Davidson, who follows Quine in many other respects, disagrees that meaning talk can be fully naturalized, and claims that to account for thinking beings and meaningful talk we have developed a battery of specific 6 See also Peregrin (2005).
Thus, in my view, the concept of rule, far from being 'supernatural' itself, enables us to account for the specificity of human language, meaning, and reason, without invoking any 'supernatural' concepts.
What I want to suggest is that the difference between being meaningful in the sense of being a suitable means for a particular end (like a hammer) and being meaningful in the sense of being expressive of a meaning (like a word) can be elucidated in terms of the difference between those practices which are straightforwardly end-driven and those which are partly governed by deliberate rules.
A spectatorial theory of meaning?
Of course not (unless by meaning something by a word we understand furnishing the word with a mental content).
No; for, of course, I am an interpreter myself, and though to find out what a word means I need to interpret its users, once I am acquainted with it, I may start to take the word as meaning what it does and in the end finally perceive it as the embodiment of the meaning.
Just as I can retrieve from my memory the function of a long unseen tool which I worked with long ago, so I can retrieve the meaning of a word whose meaning I encountered long ago.
Once I have learnt the meaning of a word, I no longer need to actually look at people using it to know that the word is governed by such and such rules and hence that it means thus and so.
Thus, though it is the process of interpretation that is constitutive of meanings, this does not mean that getting hold of a particular meaning must always involve me in interpretation.
However, in the course of my becoming acquainted with the language, in the course of my learning it, my ability to tell what a sound means becomes non-inferential – I start to perceive the sound as meaning thus and so, viz.