The theory ceiling

Theory is often referred to as an abstract discourse that aspires to elevate certain points dealing with art to a philosophical status of sorts. Anti-theory people often complain that theorists depart from the specificity of the artistic--in my case, the literary--object and the rules of critical analysis that are developed to deal with that particular object. So Judith Butler, we might say, is not such a good reader of literary texts, even though she has invaluable philosophical points to make, because she departs from the specificity of the literary object to try and use statements about it as a jumping off point for larger issues.
But perhaps theory doesn't function this way. What if, in fact, theory were more like a discourse that precisely prevents certain statements about literature or art in general from immediately, without any check, expanding themselves into philosophical reflections? What if theory actually tempers and refines statements about literature, and in fact is a constant process of laying down boundaries for a properly literary-critical analysis of literature?
I say this because I find it is often the case that New Critical (and, currently, formalist) statements about texts, which often have such an unbelievably specific notion of the literary object, often go crazy when they expand themselves into generalizations or more wide-ranging reflections. Part of what theory does is take over that process of generalization and allow it to become a complex field of debate in and of itself.
No doubt this often tempers statements about the general value or merit of a text: a New Critic might often generalize by saying that a particular text shows universal principles of the human mind, say, that it is important, nay imperative, for a humanist education to draw attention to. Theory would oppose this humanism by showing these merits are quite questionable.
But rather than see it as a process of drawing these humanist evaluations into question (which is a more pertinent way of characterizing theory when theory is seriously trying to take over--i.e. in the past), couldn't theory precisely be the process of opening up a certain area in which the operation of generalization from a text, which underlies this humanist universalizing, would be able to be articulated and refined?
Indeed, now we might have other, more accurate ways of trying to make this generalization more refined: sociological analysis is one of them (you can talk about a text as it fits into an actual market or into a practice). But I think it is useful at this juncture to think of theory as precisely operating in this area where (to put it in a different way) the micro meets the macro. In other words, theory is like a process establishing a ceiling, an upper limit, for the wild generalizations about a particular literary object, that, when reached, provokes the statement to begin to justify itself in a different way--a way that does not deal as directly with the literary object but with a sphere of more general concerns that approaches the philosphical arena. We can also reconsider it positively (that is, besides its work as a limit) as a certain vocabulary that proliferates in order to expand and give coherence to this more general zone--and not as a sort of uselessly and aggressively unspecific pseudo-philosophy: a view of theory that, I think, only produces more bad and overgeneral theory, because it instills a notion that theory's job is not to be specific at all, when it is, in my view precisely an attempt to be specific about the general.