Whatever else Clements means, it unquestionably holds that the but - for test is a test based on «necessity», not merely «
substantial connection» beyond de minimus, and that «material contribution» is not a reference to some method of
establishing factual causation on the balance of
probability.