I like your blog, but this article is of shockingly low quality. It reads as if someone never read a single piece of Kant, let alone any Neo-Kantian like Korsgaard or O'Neill, and instead thought reading 20 minutes of wikipedia was enough to discredit the most influential moral philosopher in history.
Let's take the part on the Formula of Humanity, because it's one of the most egregious ones. In this part you argue that one could believe that it is sometimes rational to treat rational nature as a mere means, e.g. because it leads to pareto improvement. But THE VERY POINT of Kantianism is that if you didn't take rational nature to be unconditionally valuable, then no end that you ever set for yourself could ever be worth pursuing (Korsgaard very nicely summarises the argument in The Sources of Normativity, that's why I referenced here).. It's essentially as if you are trying to reach something by climbing on a ladder, while at the same time kicking the ladder so it falls down: Unconditionally valuing rational nature is supposed to be the grounding of every other action you rationally pursue.
If you wanna criticise Kant you need to engage with his argument, not with a consequentialised distortion of it.
Thank you for clarifying! I have looked up the thing you referred to. I suppose there are two points I would want to press here (please correct me if I'm misunderstanding the argument):
1) I'm still not sure about the inference made. In particular, the inference from our identity as humans being the source of value/reasons to humanity being an end in itself.
It is unclear to me why we cannot simply say that our having ends or reasons derives from our rational nature, but that we do not value rational nature as an end in itself. That is, it seems underdetermined what is to be an end in itself.
In other words there seems to be nothing contradictory about having, say, the promotion of pleasure as an end in itself while recognizing that the source of my reasons is rational nature. It doesn't seem like I'm doing anything self-defeating or kicking the ladder that I'm climbing away.
Similarly (of course an imperfect analogy), I might only value winning in football because I have a brain, but that just means (even granting the constructivist framework, it seems) that I could not have had that value without my brain, not that I value my brain--it is conflating the object of my value with what confers value on it.
2) Even if we do grant that rational nature is an end in itself, it is further unclear to me that that is sufficient to reach the a non-consequentialized conception of this. That is, it is unclear to me why we get the conclusion that we should honor rational nature, rather than promote it. That is, it seems to me that the argument underdetermines in what sense we should have rational nature as an end in itself (my point is not that it reaches the promoting sense).
Perhaps the idea is just that that is what having as an end in itself is supposed to mean (your other comments seem to suggest this), but then why do we not get the conclusion that humanity should be a "schmend" in itself, where being a schmend in itself means that it should be promoted?
You're obviously more knowledgeable about this than me, so please tell me what I'm misunderstanding and/or where I'm going wrong.
The problem is that the whole idea that consequentialism treats people as "mere means" or only values them conditionally is false. Utilitarianism explicitly values all people unconditionally and always treats them as ends in themselves. Of course, it also allows you to treat people as means, but not "mere means". So if Kant is correct that there's a contradiction in treating people as mere means, that actually weighs in favor of utilitarianism and other versions of agent-neutral consequentialism that value all rational agents for their own sake.
Are you referring to the SEP article that says, "according to Kant, it is both a necessary and a sufficient condition for one’s treating persons in a morally permissible way that one treat them as ends in themselves"? Because that agrees perfectly with my interpretation of the mere means argument. The article also explicitly denies your claim that Kant is very specific about what he means.
There is a debate what's the right way to interpret Kant, but the point is that any popular Kantian theory of "treating as a mere means" is clearly incompatible with utilitarianism. Again, if you had read the article, or would be at all familiar with Kantian literature, then you'd know.
There's a paper somewhere about "finagling the categorical imperative", i.e., in order for the CI to work, we have to engage with it with the good will that the approach presupposes (otherwise we'd need to provide a more substantial critique of why goodwill alone isn't sufficient for Kant's purposes). If I find it, I'll try to remember to link it over to you.
I'd be concerned that the "egregious specifics" problem might be undermining the overall point by engaging with the CI without the goodwill it asks for, without explaining why we have abandoned goodwill at this point.
Your discussion of the deduction of the categorical imperative seems to be based on the argument in the Groundwork, but your characterization of step (5) as arguing from the datum of moral experience to the categorical imperative is more characteristic of the "fact of reason" argument in the second Critique. The "fact of reason" deduction proceeds in reverse order relative to the Groundwork deduction.
In the Groundwork III deduction, Kant's argument is roughly as follows: theoretical spontaneity -> membership in the intelligible world and hence possession of a transcendentally free will determinable by pure practical reason -> the intelligible world qua ground of the sensible world is also the ground of the laws of the sensible world -> categorical imperative.
The premise of transcendental freedom is the solution to your question about how he derives the contradiction in volition test. To determine your will independently of all sensible desire as such, you must be able to will that your maxim become a universal law. The reason is that if you couldn't will that it become a universal law, then your will would be implicitly determined by some sensible motive after all (e.g. a maxim of rational egoism).
My own reading of the idea of a maxim is a rule with the following structure. I will undertake action A in situation B for motivation See. Given this, it automatically becomes obvious why you don’t need to know the rational goal beforehand to get a contradiction in the will. For example, if your rule is lying, whenever somebody would be likely to use truthful information to hurt shrimp, then presuming everybody following the rule they would lie whenever truthful information would be used to hurt shrimp. But if everybody follows this rule, then people would know that whenever truthful information could be used to hurt shrimp, there’s no point believing what somebody says. But the motivation for your rule is very obviously to deceive people. Otherwise, you simply remain silent, so it becomes logically impossible for your rule to fulfil your motivation, which is a contradiction since the motivation is part of the rule.
The basic intuition seems to be that if it’s rational for you decide to do something, then that decision procedure isn’t just rational for you. It’s rational for everybody. Of course, this still doesn’t work because you can simply build into the rule. The condition that you only lie when everybody else would not act on the same rule that you are acting on. Since in that case, there is no contradiction in making it a universal law. Of course, there are still some things that can’t be made universal rules, for example, defecting in a prison dilemma against a copy of yourself, and in any case, the rule still permits a lot of different maxims that are mutually contradictory, so while it’s a reasonable filter for a moral theory, it doesn’t uniquely determine a single set of morals.
Honestly, I think if you recognise that the exact same decision software running on your brain would produce identical results in identical situations, then the categorical imperative becomes common sense, although that doesn’t mean, it’s actually practically possible for a human to perfectly comply with it. Still, I think it’s a reasonably good guideline for moral behaviour that you should not act on reasoning that you don’t want everybody to act on. And I will note that this still would outlaw a lot of common bad behaviour.
I like your blog, but this article is of shockingly low quality. It reads as if someone never read a single piece of Kant, let alone any Neo-Kantian like Korsgaard or O'Neill, and instead thought reading 20 minutes of wikipedia was enough to discredit the most influential moral philosopher in history.
I'm sorry you feel that way! What in particular am I misunderstanding?
(I'm not intending to respond to neo-Kantians, but obviously it sounds like I'm gravely misunderstanding or misrepresenting Kant himself)
Let's take the part on the Formula of Humanity, because it's one of the most egregious ones. In this part you argue that one could believe that it is sometimes rational to treat rational nature as a mere means, e.g. because it leads to pareto improvement. But THE VERY POINT of Kantianism is that if you didn't take rational nature to be unconditionally valuable, then no end that you ever set for yourself could ever be worth pursuing (Korsgaard very nicely summarises the argument in The Sources of Normativity, that's why I referenced here).. It's essentially as if you are trying to reach something by climbing on a ladder, while at the same time kicking the ladder so it falls down: Unconditionally valuing rational nature is supposed to be the grounding of every other action you rationally pursue.
If you wanna criticise Kant you need to engage with his argument, not with a consequentialised distortion of it.
Thank you for clarifying! I have looked up the thing you referred to. I suppose there are two points I would want to press here (please correct me if I'm misunderstanding the argument):
1) I'm still not sure about the inference made. In particular, the inference from our identity as humans being the source of value/reasons to humanity being an end in itself.
It is unclear to me why we cannot simply say that our having ends or reasons derives from our rational nature, but that we do not value rational nature as an end in itself. That is, it seems underdetermined what is to be an end in itself.
In other words there seems to be nothing contradictory about having, say, the promotion of pleasure as an end in itself while recognizing that the source of my reasons is rational nature. It doesn't seem like I'm doing anything self-defeating or kicking the ladder that I'm climbing away.
Similarly (of course an imperfect analogy), I might only value winning in football because I have a brain, but that just means (even granting the constructivist framework, it seems) that I could not have had that value without my brain, not that I value my brain--it is conflating the object of my value with what confers value on it.
2) Even if we do grant that rational nature is an end in itself, it is further unclear to me that that is sufficient to reach the a non-consequentialized conception of this. That is, it is unclear to me why we get the conclusion that we should honor rational nature, rather than promote it. That is, it seems to me that the argument underdetermines in what sense we should have rational nature as an end in itself (my point is not that it reaches the promoting sense).
Perhaps the idea is just that that is what having as an end in itself is supposed to mean (your other comments seem to suggest this), but then why do we not get the conclusion that humanity should be a "schmend" in itself, where being a schmend in itself means that it should be promoted?
You're obviously more knowledgeable about this than me, so please tell me what I'm misunderstanding and/or where I'm going wrong.
The problem is that the whole idea that consequentialism treats people as "mere means" or only values them conditionally is false. Utilitarianism explicitly values all people unconditionally and always treats them as ends in themselves. Of course, it also allows you to treat people as means, but not "mere means". So if Kant is correct that there's a contradiction in treating people as mere means, that actually weighs in favor of utilitarianism and other versions of agent-neutral consequentialism that value all rational agents for their own sake.
Kantians mean something very specifing by "using as a mere means", there's an entire SEP article on it.
Are you referring to the SEP article that says, "according to Kant, it is both a necessary and a sufficient condition for one’s treating persons in a morally permissible way that one treat them as ends in themselves"? Because that agrees perfectly with my interpretation of the mere means argument. The article also explicitly denies your claim that Kant is very specific about what he means.
There is a debate what's the right way to interpret Kant, but the point is that any popular Kantian theory of "treating as a mere means" is clearly incompatible with utilitarianism. Again, if you had read the article, or would be at all familiar with Kantian literature, then you'd know.
There's a paper somewhere about "finagling the categorical imperative", i.e., in order for the CI to work, we have to engage with it with the good will that the approach presupposes (otherwise we'd need to provide a more substantial critique of why goodwill alone isn't sufficient for Kant's purposes). If I find it, I'll try to remember to link it over to you.
I'd be concerned that the "egregious specifics" problem might be undermining the overall point by engaging with the CI without the goodwill it asks for, without explaining why we have abandoned goodwill at this point.
Do you mean Sneddon's paper on maxim-fiddling? https://www.cambridge.org/core/journals/kantian-review/article/abs/new-kantian-response-to-maximfiddling/BABDAC463DBB1F51F776DD3F67C582F8#
I love that paper.
That's the exact one—you've got a better memory than me!
Your discussion of the deduction of the categorical imperative seems to be based on the argument in the Groundwork, but your characterization of step (5) as arguing from the datum of moral experience to the categorical imperative is more characteristic of the "fact of reason" argument in the second Critique. The "fact of reason" deduction proceeds in reverse order relative to the Groundwork deduction.
In the Groundwork III deduction, Kant's argument is roughly as follows: theoretical spontaneity -> membership in the intelligible world and hence possession of a transcendentally free will determinable by pure practical reason -> the intelligible world qua ground of the sensible world is also the ground of the laws of the sensible world -> categorical imperative.
The premise of transcendental freedom is the solution to your question about how he derives the contradiction in volition test. To determine your will independently of all sensible desire as such, you must be able to will that your maxim become a universal law. The reason is that if you couldn't will that it become a universal law, then your will would be implicitly determined by some sensible motive after all (e.g. a maxim of rational egoism).
My own reading of the idea of a maxim is a rule with the following structure. I will undertake action A in situation B for motivation See. Given this, it automatically becomes obvious why you don’t need to know the rational goal beforehand to get a contradiction in the will. For example, if your rule is lying, whenever somebody would be likely to use truthful information to hurt shrimp, then presuming everybody following the rule they would lie whenever truthful information would be used to hurt shrimp. But if everybody follows this rule, then people would know that whenever truthful information could be used to hurt shrimp, there’s no point believing what somebody says. But the motivation for your rule is very obviously to deceive people. Otherwise, you simply remain silent, so it becomes logically impossible for your rule to fulfil your motivation, which is a contradiction since the motivation is part of the rule.
The basic intuition seems to be that if it’s rational for you decide to do something, then that decision procedure isn’t just rational for you. It’s rational for everybody. Of course, this still doesn’t work because you can simply build into the rule. The condition that you only lie when everybody else would not act on the same rule that you are acting on. Since in that case, there is no contradiction in making it a universal law. Of course, there are still some things that can’t be made universal rules, for example, defecting in a prison dilemma against a copy of yourself, and in any case, the rule still permits a lot of different maxims that are mutually contradictory, so while it’s a reasonable filter for a moral theory, it doesn’t uniquely determine a single set of morals.
Honestly, I think if you recognise that the exact same decision software running on your brain would produce identical results in identical situations, then the categorical imperative becomes common sense, although that doesn’t mean, it’s actually practically possible for a human to perfectly comply with it. Still, I think it’s a reasonably good guideline for moral behaviour that you should not act on reasoning that you don’t want everybody to act on. And I will note that this still would outlaw a lot of common bad behaviour.