⁶ This last claim is a somewhat delicate one. I am not using as a premise the claim that we cannot make sense of the possibility of substantively algorithmically decomposing the capacity to be aware of a full range of complex relational properties, by deploying a suitable vocabulary. That is part of the conclusion I am arguing for. I am claiming, first, that the ability to ignore the vast majority of complex relational properties that are irrelevant to a given inference in the sense that they fall within its range of counterfactual robustness cannot be taken as primitive with respect to a substantive algorithmic practical decomposition of discursive practices-or-abilities, and, second, that we have no idea at all how even primitive non-discursive abilities that could be substantively algorithmically elaborated into the capacity to form the complex predicates in question could be further elaborated so as to permit the sorting of them into those that do and those that do not belong in the range of counterfactual robustness of a particular inference.