I'm inclined towards the view that necessary/a priori truths lack worldly truthmakers, and are instead constituted by the epistemic facts of what it would be ideally rational to believe (truth as the end of inquiry, and all that). One common response people make at this point is to raise the possibility that there is no uniquely rational end-point to philosophical inquiry. Maybe the disputes between Kantians and Consequentialists, mereological universalists and nihilists, etc., would persist even under conditions of ideal rationality. There could be two (or more) equally maximally coherent belief sets. What then?
If the reasons in favour of either view are perfectly balanced, i.e. there really are no possible arguments that would decide the issue one way or another, then it seems intuitively right to me to say that there is no objective fact of the matter which of them is true. Right? (It's not as though we're just incurably ignorant about some contingent matter of fact. The matter under dispute is supposed to be logically necessary, true no matter how the world might turn out, so it really ought to be knowable by reason alone.) So constructivism yields precisely the right result here, it seems to me. It's independently plausible to think that a determinately true philosophical view ought to be determinately rationally superior. So if a philosophical question has no uniquely rational answer, nor does it have any uniquely true answer.
Matters become more complicated if we imagine, not a plurality of views that equally qualify against the standards of ideal rationality, but a plurality of rational standards (each of which leads to a different view, let's say). I'm not sure I can even make sense of this idea. But supposing it were true, should we conclude that there are likewise multiple truth predicates, one for each form of ideal rationality? P is true1 iff P is ideally rational1, true2 iff ideally rational2, etc. It all sounds a bit crazy, but I guess it wouldn't be so bad if there was a large amount of overlap between all the rationalities. We could then supervaluate and say that P is true, simpliciter, iff it qualifies according to all the ideal rationalities. Otherwise, it is just relatively true, depending on which form of rationality you assess it against.
Is that the best way to interpret these scenarios, do you think?