[Another incomplete draft, published at end of year. I didn't finish this, and it isn't as polished as I'd hope, but may be interesting anyway.]
Kant, eh? Famously deep and impenetrable. I am of course reading him in translation. He's not big on defining his terms and I therefore frequently have to guess what he means in important matters. This is annoying; bring back Popper, all is forgiven.
Kant feels the need to put morality on a sound theoretical footing. In this he is like many others going back to the antient Greeks. He fails, in my opinion; and in this he is also like the others. Is the task even of any value? As he himself notes:
There is no one, not even the most hardened scoundrel-provided only he is accustomed to use reason in other ways-who, when presented with examples of honesty in purpose, of faithfulness to good maxims, of sympathy, and of kindness towards all (even when these are bound up with great sacrifices of advantage and comfort), does not wish that he too might be a man of like spirit. He is unable to realise such an aim in his own person-though only on account of his desires and impulses; but yet at the same time he wishes to be free from these inclinations, which are a burden to himself. By such a wish he shows that having a will free from sensuous impulses he transfers himself in thought into an order of things quite different from that of his desires in the field of sensibility: for from the fulfilment of this wish he can expect no gratification of his sensuous desires and consequently no state which would satisfy any of his actual or even conceivable inclinations (since by such an expectation the very Idea which elicited the wish would be deprived of its superiority): all he can expect is a greater inner worth of his own person.
I include that partly maliciously so you can get a sense of his language. So for me the task is less to produce a theoretical understanding of morals, than a practical understanding1: how is it that even "scoundrels" have a good moral sense? To this end, Adam Smith's Theory of Moral Sentiments is more to the point.
Is there anything of practical value in the work? Yes, three things.
1. The Categorial Imperative (CI): Act only according to that maxim whereby you can, at the same time, will that it should become a universal law. Or some variation thereof.
2. Treating people as ends in themselves, rather than means to your end.
3. The idea that acting morally correctly from a sense of duty and / or respect for the law, is more worthwhile in some sense than exactly the same actions, performed because you wanted to anyway.
But just like Descartes, this is a thin core wrapped around by a mighty barricade of long words. Point 2 is nice, and not the sort of thing that the Plato of the Republic can say, but does not I think count as Deep. Furthermore it is problematic. If everyone is an end, how do you resolve conflicts? Are you ever allowed to kill people? Kant forbids suicide, so killing someone else would appear problematic, that would make war impossible, which you might agree with; but I don't think he did. But we resolve these problems in a practical way: the rule is not absolute, derived from axioms like geometry and eternally true: it is merely a rule of thumb, a guideline to thinking about how you might act. Although it doesn't really work in everyday life: if you go to the shops, you will probably treat the shelf-fillers and the checkout people as simple means-to-your-end; treating them as ends in themselves is almost meaningless. You will recognise that they have their own ends, if you stop to think about them at all, but that's different.
Point 3, similarly, can be pushed too far and Kant proceeds to do exactly that, asserting that moral worth comes only from acting in accord with duty, untainted by your own inclinations. Adam Smith's Theory of Moral Sentiments gets closer to the right answer, when he says that a benefactor would think himself poorly rewarded if the benefactee acted coldly from only a strict sense of duty.
Point 1 whilst an interesting thing to think about and argue about is not that different from the Golden Rule of immemorial antiquity: treat others as you would be treated. Indeed, arguably that is a better formulation not subject to the many objections to Kant's version. Inevitably, the Kant fanbois aren't happy with this, but that's the problem with reading commentary on philosophy: so much is written by fanbois.
Kant's is an axiom schema not an axiom or maxim; I think there's some confused language around that, but as it wasn't terribly interesting I didn't trouble with the details. The first obvious problem is with "act only": does this mean that all of my acts, down to the most trivial, should be so constrained? This seems implausible, but I don't see anything clarifying it. The answer I think is that Kant is so focussed on moral problems that he has forgotten everyday life. So I think he means it to apply to moral problems. But that doesn't dispose of the problems, because for example "improving oneself" is a moral duty and while the general concept can be generally willed, we can't all write blogposts about Kant; so again we have to allow some latitude for his imprecision.
At which point, we begin the fun game of "can we think of things that can be generally willed but which K wouldn't like?". One, which to give him credit he tries to deal with, is suicide. He forbids it. Why? The true answer is probably that it was part of his religious upbringing; but it is also a rather inconvenient generally-willable thing. K's arguments against are not plausible, so you'll have to read them yourself. His next example, deception, fares better. The third, "failing to cultivate one's talents", is as unconvincing as the first; and the fourth, charity, fares no better.
However, K gives only these negative examples. He provides no positive examples - at least, not in this work. So we're left with the possibility that no maxims satisfy the CI, which would be regrettable, from a pure-philosophy viewpoint. We could imagine that the negatives of his examples - don't kill yourself, tell the truth, be charitable, develope your talents - would suit. The first, alas, is somewhat questionable, and not really very useful anyway. The others verge on the bleedin' obvious, so also don't get you very far.
[I have a section heading on God here, so I clearly intended to write something, but have forgotten what.]
Kant ties himself in knots over Free Will. His problem is that morality implies choice and there is no choice without free will. For me on Free Will, see this from 2018 ("Update" section); or this from 2006. So the correct solution, for Kant's purposes, is to simply assert as an axiom the existence of Free Will. But, no, that's too simple for him, so he blathers on pointlessly page after page.
* Kant's Cats.
* Moral Knowledge: A Question of Timing - Bryan Caplan and Ayn Rand.
1. I wonder why I thought that? Now (2021/07) it doesn't seem true.