One might say that in the beginning of Meaning, Grice tries to distinguish between two different vocabularies – the naturalistic and the non-naturalistic one, given the fact that “to mean” figures in both of them. Also, in doing so, he attempts to use a neutral vocabulary, apparently.
I would say that Grice tries to distinguish between two practices – one which belongs, togheter with other practices, to the “talking about the workings of nature” heading and another, which belongs to a different set, which I would call “making sense of what we do”.
It is disputable, however, that one can distinguish between such practices from a neutral standpoint. “Practice”, like “meaning”, can be used within both types of practices just mentioned.
Each of the criteria he uses to distinguish between meaningN and meaningNN could be criticized in this way. I will look at them in turn.
(1) tells us that when “to mean” is used within the second practice, we could always say (a) “X means Y but it is not the case that Y”, altough this does not hold when “to mean” is used within the first practice.
Now, we can speak about saying (a) while we are talking about the workings of nature or while we want to make sense of what we do. In the first case, the proposed criterion does not work. Of course it could hapen that someone says “Those clouds mean that it is going to rain, but it is not going to rain”. This is not an impossible natural event.
Perhaps the person saying such a thing believes that a “Rain God” has sent those clouds “only to scare us” (for whatever reason). So we could say, as Grice himself does, when introducing his fifth criterion: “one could say (a), but she would mean something entirely different”.
However, in order to be able to see the difference we must be able to get involved in the two different practices I was talking about. Similarly, in order to understand how “X meansNN that Y” and “The fact that X occured meansNN that Y” can meanNN something entirely different we must be already acquainted with the both the practice of making sense of what we do (or say) and that of decribing the workings of nature.
The circularity is not obvious here because the relation between the fifth criterion and the conceptual distinction between the said practices is, as Brandom would put it, a pragmatically mediated semantic relation.
In fact, we do not need the distinction while talking about saying (a) within the first practice (let us call it WN; the other practice we might call MS). The criterion just does not work. The distinction between WN and MS comes into play only when we are talking about saying (a) while engaged in MS.
(2) and (3) introduce similar criteria – being able to talk about “what was meant by X” or about the agent who “meant Y by X” while using “to mean” within MS, but not when using “to mean” within WN.
Again, the criteria simply do not work if they say that some linguistic behavior is impossible, as they would do if they were formulated in the WN practice. If Grice does not notice this, it is because he is already engaged in MS when he produces the criteria, in spite of his attempt to express (2) and (3) outside the MS specific vocabulary. One could talk about what was meant by the clouds and answer the question “Who means that it will rain by those clouds?” (the Rain God) while appearing to talk about the causal relation between the sky being clouded and the event of raining, but she will mean something different from what we say when we say that “those clouds mean that it will rain”. Of course, in order to see the difference we must be able to recognize that speaker is engaged in the MS practice.
And now let us turn to (4). Of course one could write down “Those clouds meant ‘rain'”. This is not an impossible natural event. If we want to make sense of what this person does, of course we could. The quotation marks can be used in may different ways. In fact, one could use them in the WN practice very well. Suppose that from the placement of some firework tubes on the ground a person could infer that when fired, those fireworks will make some word to appear in the sky and she will express this by saying that those firework tubes mean “Happy New Year” (if this exemple is criticized for the hidden qualifications present in saying that the firework tubs mean “Happy New Year”, perhaps a better one could be devised). This is, however, of little importance. It is not difficult to see that the person who believes that clouds are like messages sent from a Rain God would have no difficulty to say that some coulds mean “it will rain”. I think we can plainly see that we are in the same situation with (4) as we were with (5), which I will not discuss again.
To conclude, when Grice introduces (1)-(5) he does not use a neutral vocabulary. I would rather say that he uses a vocabulary which can belong to both WN and MS, altough he is in fact engaged in the MS practice. In order to be able to use the criteria not only must we recognize that he is engaged in such a practice (in order to understand the criteria), but we must also be able to properly engage in MS and WN ourselves (in order to apply them correctly).
However, I do not believe that Grice is trying to convince us to distinguish between meaningN and meaningNN. We would not be able to accept the distinction if we were not able to recognize (and engage into) the two different practices which the distinction presuposes. He is only trying to conceptualize the distinction (or to “make it explicit”, to use a famous phrase).
From this perspective, the real problem is that we do not seem able to conceive the distinction from within the WN practice. If this is true, than the attempt to offer a naturalistic account for non-natural meaning is irreparably flawed, since it rests on the assumption that we can isolate non-natural meaning, in order to reduce it to natural meaning (in this respect, eliminative materialism seems a better strategy in philosophy of mind than other forms of reductionism; in fact, eliminative materialism does not reduce anything at all).
So one might wonder why does Grice, after giving an account, from a non-naturalist perspective, of “meaning” as being used in MS, open the way for the WN practice (for the completion of his analysis of meaningNN in terms of intentions). In other words, he seems to start by distinguishing two vocabularies but to end by allowing us to engage into the only practice according to which the two vocabularies are one and the same (since the concept of “intention” is not further analysed by him, its treatment might be considered the task of cognitive neuroscience, as indeed happened later on).
Of course, one could say (as Wittgenstein and Anscombe do) that intentions are not psychological objects, so in order to understand what it is to have an intention we must engage in MS and talk about intentional actions (to have the second intention from Grice’s original definition of utterer’s meaning would be, for instance, to act overtly, while the “presence” of the third intention would perhaps depend on the context in which an act is performed). In this case, however, Grice’s intentional semantics would collapse into a version of a use theory of meaning.
*
A necessary condition for refuting an empirical statement is the production of a proof that the respective statement is wrong – I could prove the statement wrong without refuting it, of course, but in this case it would be mistaken to consider my proof a message; see Grice’s example with Herod, who proves to Salome that St. John is dead by showing her his severed head, but does not tell her that St. John is dead.
Analogy: I cannot say ‘hi’ to somebody without meeting that person. Meeting X seems to be a necessary condition for saying ‘hi’ to X, but is this a similar situation? If by ‘meeting X’ I want to talk about some action (in MS), then perhaps yes. If by ‘meeting X’ I describe the occurence of an event (in WN), then no.
So it can be said, perhaps, that Grice’s third condition comes down to performing an action which has a non-communicative action as its necessary condition. The problem, of course, is that this account uses the concept of ‘non-communicative action’ which, in its turn, depends on some concept of communication. In addition, it is not clear why should this condition be a reduction of Grice’s third one (about intending that A does r at least in part due to A’s recognition of A’s intention that A should do r) to action talk.
But maybe we do not want to translate Grice’s third condition into action talk, but only to replace it.
Instead of calling something “a non-communicative action”, we could start by talking about “acting in one’s environment” (I will assume that we do not consider other persons part of our environment).
Acting in my environment is not causaly changing something in nature. It is, rather, inspecting my environment, producing an artifact from a raw material, using an artifact, making adjustments to an artifact, simulating such actions a.s.o.
“So what do i do when I meet someone else, if this is my action? The other person is not part of my environment, yet my action of meeting her is not a communicative action.”
[…]
Tags: grice, nonnatural meaning