I apologize for the anecdotal nature of this thread, and would greatly appreciate feedback including vastly different experiences. I am reflecting on my own introduction to normative ethical theories and how that introduction influenced my later understanding. I can't recall the specifics and my memory is admittedly fallible, but I will attempt to recount what I take to be the important details.

Throughout my education the introduction to normative theories is done as a contrast whereby consequentialism (utilitarianism) is introduced first and as a theory that weighs moral rightness on the respective consequences of the particular brand of consequentialism. Deontology is introduced second and, in contrast to consequentialism, as a system that focuses on the moral rightness of classes of actions themselves irrespective of consequences (e.g., lying is always wrong). While this is a somewhat superficially accurate description, it is almost entirely inaccurate in the spirit and aim of deontology. Unsurprisingly, that introduction is enormously influential in a misunderstanding of deontology (Kant in particular).

Although the instruction in class is often accompanied by reading a text, likely the Groundwork, Kant can be dense and difficult for the uninitiated to read and understand. The gaps in understanding are filled in by the instructor, and are aimed towards the superficially accurate description above. The student is engaged in a sort of confirmation bias to conform the text with class instruction, and ignores the details in Kant that point away from that understanding.

The misunderstanding is exaggerated in the inquiring murderer example often cited as a criticism of deontology. Specifically, the criticism is that a moral agent that is duty-bound to truth-telling to the murderer over the safety of one's friend has a misplaced and inhumane priority. Our intuition, it seems, is that in such cases it is not only permissible but morally required to lie to the murderer if that is the only means by which to keep one's friend safe. The superficial understanding of Kant cannot make room for that nuance insofar as the class of action (e.g., lying) is ALWAYS prohibited.

The misunderstanding arises as a reductio of sorts of the Categorical Imperative and has textual confirmation in Kant if one is looking for that interpretation. The Categorical Imperative prohibits only maxims and not actions (whether class or specific), but the application of the Categorical Imperative strips away the particulars of a maxim to give a generic response to prohibit maxims as a class of prohibition. Such that, allegedly all lying maxims will be defeated by an inconsistency inherent in lying itself, which will cash out as a general prohibition of lying. However, the last step in the moral reasoning chain (i.e., a general prohibition of lying) is not a central feature of the Categorical Imperative and is not contained in the a priori or metaphysical grounding of the moral law. It is a sort of moral laziness to simplify moral judgement.

The perniciousness of the misunderstanding is in how much it gets right. The application of the Categorical Imperative does in many cases cash out as a general prohibition against a class of actions (e.g., lying) and lying in general relies on a maxim that has an internal inconsistency common across all lying maxims. However, this understanding completely misses the spirit of the Categorical Imperative. The Categorical Imperative is a decision making procedure with the aim of preserving rationality. The Law provided by Kant, "act only on those maxims that you can at the same time will to be a universal law", is not the Categorical Imperative per se but is an expression of the Categorical Imperative. The Categorical Imperative can be expressed in numerous ways (e.g., Universal Law, Law of Humanity, Veil of Ignorance, Discourse Ethics), but all must perform the same function: preserve rationality.

The duty and commitments of the moral agent are to rationality itself. When the inquiring murderer (or a Nazi) is at the door, the moral agent is not committed to the Kantian expression of the Categorical Imperative, but to preserving rationality (which is being threatened by the murderer). The moral agent must act to protect rationality, which may include lying. The Kantian expression of the Categorical Imperative can make room for the inquiring murderer, but it requires a deeper understanding than the typically taught prohibition on lying.

A problem arises in introductory courses on ethics, as the prevalence of the superficial understanding persists due, in part, to the ease with which it can be taught and understood without much thought to what it gets wrong, which I contend has severe and false implications.