Wednesday, July 20, 2005

Willing and Normativity: Part One

[Guest post by Pat Smith]

(Nota bene: What follows is a story that is unapologetically Korsgaardian, so it should not be taken as representative of Neo-Kantian constructivism. They all have their unique elements.)

How does the issue of normativity arise? It arises because we are self-conscious causalities. That is, we face the problem of morality because we have to decide what to do. If we weren’t causalities, like a rock or a chair, then there would be no problem of morality or of normativity. If we weren’t self-conscious about our causality, like a worm or a dog, then likewise the issue of normativity wouldn’t arise. The real problem of morality is that we can reflect upon our values and might decide that we aren’t justified in having them.

From the standpoint of the agent (first-person), the agent is presented with many desires (or as Kant calls them, incentives) vying for attention. But above those desires there is something that is you, something that ultimately has to endorse one of those desires in order to act. If I desire to go to a party, but also desire to study and do well on the test, I cannot simultaneously endorse (i.e. act upon) both desires. In order to act at all, I must pick the desire that I act upon, I must choose (We’ll get to the incompatibilist/ compatibilist discussions a little bit later).

To see why this must be so, let’s consider a paradigm case where it isn’t: addiction. In the most extreme forms of addiction, one desire simply takes control of you. In this case, a behavior is imposed on you to the point where it seems improper to say that you are the one acting at all. To be a self-conscious causality is to see oneself as a cause that acts upon reasons based upon desires that one reflectively endorses.

So, when we will an action we are responding to a reason, but it is a reason that results from us endorsing a particular desire. Now, if we stop here we are still far away from Kant and certainly far away from any kind of normativity. To get to Kant, we have to think on the argument in the beginning of the Section of 3 of the Groundwork.

Kant puts the reflective problem I mention above in terms of freedom. A free will is an autonomous causality. Because we are causalities, we have to operate according to laws. But we are autonomous causalities, so we cannot have an external law be imposed upon us. Nothing can determine, in advance, what the content of that law will be, so the only requirement of an autonomous will is that it will universal laws (which is redundant, laws that govern causes must be universal). Now consider the first formulation of the categorical imperative: "I ought never to act except in such a way that my maxim should be a universal law." A free will must act according to universal laws, and that is all the (first formulation) of the categorical imperative demands of us: that when we act, the only thing that determines our will must be a law.

Furthermore, consider what happens when we try to deny that our willing must be, in some important sense, universal. Well, if willing doesn’t have to be universal, then it can be one hundred percent particular. But particularistic willing is a very strange animal indeed. Particularistic willing is not "willing a different maxim" for each situation because every situation might be importantly different, thus justifying a different universal law for each situation. Nor is it endorsing whatever your strongest desire happens to be because you are still following the universal law: "I will do whatever my strongest desire is." And for similar reasons, particularistic willing cannot be something like egoism. Rather, particularistic willing must involve simply bouncing from one desire to another as soon as it seizes you, without reflection. You cannot see that desire as being representative of a general type of incentive; you must be entirely taken with the utterly particular nature of the incentive. But this is makes particularistic willing impossible. To quote Korsgaard:

"This means that the person who wills [particularistically] is at each moment identified entirely with ultimately particular incentive which he endorses. But that in turn means that particularistic willing eradicates the distinction between a person and the incentives on which he acts. And I think that means there is nothing left that is the person, the agent, whose will is distinct. He is not one person, but a series, a mere conglomeration, of unrelatedly causally effective impulses. There is no difference between someone who has a particularistic will and someone who has no will at all… (My emphasis)."

Remember our reflective structure: above our incentives there is something that is us. But if we will particularistically, then that part disappears leaving us unable to will anything at all. So particularistic willing is impossible.

Okay, let’s take stock at this point. The reflective structure of ourselves as self-conscious causalities gives rise to the requirement (both through Korsgaard’s and Kant’s arguments) that our willing use universal laws. And since this is the sole requirement of the categorical imperative, we can conclude that the categorical imperative must be our law. The categorical imperative is at least part of the constitutive standards of willing.

But this isn’t the end of the story because we need to demonstrate that the range of the categorical imperative for a free will (i.e. self conscious causality) is that of all rational beings. But that is the subject for the next post.


  1. I'm curious about the place of cognitive and neuroscience in the arguments you're making. They don't seem to play any role at all, which makes sense, as this is a philosophy blog, but the arguments you're making are arguments that rely heavily on ideas about how people think. Cognitive and neuroscience evidence isn't necessary to support these arguments, obviously. But in the presence of direct and evolving evidence in these sciences about the mechanics and processes of how people think, it seems important to place this body of knowledge in interaction with the arguments you're making. For example, if cognitive science shows that brains simply don't process the types of decisions and stimuli you're talking about in ways consistent with your reasoning, that would create a problem. More plausibly, understanding how emotions, reasoning, etc play out physically in the brain in decision making can help refine, strengthen, and clarify one's thinking about certain points, for example yesterday's commenter question about normative adequacy, or the contrast between universal and particularistic willing.

  2. I'm still a bit unclear on exactly what particularistic willing is. Is the following an example?

    Suppose I'm at a party and I see an interesting dessert sitting on a big plate. I have a strong desire to eat that dessert. I don't have any occurrent mental states of the form, "People at parties ought to eat desserts like that" or "I ought to satisfy my hunger whenever possible." My inventory of occurrent mental states just includes a desire to consume that dessert and a belief that by going there and putting it into my mouth I can eat it. Perhaps I also have dispositions to eat desserts of that kind, or to eat stuff in general, but my desire at the moment is only for that particular instance of dessert.

  3. Jumping the gun a little here, and assuming you accept that my example is a case of particularistic willing, I'd want to make a reductive move where Korsgaard makes an eliminative one:

    He is not one person, but a series, a mere conglomeration, of unrelatedly causally effective impulses.

    Why can't we reduce a person to this conglomeration of impulses (or something including this conglomeration of impulses and other stuff)? Perhaps what you've got is a fickle or complicated person, but I think there's still a person here.

  4. To Neil:
    I don't think it is an example of particularistic willing. Or at least, not necessarily so. I still think there is the further thought: "my wanting the cake is a reason for me to be move over there and try to eat it." In which case, it isn't particularistic willing.

    To anonymous: This account is separate from deterministic issues. Even if neuroscience definitely showed that this kind of deliberation is reducible to physical states (how would they show that?), it still wouldn't solve he first personal problem of having to decide what to do.

    I think my account is completely separate from neuroscientific issues at least until we get to the issue of moral/political responsibility.

  5. Pat, you don't think that it's necessarily a feature of cases like this that we have that further thought, do you? It seems to me that in some cases we don't have that extra thought.

    If you think it's possible to act without the additional thought, consider a case in which we do. Is that particularistic willing?

  6. I think it is certainly possible for an incentive to present itself without the further thought.

    What I don't think is possible is for a person to act on that incentive without the further thought. THis is because action, as opposed to reflexive or coerced movements, is a movement that is originated or started by the agent.

    So I think what is being confused is the unmediated and immediate presentation of the incentive as a possible reason for acting and the eventual acting after that incentive is endorsed.

  7. action, as opposed to reflexive or coerced movements, is a movement that is originated or started by the agent

    I agree with this. However, I'd say that having a desire and a belief that causally interact (in the right way) to generate movement is a way for an agent to originate or start an action. If you have a view on which selves are reducible, at least in part, to their desires, you can count this as action while holding that agents must originate actions. No further reflective endorsement is required.

    Judging by the phenomenology of action, I'd want to say that this is what happens most of the time when I act. Usually, I don't think anything like, "my wanting x is a reason for me to do y" -- I just desire x, form the belief that doing y will bring about x, and bam! I'm doing y.

  8. I think I can mostly grant your picture. But even so, it is implicit to your action. Look at the second step:

    "I form the belief that doing y will bring about x." This belief formation is where the universalistic requirement comes in.

    So, that isn't particularistic willing. It involves the creation of a hypothetical imperative, which equally involves a certain amount of categorization. Let me see if I can find the appropriate quotation from it is:

    "Particularistic willing is not just a matter of willing a different maxim for each different occasion. Instead, particularistic willing is a matter of willing a maxim for exactly this occasion without taking it to have ANY implications for ANY kind for any other occasions, whether they are like this one or not. You will a maxim thinking that you can use it just this once and then so to speak disacard it; you don't even need a reason to change your mind."

    I would wager that the belief (maxim) you form is not of this form.

    Now, this process happens essentially instantly in many cases where there is no opposition.

    But imagine that you are choosing between conflicting desires A and B. Then you can see how the Korsgaardian picture operates.

  9. And I still maintain Neil, that when you decide to walk over to the table to get the are deciding to do that.

    You could stay here, you could go to the bathroom, or get the crabcakes. But you want to eat the cake and you take that as good enough reason not to go some place else.

    This is more obvious when desires are in conflict, but it still seems that is still happening even when things are not obviously conflicting.

  10. Can you explain how a universalistic requirement comes in when I form the belief? I thought the formation of this belief was merely a normal instance of belief-formation about some causal connection, no different from the formation of a belief that fire will cause smoke. Do universalistic requirements come in whenever I form beliefs about causal connections?

    I'm also wondering what work the hypothetical imperative does here. It seems to me that once one buys a view where belief-desire pairs are sufficient to cause action, all the explanatory work is done and we have no further need of imperatives. Is hypothetical imperative formation just another way of talking about something in that picture? (The causal belief, which would otherwise be the most likely candidate, doesn't seem to have imperative form.) Or is there something else that the belief-desire picture doesn't account for?

    Is the strangeness of particularistic willing just a result of the fact that 'x' and 'y' in most of our practical deliberations are broad enough to cover multiple possible situations? (I'm guessing that this isn't what you mean.)

    In response to the stuff about deciding -- I agree with you, if 'deciding' is given a weak enough reading. If 'deciding' means 'deliberating', however, I'd have to argue that I can act without deliberation.

  11. I guess what's confusing me about particularistic willing is that I don't usually think about my mental states as I generate them, so I don't think of them as 'having implications for further situations' or 'having no implications for further situations.' I just have the mental states themselves. As a result, I'm trying to understand Korsgaard as talking about the nature of the maxims (or whatever these mental states are) themselves, rather than our behavior in taking sidelong looks at our own mental processes.

  12. Part of the problem here is that you think of action as being a production.

    You take a desire, you smush it together with a belief about that desire, and then you act.

    I think this is most plausible when we are talking about an instinct. A desire for food or water directly often appears unmediated in the way you describe if you are really, really hungry or thirsty.

    In fact, when we act that way (like seeing some food when starving) and we directly, without any forethought to go eat it. But that looks a lot like how non-human animals behave. I think at that point it is very plausible to say that we aren't acting as self-conscious wills at all. Like addiction, it overcomes our reflective natures.

    But is that how we ordinarily act? Absolutely not. Even at a party you describe, it simply does not strike me as plausible that what you are thinking is something along these lines:

    "At time T ONLY and this specific motivation M ONLY, I will go to that Table T ONLY, I will eat this Cake C
    And this has no consequences to how I think about other situations that I might face, no matter how similar."

    I just don't think that's the belief you form. And this is what particularistic willing would involve.

    I think the belief you form is, or is implicitly based on (maybe you make a heuristic shortcut) based on a general principle.


    Not only that, but I think there is a gap in the explanation based on our reflective structures.

    It seems to me that the following is incomplete:

    1) I want cake
    2) I believe that I can get the cake by walking over to the table and eating it.
    3) I act.

    I think there simply has to be the idea in there: "That my hunger is a sufficient reason for me to act." Or something similar.

    But let us suppose that the unmediated model makes a certain degree of sense when it comes to those who are really hungry. I can't imagine it makes a great deal of sense with any more complex action

    But one more thing.

    You say in your belief you perceive a causal connection.

    But causal connections are necessarily a percetpion of a universal law, so I think there is a way where you can't will particularistically in that sense.

    See, even in your account I think you are still seeing yourself as the cause of the action, as engaging in a causal linkage that you originate?

    That is what I mean by "originating" or "deciding" to act. You are seeing yourself as the cause of a certain set of events.

    I hope this helps somewhat.

  13. "At time T ONLY and this specific motivation M ONLY, I will go to that Table T ONLY, I will eat this Cake C
    And this has no consequences to how I think about other situations that I might face, no matter how similar."

    Actually, this does seem more or less right to me. I don't explicitly think out the "ONLYs", but I certainly don't see myself as making a decision that has implications for future situations. I'm just thinking about that piece of cake, not any future pieces of cake. No future situations or pieces of cake cross my mind. While I agree with you that I don't think to exclude future cakes from consideration, I don't think to include them either, so I think the natural thing to say here is that they aren't included.

    This isn't limited to instinct. When I'm editing a paper and I see a misspelling, the phenomenology is similar. There is no thought of future situations and misspellings. Certainly, I don't think to exclude future situations, but all I'm concerned with at the time is the one misspelling. I doubt you would say that correcting misspellings is instinctual.

    Now, it is certainly true here that I have a general aversion to misspellings. Most of my desires are, similarly, about a broad class of things, and not particular entities. But I don't think this is Korsgaard's point.

    I don't really see why the "sufficient reason for me to act" premise is necessary. But while we're talking about phenomenology, let's bear in mind that the agent doesn't think about his desire for cake. He just desires the cake and goes "mmm, Cake". (Or he is averse to the misspelling and goes, "ack! Misspelling!") As Pettit and Smith argue, the desire is "backgrounded" rather than sitting in the foreground of deliberation. The agent thinks about the goodness or the badness of the entity, and he is caused to do this by his desires and aversions. So from the agent's point of view, it goes like this:

    1. Ack! There is a misspelling!
    2. If I change that word, there will be no misspelling.
    Conclusion: change the word.

  14. I think at this point we are simply arguing about the phenomenology: how acting appears to me versus how it appears to you.

    I don't think the mispellings example works for you because there has to be included, at least implicitly, that mispellings are bad, embarassing, and to be avoided.

    Or else, why are you acting?

    Analogously, you must take your hunger as a reason to act, or else, why are you acting?

    Now, I don't think I am commited to saying that this elaborate deliberative process happens every time.

    I think we make heuristic shortcuts where we don't go through an elaborate deliberative procedure.

    I think it is because we make universal claims that these immediate claims work.

    We make the universal claim that typos in academic papers are bad, so when we see one, we can act immediately to take it away.

    We make the general universal claim that we can eat food at parties, so we don't have to think about it when we see cake at the party.

    But, we run across things that do bring this deliberative structure.

    Suppose you are going for the cake and you see it is the last piece and someone else is reaching for it. Or a close friend has just come into the party.
    I don't think your account can explain normative failure or a failure to act.

    For your account, whatever our strongest desire is, combined with a belief, leads us to act. There is no sense in which we ought to do something or act differently. No normativity.

  15. Further, I think your description of an action is really action at all.

    In your account, we are only agents in the sense that our bodies and minds are the locations of external influences.

    But if we are simply a passive spectator to the operation of desires, flinging us this way and that, to that typo or that cake, then we aren't really AGENTS at all.

    And if we aren't agents, it is hard to say we are engaged in action.


Visitors: check my comments policy first.
Non-Blogger users: If the comment form isn't working for you, email me your comment and I can post it on your behalf. (If your comment is too long, first try breaking it into two parts.)