ‘disjunctive syllogism’ – or even to have an idea of good when I use the term good. I think I can get him to have a concept of the term ‘toy’. At least, I can use the term to create behavior that I can predict towards the object I have in mind when I use the phrase ‘get your toy’ when speaking to my dog.

If we cannot get different people to have the same thought when they receive the term ‘good’, then we cannot have communication. Somehow, there must be a way to accomplish this – which means that, somehow, there must be a way to define the term ‘good’.

One may object to the idea that the definition of a term looks to creating a common idea in the thoughts of those who use it by claiming we have terms that may well lack a common idea. How do we know, for example, that what you see when you see red is the same thing that I see when I see red?

Yet, what matters in the definition of ‘red’ is not this ‘qualia’ or the sensation of redness. It is found in the fact that, while we are working together defusing a bomb, and I say, “cut the red wire,” you correctly identify the wire that I want you to cut. The qualia of redness is entirely irrelevant to the meaning of the term.

Ironically, this is a place where Moore‘s naturalistic fallacy actually has an application. A person experiences a certain sensation with all things red and makes the mistaken inference that the term ‘red’ refers to that sensation. But it does not- because others can be using the same term without having the same sensation. And when one is corrected for having misused the term, it is not on the basis of using it to refer to the wrong sensation. It is because of a failure to identify the correct objects.

The attempt to define ‘good’ is an attempt to get a common, useful idea to show up in the thoughts of different people when they use the term in a certain set of contexts.

No comments: