Source: Prof Michael Sandel, Justice: ..., Episode 06: "MIND YOUR MOTIVE"
Question: Despite Prof Sandel's answer below, I still don't understand why JSM was wrong.
Does Kant mean: you can appeal only to consequentialism while test[ing] an action?
But if you use consequentialism as a test, then you're still appealing to consequentialism?
Motivation behind question:
John Stuart Mill [JSM] made this criticism of Kant: If I universalize the maxim and find that the whole practice of promise-keeping would be destroyed if universalized, I must be appealing somehow to consequences. If that's the reason not to tell a false promise[what?]. So JSM agreed with that criticism against Kant, but JSM was wrong.
Kant is often read as appealing to consequences: The world would be worse of if everybody lied. Because then nobody could rely on anybody else's word. Therefore you shouldn't lie.
That's not what Kant is saying exactly, although it's easy to interpret him as saying that. I think what he's saying, is that this is the test. This is the test of whether the maxim corresponds with the categorical imperative. It isn't exactly the reason. It's not the reason. The reason you should universalize to test your maxim, is to see whether you are privileging your particular needs and desires over everybody else's. It's a way of pointing to this feature. This demand of the categorical imperative, that the reasons for your actions shouldn't depend, or their justification, on your interests, your needs, your special circumstances being more important than somebody else's. That I think is the moral intuition lying behind the universalization test.