Friday, 9 March 2012

Korsgaard's Kantian Moral Theory (Benjamin Mitchell-Yellin)

The following is a guest-post by Benjamin Mitchell-Yellin, which provides a succinct summary of Korsgaard's Kantian Moral Theory:

According to Christine Korsgaard, our first-person experience of agency reveals a requirement. We need reasons to act and to live. You are a creature that stands at a “reflective distance” from your motives. You are not simply determined by your desires to act this way or that, but you can and do choose which desires to act on.
When you deliberate, it is as if there were something over and above all your desires, something which is you, and which chooses which desire to act on. This means that the principle or law by which you determine your actions is one that you regard as expressive of yourself” (Korsgaard (1996), p. 100).
You regard your reasons as self-imposed.

In deliberation, on Korsgaard’s view, you form conceptions of your “practical identity,” and these are the source of your reasons and obligations. A conception of your practical identity is “a description under which you value yourself, a description under which you find your life to be worth living and your actions to be worth undertaking” (101). When you choose to act on a given desire, you do so because this is justified by some or other of your practical identities. Typically, one has many practical identities—a philosopher, a parent, a neighbor, a citizen, a student, a teacher—and each one gives rise to reasons and obligations. “Your reasons express your identity, your nature; your obligations spring from what that identity forbids” (Ibid, p. 101).

There is one practical identity, however, that has pride of place on Korsgaard’s view. Most of your practical identities are contingent. You could have valued yourself under different descriptions (and might do so in the future). But your identity as a human being is necessary. Korsgaard argues for this claim as follows:
It is necessary to have some conception of your practical identity, for without it you cannot have reasons to act. We endorse or reject impulses by determining whether they are consistent with the ways in which we identify ourselves. Yet most of the self-conceptions which govern us are contingent. … Because these conceptions are contingent, one or another of them may be shed. … What is not contingent is that you must be governed by some conception of your practical identity. For unless you are committed to some conception of your practical identity, you will lose your grip on yourself as having any reason to do one thing rather than another—and with it, your grip on yourself as having any reason to live and act at all. But this reason for conforming to your particular practical identities is not a reason that springs from one of those particular practical identities. It is a reason that springs from your humanity itself, from your identity simply as a human being, a reflective animal who needs reasons to act and to live. And so it is a reason you have only if you treat your humanity as a practical, normative, form of identity, that is, if you value yourself as a human being. (Ibid, pp. 120-121)
The claim that your human identity is a necessary part of your self-conception is central to Korsgaard’s moral theory. Your human identity both “stands behind” your particular practical identities and is the source of your moral reasons and moral obligations. These points merit extrapolation.

The central idea of Korsgaard’s view is that you impose reasons and obligations on yourself by valuing yourself under certain descriptions. Each particular description under which you value yourself is a particular practical identity of yours—teacher, parent, citizen—and imposes a consistency constraint on candidate motivations for actions. You have reason to act on those motivations consistent with one of your practical identities and are obligated not to act on those motivations inconsistent with any of your practical identities.

But the fact that you have any practical identities at all means that you value yourself under the description of one who needs reasons to act and to live. That you have contingent practical identities entails that you have the necessary practical identity of a human being. Your human identity is explained in the same way as all other practical identities: you value yourself under a certain description. But it is special in two ways. First, your human identity is implicitly affirmed in the adoption and maintenance of all of your particular practical identities. It “stands behind” them. Your particular identities are normative only given that it is normative. The normativity of any particular practical identities is parasitic on the normativity of your human identity. And your human identity requires that you have some particular practical identity or other. Without these particular identities you would not have reasons to do particular things. Your human identity is necessary, in other words, because you do act for reasons, and this presupposes that you value the need to do so. Second, your human identity is the source of all of your specifically moral reasons and obligations. Your human identity gives you reason to value others’ humanity, and it obligates you not to flout the value of others’ humanity.

This introduces a distinction between moral and non-moral reasons and obligations. One has a moral reason to act on those motivations consistent with one’s human identity and a moral obligation not to act on those motivations inconsistent with one’s human identity. One has a non-moral reason to act on those motivations consistent with one of one’s particular practical identities and a non-moral obligation not to act on those motivations inconsistent with one of one’s particular practical identities. This makes room for conflict between obligations. For example, a non-moral obligation may conflict with a moral obligation when it would be both inconsistent with one’s human identity to act on a given desire and inconsistent with some particular practical identity not to.

The resolution of such conflicts is dictated by the structure of Korsgaard’s view. From the two special characteristics of your human identity, it follows that morality is both rationally inescapable and overriding. Morality is rationally inescapable because your human identity, the practical identity that underwrites moral reasons and moral obligations, is necessary. It is not one that can be shed. Morality is overriding because, in a case of conflict, the conflict must be resolved by shedding the source of one or the other conflicting reasons or obligations. But since the source of your moral reasons cannot be shed, it will always be the case that a conflict between a moral reason or obligation and a non-moral reason or obligation will be resolved in favor of the moral reason or obligation. Morality always wins the day because the source of its normative force is a necessary feature of human agency.



Many thanks to Ben for the clear and concise summary of Korsgaard. I am quite struck by her definition of a human being as "a reflective animal who needs reasons to act and to live." I remain unconvinced. There is a great deal of empirical evidence that what we call “reasons” are often after-the-fact rationalizations of our actions. If this is right, then this may be evidence that we would still be motivated to act even in the absence of reasons.

And at the risk of going off on a tangent, I might add that I also don't get the assumption (taken to be nothing less than a truism by many) that our ability to reason is our most defining or fundamental trait as human beings. Sure, our reasoning ability may be far superior to that of all other members of the animal kingdom. But it does not seem to follow that being human is most fundamentally about being a reasoner. Dolphins are believed to have (by far) the best hearing of any animal. Does it therefore follow that being a dolphin is most fundamentally about hearing? Why then should the fact that we can reason better than other animals entail that being a human is most fundamentally about being a reasoner?

Moreover, to say that we reason better than all other animals is not to say that we reason particularly well. The bar may well be quite low. Reasoning may play such an insignificant role in animal life and action in general that saying we are better at it than everyone else is not really saying very much. For example, Daniel Willingham has argued (and many cognitive scientists would agree) that the human brain is not designed for thinking, but rather to save us from thinking. Most of the brain's greatest feats are processes that take place far outside the reach of our reflective access. There are many things, such as allowing us to see a 3-dimensional world, that the human brain is better at than reasoning. Thus, to say that humans are better at reasoning than all other creatures is not to say that reasoning is the thing we do best.

But I digress. My question, simply put, is what underlies the assumption that being a human being is most fundamentally about being a reasoner or rational actor? Can anyone offer me some guidance, here?

Nick said...

Avery, Chris has refined her view in her (2009), according to which we constitute ourselves AS human beings by acting rationally. It's still hard to swallow, but you get a little more detail, there.

Personally, I've never been able to swallow this move:

"But this reason for conforming to your particular practical identities is not a reason that springs from one of those particular practical identities."

Williams' "one thought too many" problem is delicately concealed in this text. If you are acting in service of some identity-conferring project (like being a good husband), Korsgaard's position implies that you need a higher-order reason to do so. That is to say, ALL of your reasons for action that derive from you being somoene's husband are parasitic on some higher-order reason, which she thinks springs from your "humanity".

I think this is absurd: I do not need a reason to be someone's husband... I AM someone's husband, and that fact is all I need to act on husband-based reasons. So it is for any practical identity: it is its own source of reasons.

Peter Strempel said...

Avery, if the cogito ergo sum recognition is not enough to convince you about humanity as being about the ability to reason abstractly, what about the the very acts we are engaged in here?

The debate about what it is to be human is what makes us unique in the universe we know of. That debate is about reasoning for its own sake, not for survival or procreation. Is that not enough?

Ben M-Y said...

Avery, thanks for posting this on your blog. I am happy to get some feedback on my description of Korsgaard's view. And I am happy as well to have a forum to try out some defenses of it. I am not sure if I think hers is the right moral theory, but I am sympathetic to many of its elements.

Let me attempt to answer some of the concerns raised the above comments.

Avery: Perhaps it will help to answer both of your worries--the worry about confabulation and the worry about the fundamentality of human reasoning--to recall the familiar Kantian move of considering questions about human agency and morality from within a certain perspective (the practical perspective, the deliberative perspective). If we consider the claim that we need reasons to act from within the perspective of one engaged in the project of deciding what to do, then it is not clear how the evidence about ad hoc rationalization is supposed to bear on the issue. For all this evidence shows, I take it, we may be deluded about the sources of our behavior even from within the deliberative perspective. That is, when we consider what to do, it may seem to us as if we are considering reasons, even if a more accurate (empirical) description of what we are up to is that we are considering ways of making sense of what we already have decided to do (are inclined to do, want to do, are doing). Similarly, if the perspective we adopt on ourselves as agents essentially involves a conception of ourselves as reasoners, this may explain the assumption that our ability to reason is our defining feature, as you put it. The fact that we are good at reasoning need not enter into the explanation at all. Rather, it is the fact that we must conceive of ourselves as reasoners that does the work.

Of course, you may want to reject this appeal to a deliberative perspective or to reject the account of what this perspective commits us to. But, insofar as this is the Kantian's move, I think she has the resources to handle your concerns.

Nick: You are correct about Korsgaard's refinement of the view in later work. That's helpful to have in mind. But I am not sure that it helps to answer Avery's worries. In fact, it may only exacerbate them.

As for the one-reason-too-many criticism, I think it might arise if Korsgaard appealed to a higher-order reason stemming from one's human identity. But I do not think that this identity gives one a higher-order reason on her view. Rather, I think that her view is that one's human identity is one among many practical identities, all of the same "order." You do not need a further reason to have any particular practical identity, such as that of a husband. Rather, your having this practical identity commits you to valuing having reasons. But this is just to say that it commits you to valuing your human identity--as one who needs reasons.

The normative force of one's human identity is not "positive" in the way you suggest: it does not get added to each and every reason/obligation stemming from a particular practical identity. Rather, the normative force of one's human identity is "negative": it provides a constraint on which particular practical identities one can consistently adopt. If one adopts a particular identity that conflicts with an identity one already endorses, then the conflict must be resolved by shedding one or the other of these identities. In cases of conflict between one's human identity and a particular identity, the conflict can only be resolved by shedding the particular identity. One's human identity is necessary, so it cannot be shed.

This does not commit Korsgaard to the claim that one has a reason to engage in the project of romancing one's spouse because one is a husband AND because one is a human being. It does, however, commit her to the claim that when one takes oneself to have reason to romance one's spouse, this commits one to endorsing one's identity as a husband AND one's identity as a human being.


Nick, thanks for the update on Korsgaard's view! I actually think that your elaboration goes some distance towards addressing the worry I raised, and perhaps your worry as well. Let's begin with my worry. Korsgaard's "old view" seems to be that while one can decide to stop being a husband (by, say, getting a divorce), one cannot decide to stop being a human being. Thus, being human is a practical identity that rides shut gun with every other practical identity one may choose to take up. This provides a compelling rationale for the primacy of moral reasons. Cases in which one has non-moral reasons that conflict with moral reasons, and in which one is therefore rationally required to give up one of the practical identities that give rise to the conflicting reasons, the moral reasons (being rooted in our practical identity as human beings and which therefore cannot be given up) will always win out in the end.

However, this framework (despite its other virtues) does not yet give us any reason to think that being human has the normative import Korsgaard supposes. This is because there are other competing conceptions of what it means to be a human being available. For example, from a brute Darwinian point of view, to be a human being is fundamentally to be an effective reproducer, and that sometimes (but does not always) involve being a good reasoner. So why couldn't we say, pace Korsgaard, that we constitute ourselves as human beings by being effective reproducers?

Now enter your "updated view". Korsgaard seems to be committed to saying something even stronger than the preceding adumbration suggests. Her point seems to be that I could not be a husband unless I was a human being, and so the former practical identity is dependent upon the latter. That is, one could not have the non-moral reasons that come from being a husband unless one already had the moral reasons that come from being a human being; given that being a human being is a necessary condition for being a husband. This picture does seem to have some initial plausibility to me. Moreover, it goes some distance towards ameliorating my worry since being an effective reproducer (qua the brute Darwinian view) is not a sufficient condition for being a husband. After all, dogs and cats are effective reproducers. However, dogs and cats can never be husbands (i.e., they could never take up the kinds of normative commitments being a husband involves). Thus, we have a reason to prefer Korsgaard's characterisation of being human (qua rational actor) over the brute Darwinian view of being human (qua effective reproducer).

Perhaps this may help address your worry as well, Nick. Since being a human being (qua rational actor) is a necessary condition for being a husband, then there seems to be a sense in which the reasons derived from the latter are dependent on the former. What do you think?


Peter, you raise a very interesting question, here. I will like to suggest that the answer to your question may not be as straightforward as it may initially appear. To begin with, I am not sure if being able to have the cogito thought is sufficient for being a human being. For example, suppose I were an incredibly advanced and super-sophisticated sentient computer program existing in a virtual reality, but tricked into thinking (by my designers) that I was really a carbon-based life-form with a certain phylogenetic heritage; namely, ours. It seems as though I could, in principle, have the cogito thought, even though I would hardly qualify as a human being.

Nevertheless, let us suppose that enjoying the cogito thought is at least necessary for being a human being. But so is being a mammal. The question then becomes, why is enjoying the cogito thought, rather than being a mammal, taken to be that which makes us human? Well, one answer is that while we are not the only mammals, we are the only one's that can enjoy the cogito thought. But here's another thought experiment. Suppose we were to discover another mammal that could enjoy the cogito thought. Would that, ipso facto, make them human beings? I doubt it. (Imagine that dolphins were one day to evolve to the point of becoming as smart as us, I doubt we would then consider them human beings.)

Moreover, perhaps enjoying cogtito-thoughts is not even necessary for being human. Consider: it seems plausible that there are individuals who suffer from such extreme forms of mental retardation, that they are unable to enjoy cogito-thoughts. Would we still consider such individuals human beings? Well, I doubt too many of us would feel comfortable saying that they are not. We certainly would not be okay with them being caged, or treated like animals. Furthermore, I suspect that most non-philosophers would agree that such individuals should to be afforded the basic dignity and respect that comes with being human, a dignity and respect that they would not afford a chimp with comparable or even superior intelligence. So, what should we say about such cases. Well, it seems natural to conceive of such individuals as being defective. However, a defective human being is still a human being. Despite their defects, we still seem to think (at least pre-philosophically) that they have fulfilled whatever necessary conditions there are for being human. Consider: we would never call a really smart chimp a defective human being. This is instructive; for it suggests that there is some necessary condition for being human that the super smart chimp fails to meet that the person suffering from severe mental retardation does not.

My suggestion is this: Perhaps we are actually dealing with different conceptions of what it means to be human here. Perhaps Korsgaard has a notion of being human that is fundamentally tied to being a rational actor. One that does not include individuals with severe mental retardation, and one that perhaps includes highly evolved dolphins. But there is also our everyday pre-philosophical conception of what it means to be human; a conception that includes the severely mentally retarded and excludes highly evolved dolphins. The question then becomes, why should we buy into Korsgaard's conception of a human being when it clearly departs from our everyday pre-philosophical conception?

Now, I am not here suggesting that our pre-philosophical conception of being human (one that does not seem to require the ability to enjoy cogito-thoughts) is preferable to Korsgaard's conception (one that does). My point is simply to highlight, in response to your question, that the issue is not as straightforward as it may initially seem. There seems to be more than one competing conception of what it means to be human, and my suggestion is that it is not immediately obvious that Korsgaard's conception is the right one.


Ben, Thanks for joining the conversation! (I forgot to refresh my browser, so I did not notice your comment before posting my last two replies.) My problem is not so much with the deliberative perspective or even the Kantian move. Rather, I worry that Korsgaard, unlike Kant, may not have the resources to make the Kantian move. This is because, unlike Kant, Korsgaard takes as her starting point, being a human being, as opposed to being a rational agent. Now, she does seem to assume that these are one and the same. However, this is precisely the assumption I have been calling into question. As illustrated in my previous comment, there seems to be a pre-philosophical conception of what it means to be a human being according to which being a rational actor (at least of the kind Korsgaard seems to have in mind) is neither necessary nor sufficient for being human. This is not a problem for Kant since, if I understand him correctly, he does not ground morality in being a human, simpliciter. Rather, he grounds it in being a certain type of rational agent, who (in our case) happens to be human. In short, Korsgaard seems to begin with a controversial (and perhaps question-begging) conception of what it means to be human.

Nick said...

Hi Ben,

Yeah, thanks for clarifying, but the basic problem as I outlined it remains precisely the same. The fact of the matter is that all of my project-related reasons are contingent on my conforming to the principles that constitue my "human identity". It doesn't matter if we call that identity additive, or higher-order, or anything else we like. The fact remains that in order to pursue my projects I have to (to use Wolf's memorable phrase) "apply to a principle for permission".

I also think it's worth reflecting on the non-obviousness of many of the things you say here. Just to take a few examples:

1. "Rather, your having this practical identity commits you to valuing having reasons."

- I, for one, have never encountered myself 'valuing having reasons', and I bet you haven't either. We have no conception of what it might be like to lack reasons for action, and as such it's hard to say that we value one state of affairs over the other.

2. "If one adopts a particular identity that conflicts with an identity one already endorses, then the conflict must be resolved by shedding one or the other of these identities. "

-Now, this is just plain crazy. Two general committments may conflict, and I may simply learn to live with the conflict rather than "shed" one entirely. This is clearly not a requirement of practical rationality, though Korsgaard occasionally makes it sound as though it is.

3. "One's human identity is necessary, so it cannot be shed."

My species-membership is indeed necessary, but my reflective acceptance of the hypothetical and categorical imperatives is most certainly not. Much hangs on what the word "human" is doing, here.

4. "this is just to say that it commits you to valuing your human identity--as one who needs reasons."

-Finally: I have needs, and those needs seem to generate reasons, but saying that I "need reasons" is basically just inviting grammatical confusion. I don't accuse you of inventing these confusions, they're Kris's responsibility. But I do think it's worth reflecting on how odd they are.