So Gillian Russell over at logicandlanguage.net very nicely mentioned the birth of this blog. She also expressed a great deal of puzzlement at my claim that philosophy of mind offers an explanation of problems like the liar.

<

p> Well I suppose I shouldn’t make bold statements like that without explanation so here goes. So first I am going to argue that it is not sentences which inherently have truth values, nor propositions. Rather it is mental representations. Essentially I agree with Searle when he defends the thesis that many experiential states like belief inherently have conditions of satisfaction. Certain of these experiential states we judge to be correct or incorrect (true/untrue) depending on whether their conditions of satisfaction are satisfied or not.

<

p> Sentences then are said to be true if they induce the right kind of mental representation and the conditions of satisfaction of that representation are met. They are false if the conditions of satisfaction are not met and neither true nor false if they don’t evoke the right kind of mental state. Admittedly more needs to be said about what the right kind of mental state is here. It certainly isn’t belief as we can judge the truth or falsity of sentences independently of belief but this I will leave to later.

<

p> Now the reason the liar and strengthened liar are problematic is that as well-formed sentences they seem like the sort of thing which ought to be true or false or correspond to a proposition. If we could give a compelling justification of why liar sentences should not get truth values or correspond to entities which get truth values this would address the difficulty. We could then insist that sentences like the liar as well as the strengthened liar aren’t true or false anymore than ‘asdfujknsdfkj’ is true or false. In other words despite their apparently acceptable surface structure they really aren’t making claims.

<

p> Since I understand sentences to be true or false depending on the conditions of satisfaction of mental representations the next move is pretty obvious. I deny that liar sentences correspond to the correct sort of mental representations. Essentially my claim is that these sort of mental representations can only have conditions of satisfaction which refer to external reality or which refer to the correctness of already formulated mental representations.

<

p> So in other words I am arguing that these mental representations are inherently stratified in a similar way to what Quine requires for comprehension instances in New Foundations. Admittedly one might ask what extra work does the theory of mind stuff do, why not simply insist on this directly in our theory of truth. My response is that these restrictions seem ad hoc and unmotivated in a theory of propositional truth. However, I think an understanding of theory of mind can independently motivate this condition.

<

p> In effect this lets us justify these restrictions on truth by reference to empirical observations. I certainly find that when I hear a liar sentence I don’t create a mental representation with conditions of satisfaction like I do when I hear a normal descriptive sentence. This no longer needs to be justified as some reasonable property that truth must possess but is now simply an empirical fact about our mental experiences. Sure one might wonder what things would be like if this wasn’t the case but then one just isn’t talking about the same notion of truth.

<

p> Sure more work needs to be done to spell this all out satisfactorily but this is at least explains why I think the solution lies somewhere in the philosophy of mind.

<

p> For more explanation and some compelling arguments in favor of mental events inherently having conditions of satisfaction read Searle’s Intentionality.