By In Ideas, Moral Psychology Comments (3)

Both Humeans and Kantians about Motivation are Wrong

Both Hume and Kant advocated extreme and implausible views of motivation; the same is also true of many of their contemporary followers. The truth about motivation lies in between these two extremes.

A. Hume. According to Hume (Treatise III.i.1.12):

Reason, in a strict and philosophical sense, can have an influence on our conduct only after two ways: Either when it excites a passion by informing us of the existence of something which is a proper object of it; or when it discovers the connexion of causes and effects, so as to afford us means of exerting any passion.

Hume’s examples of these two motivational processes are as follows (Treatise II.iii.3.7):

  1. The “supposition” that a fruit is “of an excellent relish” may lead one to “desire” or have a “longing” for the fruit.
  2. The “supposition” that “certain actions” are “means of obtaining” a “desired end” may lead one to “will the performance” of those actions.

Different interpretations of these two processes are possible, but the following interpretation seems to me most plausible:

  1. In the first process, a belief to the effect that one’s situation has a certain feature triggers a certain disposition to have a desire—specifically, a disposition to desire something that is related in a certain systematic way to that feature of one’s situation.
  2. In the second process, having both (a) a belief about how a certain available action will promote a certain end and (b) a desire for that end motivates one to perform the action in question.

Thus, the first process presupposes a disposition to have desires, and the second process presupposes a desire for an end. In both cases, Hume insists that there can be nothing “contrary to reason” about having or lacking any such disposition to have desires or any such desire for an end. “It is not contrary to reason to prefer … my own acknowledg’d lesser good to my greater…” (Treatise II.iii.3.6).

Hume’s theory is extreme and implausible because he does not see that some of these dispositions are in a sense rationally required. In particular, consider the disposition of being disposed to desire to perform an act whenever one believes that one ought to perform the act. Arguably, this disposition is necessarily present in any agent who is capable of having beliefs about what she ought to do, and in a non-accidental way avoids acting akratically.

To be a rational agent, one must not just avoid acting akratically by a lucky chance: one must avoid acting akratically in a non-accidental way. So, arguably, this disposition is necessarily present in every rational agent who is capable of having beliefs about what she ought to do. In that sense, lacking this disposition would indeed be “contrary to reason”.

B. Kant. According to Kant, when our motivation is moral (i.e. when we “act from duty”, which as close as we imperfect beings can come to having a “good will”), we are motivated directly by the moral law itself (as explained at the conclusion of Section 1 of the Groundwork).

The moral law, for Kant, is simply a principle of practical reason. That is, it is a way in which the will of any being who is completely rational will be motivated. Such a completely rational being will reject any non-universalizable maxim that comes before their mind, simply because it is non-universalizable; and being motivated in this way, we are told, will inevitably involve one’s adopting certain ends, such as one’s own perfection and the happiness of others (Groundwork, Section 2).

This principle is meant to be a fundamental feature of all completely rational beings, in a way that cannot be explained by the feelings or sentiments that these beings happen to have. Admittedly, when your motivation is moral in this way, certain “moral feelings” will be produced in you – such as “respect” for the law and the like (as explained in the 2nd Critique, Chapter 2). But these feelings are explanatorily downstream of what is fundamental – the primary fact that you are motivated by the moral law itself.

Kant’s theory is extreme and implausible because this idea of being motivated to adopt an end in this “pure” way, in a way that fundamentally owes nothing whatsoever to the feelings and sentiments that the agent has, seems to be a myth. Without feelings and sentiments to work on, practical reason would be empty and lacking in any direction. To use one of Kant’s own images (1st Critique, B 8), it would be like a dove that tried to fly by flapping its wings in a vacuum.

C. An intermediate view. The intermediate view that I have in mind differs from the Humeans’ view because it allows us to be motivated to act directly by our normative beliefs. However, contrary to what Kant maintains, the intermediate view implies that those normative beliefs must themselves necessarily be sensitive, in some way or other, to our feelings and sentiments.

Nonetheless, contrary to what Hume and his followers maintain, this sensitivity of our normative beliefs to our feelings and sentiments can take a holistic form. The normative beliefs of a rational agent will collectively form a body of beliefs that optimally coheres with the totality of the agent’s feelings and sentiments and other beliefs. But there is no requirement that each individual belief must be directly supported by a particular feeling or sentiment that corresponds to that belief. So there is no reason to think that whenever we are motivated by a normative belief, this belief is simply channelling the motivational force of some pre-existing passion.

It doesn’t matter whether this intermediate view is classified as “Humean” or “anti-Humean”. That is a purely terminological question. What matters is that both Kant and Hume and many of their followers had extreme and implausible views about motivation, and that the truth lies in between these extremes.

, ,

3 Responses to Both Humeans and Kantians about Motivation are Wrong

  1. Jussi Suikkanen says:

    Brief comment about the difference between your view and Hume’s and also about your objection to Hume and your reading of Hume too. It looks like you are assuming that, when our normative beliefs motivate us, it is the faculty of Reason that is responsible. But, I take it that the alternative expressivist interpretation of Hume is to think that, according to him, normative beliefs themselves are some kind of calm passions/desires. What he says about the faculty of Reason in your quote would then be compatible with your thought that normative beliefs can rationally require certain dispositions to have desires (and also certain desires themselves I assume). In this situation, Hume could rely on coherence relations between desires to account for the relevant rational requirement – it would be constitutive of rationality to have coherent set of desires and thus rationality qua coherence would require a disposition to desire in accordance to one’s normative judgments which themselves are desire-like. This reading of Hume would also be compatible with the idea that normative judgments themselves should be sensitive to feelings and sentiments.

  2. Frank Jackson says:

    About 1. Maybe the best way to read Hume’s thought is: it is Jeffrey on preference minus the math.

  3. Nice post, Ralph.

    Three questions for you, if I may.

    (1) You say “In particular, consider the disposition of being disposed to desire to perform an act whenever one believes that one ought to perform the act. Arguably, this disposition is necessarily present in any agent who is capable of having beliefs about what she ought to do, and in a non-accidental way avoids acting akratically.”

    But, why think this disposition you appeal to is “necessarily” present? Consider Leroy. Leroy is capable of having beliefs about he ought to do. In fact not only is he capable, but he in fact has lots of these beliefs throughout each of his days. He is disposed to rank his options at any given time but he is not disposed to desire to perform what is best after ranking them, that is, he is not disposed to desire to do what he is obligated to do by the tenants of his moral theory. Instead he is disposed to desire to do more than what he deems to be the worst of his options because he is instead disposed to desire to not be the worst moral agent. So maybe you could explain more as to why you think this disposition is NECESSARILY present for agents that are merely capable of having beliefs about what one ought to do?

    (2) This is a question about the nature of the intermediate view. Are you suggesting that *all* normative beliefs are necessarily sensitive to our feelings and sentiments or that *some* of these beliefs are sensitive to our feelings and sentiments? It seems like you want to endorse the former but why not settle for the latter? Maybe beliefs about what we ought to do in contexts involving innocent bystanders are necessarily sensitive to our feelings and sentiments, but, beliefs about what we ought to do in contexts involving matters of justice and deserved punishment are not *necessarily* sensitive to our feelings and sentiments because those beliefs are motivated by some just-world assumption. Maybe what I am getting at is that some beliefs seem to be immune from this sensitivity you claim is necessary when you say: “contrary to what Kant maintains, the intermediate view implies that those normative beliefs must themselves necessarily be sensitive, in some way or other, to our feelings and sentiments.”

    (3) This is a question aimed at clarification. You write, “The normative beliefs of a rational agent will collectively form a body of beliefs that optimally coheres with the totality of the agent’s feelings and sentiments and other beliefs.”

    What does it mean for a “body of beliefs” to “optimally cohere with the totality of the agent’s feelings and sentiments and other beliefs”? What beliefs are separate from our body of beliefs?How does a body of belief cohere with “the totality of an agent’s feeling and sentiments”?

    FWIW, I think that some intermediate view is probably right but the one on offer here seems a bit strong with a necessity requirement that all normative beliefs be sensitive to our feelings. It seems that all we would need for an intermediate view to get off the ground is that some of our normative beliefs are necessarily sensitive to our feelings and sentiments and an investigation as to why some are sensitive and others are not could be a fruitful way to move forward.

    I’m thinking here that normative beliefs that are grounded in strongly entrenched metaphysical commitments make one less sensitive (or not sensitive at all) to our feelings and sentiments.