helpful lies or unhelpful truths

I like to think that a Pragmatist-like concept of truth is incisive and appealing. Isn't it tidy to connect a proposition's level of truth more or less directly with the methods or actions for confirming it? But objectors understandably prefer a simpler concept of truth that disconnects the subjective role even further: they'd rather say that real is real and truth is truth. What about helpful lies or unhelpful truths? Don't these two categories show the limitations/weaknesses of Pragmatist-like truth?

My immediate reply is to admit that I'm not presenting an infallible path to truth, and if the alternatives claim such an infallible path then the alternatives are mistaken. My longer reply is a Pragmatist-like analysis of those two categories. In practice, how are specific propositions classified as helpful lies or unhelpful truths?

Obviously, the secondary classifications "helpful" and "unhelpful" indicate that the propositions are either aids or hindrances for accomplishing particular goals. However, the primary classifications "lies" or "truths" force the follow-up question of how we know to sort a given proposition into one of those two. Although the original argument merely assumes that the relevant propositions can be definitely classified as lie or truth, this tactic won't work in general because personal assumptions are so often wrong. The sensible conclusion is that there must have been some primary procedure for us to know for certain that the proposition under consideration is lie or truth, and we must have gained that knowledge independently from the proposition's secondary classification of helpful or unhelpful.

Hence classification of actual propositions as helpful lies or unhelpful truths presumes two procedures with conflicting results, a primary and a secondary. An occurrence like this surely demonstrates that the secondary procedure, i.e. that single "helpfulness" test, is a flawed test for truth. But the invalidation of one procedure by another doesn't invalidate the overall concept of Pragmatist-like truth.

I don't argue that true implies "always helpful for all goals" and false implies "always unhelpful for all goals". Reality is too complicated. Propositions are too numerous. Goals are too flexible. That's exactly why more than one confirmation procedure is a good idea anyway and the human judge resolves or prioritizes conflicting results. Moreover, helpful lies and unhelpful truths can be essential mid-points on the journey to propositions of greater refinement. A partial answer could be helpful for some purposes regardless of its ultimate disqualification, and the unhelpful nature of a puzzling piece of data could challenge incorrect preconceptions. The dread of (temporary) contradictions isn't part of Pragmatist-like truth, which is intended to evolve. Intolerance of contradictions is more emblematic of fastidious otherworldly concepts of truth.