Categorical Imperative

It sometimes seems to me that Kant's Categorical Imperative is a sciencefictional conceit. From Grounding:

Kant asserted that lying, or deception of any kind, would be forbidden under any interpretation and in any circumstance. In Grounding, Kant gives the example of a person who seeks to borrow money without intending to pay it back. This is a contradiction because if it were a universal action, no person would lend money anymore as he knows that he will never be paid back. The maxim of this action, says Kant, results in a contradiction in conceivability (and thus contradicts perfect duty). With lying, it would logically contradict the reliability of language. If it is universally acceptable to lie, then no one would believe anyone and all truths would be assumed to be lies. The right to deceive could also not be claimed because it would deny the status of the person deceived as an end in himself. And the theft would be incompatible with a possible kingdom of ends. Therefore, Kant denied the right to lie or deceive for any reason, regardless of context or anticipated consequences.

It's the absolutism here that is SF, I think; the argument from Bizarro world ('argumentum ad Mundo Bizarro'?). The world is this way; imagine what it would be like if it were the exact opposite.

One problem is that this is predicated upon a sort of one-to-one mirror transformation. If everybody always told the truth, then one would always know where one stood; but if nobody ever told the truth, then one would still always know where one stood. The real ethical problem comes in a world in which some people are always truthful and some habitual liars and you can't be sure which is which, and more to the point where most people sometimes lie and sometimes tell the truth. The key here is inconsistency, precisely the thing absent both from Kant's Categorial moral alternate reality and its Mirror-Universe evil double.