Understanding Grice's Theory of (Non-Natural) Meaning

Question

I am trying to understand Paul Grice's famous essay "Meaning".

So consider a computer system which is fed a dictionary of every English word in existence. The computer system then randomly spits out a finite number of words. Let us suppose this random assortment of words is -- remarkably -- the string "Paul Grice was a 20th century philosopher."

Now this sentence seems to be meaningful. Yet it doesn't seem fair to say that the computer had any intention of getting us to believe that "Paul Grice was a 20th century philosopher." Then am I correct in assuming that Grice would say this sentence is meaningless? Or would he say that the sentence has no "non-natural" meaning to it? What exactly would that mean?

In any event, assuming the answer is "yes -- the computer's sentence is indeed meaningless", this seems to be an absurd position to take, for clearly the sentence "Paul Grice was a 20th century philosopher" is meaningful (and true). So how is this not a counter-example to his theory of (non-natural) meaning (or at least what could a defender of Grice's theory say against this objection)?

Answer 1

The sentence is obviously meaningful to you, english speaker. But imagine a language where "philosopher" means musician: a speaker of this language would interpret the same sentence differently. Either you have to accept that a sentence can have more than one meaning (perhaps infinitely many if non-existent, ad-hoc languages are accepted: actually the sentence can mean anything you like, which is absurd) or you have to rely on the speaker's intention to fix the "real", intended meaning, which is precisely what Grice does.

Answer 2

I have looked through both H.P. Grice's 1975 article 'Logic and Conversation' and his 1957 article 'Meaning' because I think his thoughts in 'Meaning' led him to the famous implicature in 'Logic and Conversation'. This is how he introduce his implicature in 'Logic and Conversation':

Suppose that A and B are talking about a mutual friend, C, who is now working in a bank. A asks B how C is getting on in his job, and B replies, "Oh quite well, I think: he likes his colleagues, and hasn't been to prison yet" At this point, A might well inquire what B was implying, what he was suggesting, or even what he meant by saying that C had not yet been to prison [...] I think it is clear that whatever B implied, suggested, meant, etc., in this example, is distinct from what B said, which was simply that C had not been to prison yet.

Excerpt from 'Logic and Conversation' in The Philosophy of Language by A.P. Martinich

There is a literal reading and there is a suggested or implied reading. I highly doubt that Grice would say that 'Paul Grice was a 20th century philosopher' is a meaningless sentence. What he might say is that it lacks a implied or suggested reading. In its literal reading, the sentence 'Paul Grice was a 20th century philosopher' is understood without any problems. It is not a nonsensical sentence with randomly arranged signs.

SEP (Stanford Encyclopedia of Philosophy) has a section called meaning which is about that 1957 paper with the same name in their article about Paul Grice that I think would be helpful.

Edit:

This is what William G. Lycan has to say about Grice:

[…] a linguistic expression has meaning only because it is an expression – not because it ”expresses” a proposition, but because it more genuinely and literally expressses some concrete idea or intention of the person who uses it.

"speaker-meaning" is, roughly, what the speaker in uttering a given sentence on a particular occasion intends to convey to a hearer.

'Philosophy of Language' by William G. Lycan (Routledge)

It seems that the sentence 'Paul Grice was a 20th century philosopher' have no speaker-meaning.

Answer 3

Communication, non-natural meaning, is about the recognition and expression of intentions. Context plays a very important role here. As listeners, we normally use the context in which a sentence is used to extrapolate its meaning, in other words the intentions of the speaker.

In the contextless vacuum in which the sentence is presented to us in this particular case, we interpret it in the way that would have the most relevance for us - with the least effort in terms of processing it and inferring its meaning. This is because we are continuously cognitively searching for relevance, especially when engaging with language. In the context that the sentence is presented, we just imagine a speaker's intention to assert:

• Paul Grice was a 20th century philosopher.

... and this being a philosophy website and all that, we are able to identify the guy that we're discussing, and attribute some kind of truth value to the statement. We derive a lot of relevance from it. But note - if we look up Paul Grice on Facebook, for example, we'll find several hundred Paul Grices, which is problematic.

Let's imagine that someone has been talking about an old guy in a retirement home. This guy is a bit dopey and openly lets off wind etc in front of the other residents and is generally also a bit uncouth, a bit rough round the edges - and his name is Paul Grice. People are sometimes mean to him. Now the woman that's speaking about him is at the cafe with her friends. To improve his standing with them she decides to lead them on by making them think that he was a famous and erudite man in his previous existence - which he wasn't. So she says to her entourage:

• Paul Grice was a 20th century philosopher.

I think we'd agree that this sentence was false. Now, if we go back to the 'sentence' that was trundled out by the computer, it is less easy to say that this sentence is meaningful. We need a context and an utterer to be able to decode who the 'intended' referent of Paul Grice is. It's not possible in this instance to say there is one. There is nothing to choose between the Paul Grice of the OP's original question, Paul Grice in the retirement home and the other thousand contenders for Paul Grice on Facebook.

A sentence is just evidence of an utterer's intention to change their listener's state of mind. Even just to be able to decode the so-called 'sentence meaning' we need to input all the evidence we have about the speaker's intention to bring the required reading to life. So the claim that "Paul Grice was a 20th century philosopher" means something is, in some tangible sense, incorrect in this case, if by that we mean that we can decode what the speaker's intended proposition was. We can't.

What we really mean when we say that the sentence has meaning, is that we are able to see what it could mean, and also that the words that were produced by the computer have a certain relevance for us. But, it is perhaps, now we've considered it this way, easier to see it as an instance of natural meaning. We've peered at a load of inky marks and seen what someone might have meant if they'd produced the same ones. But this is a bit like looking at a cloud and seeing what looks exactly like a sketch that someone might have done of a dragon. We find it relevant and can make all kinds of inferences from it, but it is not an interpretation of an ostentatious attempt to change our state of mind by another speaker.

Answer 4

Chomsky's famous sentence "Colorless green ideas sleep furiously." is an excellent example of the kind of meaning you attribute to your statement. He tried very hard to construct a sentence that would in fact not have meaning, only to find that many people find it poetic and almost profound.

He meant nothing, and worked to evade meaning. From the proposed perspective, this is something at which one simply could not fail. The sentence does not attempt to communicate, so it lacks meaning. But we can still inject meaning into the statement.

This harkens back to Frege's differentiation between Sinn (real sense) and Bewiss (apparent meaning). Things can clearly have the latter, while lacking the former. But then it is the interpreter communicating to himself, rather than the source communicating to the recipient.