In a very interesting, if rather bizarre, paper called “Berkeley’s Metaphysical Grammar,” Colin Turbayne develops an interpretation of Berkeley’s ‘language of nature’ theory which takes extremely seriously Berkeley’s remark, in the New Theory of Vision, “that visible figures represent tangible figures much after the same manner that written words do sounds” (sect. 143). The relation of vision to touch is, in other words, the same as the relation of written English to spoken English. A particular visible idea signifies a particular tangible idea not in the way a word signifies its referent, but rather in the way a written word signifies its spoken counterpart.
Now, there is one text where Berkeley seems to say the opposite (Alciphron 4.10), but let’s set that aside. It seems there is a more obvious and pervasive problem: Berkeley is constantly saying that our visual ideas inform us about tangible ideas (see, e.g., PHK 44). Does this not imply that the relation is like that of ‘dog’ to dogs, rather than like that of the written word ‘dog’ to the sound dog?
On reflection, perhaps not. Here are two examples. First, consider this sentence: ‘the president released a copy of his remarks to the media in advance, to inform them of what he was going to say.’ Here the written words have the purpose of informing about the spoken words, but the written words are not, in terms of their content, about the spoken words. Rather they are about whatever the spoken words are about (why warrantless spying is great, and credence .51 is sufficient for reasonable belief, or whatever else presidents talk about).
Second, consider the purpose of a script for an actor in a play. The script informs the actor of her cues – i.e. of what the other actors will say – and instructs her to respond by saying certain other things. But again, the words in the script are not about what the actors are going to say. They are about whatever is going on in the play (politically motivated cross-dressing or running away from bears, or whatever else they do in plays).
So, while Berkeley’s language of nature theory is admittedly pretty weird, it looks like it can evade this particular objection. This also helps to evade a different objection, namely that the language of nature only talks about itself, which would make it a pretty odd sort of language. If Turbayne is right, and visible ideas don’t actually refer to tangible ideas, then we don’t need to say that one part of the language is about another part. Instead we can say, as I have argued we should, that the language is about minds.
(cross-posted at blog.kennypearce.net)