Aboutness, Questions and the Original Question.

What we need to have to settle the debate at hand is an analysis of ‘For any given question Q, Q is about X’. One way to start is by examining the components of such an analysis. This entails looking at what questions are, that is what kind of structure/entity they are, and what questions are about, that is what counts as answers to them.
One prominent analysis of questions and answers is given in Belnap and Steel’s The Logic of Questions and Answers. They hold that (for a certain class of questions) the answer to a question depends on what the asker had in mind for an answer. For example, if someone were to ask ‘Why did Ben steal the car?’, it seems that many different but equally acceptable answers could be given depending on the intent of the asker. If the asker was sociologically minded, an acceptable answer might be ‘Ben stole the car because he belongs to the lower-middleclass, is a Catholic and lower-middleclass Catholics are prone to thievery.’ If the asker was street-thugidly minded, an acceptable answer might be ‘Ben stole the car because he was the only person in his gang that could hotwire cars.’ The point here is that the answer to a question and question-aboutness are relative to the reference classes that the enquirer considers relevant.
Now let us apply this sketch of an analysis to the original question (OQ): ‘Was mars always either dry or not dry?’.
1) Presumably, (OQ) is a question like the above and hence depends on the reference classes the asker has in mind.
2) If (OQ) is a question that depends on the reference classes the asker has in mind, then it is easy to see how the question may be about any subclass of a set like {vagueness, matter, the laws of logic, space travel, H2O, language, concepts …}.
3) If (1) and (2), then (4).
4) The majority of questions, including (OQ), could be philosophical ones while also being about things other than language or concepts.
5) If (4) then philosophers are not significantly restricted in the kinds of questions they can pose.
-So far so good for Williamson. On this analysis of questions and answers, philosophical questions needn’t be solely linguistic or conceptual and our methodology has not been restricted at this level. However,…
6) If (1), (2) and (3) [i.e. the inference to (4)] holds, then analogously the questions of chemistry and other non-armchair questions are philosophical questions.
7) It is false that the questions of chemistry and other non-armchair questions are philosophical questions.
8) If (6) and (7), then [either ~(1) or ~(2) or ~(3)].
9) Either, ~(1) or ~(2) or ~(3).
On the analysis of questions given above, it seems like the right move here is to deny (3). Though it is true that questions like (OQ) can be philosophical, they seem to be so only in virtue of the asker fixing the meaning of the question with respect to linguistic or conceptual reference class elements.
I think this argument illustrates a fatal flaw in Williamson’s argument in chapter 2. He presumes that showing (OQ) to be both philosophical and not necessarily about language, is sufficient to show that philosophy is not solely interested in matters of language or concept analysis. However, now that we have seen something about the structure of questions, it is a very live possibility that (OQ) is philosophical in virtue of being asked with a subset of the reference classes {vagueness, the laws of logic, concepts, semantics …} in mind. Alternately, (OQ) is not about language or concepts only in virtue of being asked with respect to a subset of the reference class {matter, space travel, H2O, …}. But now these two ways of asking (OQ) appear disjoint, the latter is no longer a philosophical question. In brief, it is a very live possibility that (OQ) is philosophical only insofar as it is asked with a reference class containing LINGUISTIC or CONCEPTUAL elements. After an in-depth analysis of questions, answers and aboutness, Williamson’s argument is unconvincing.