an ethical dilemma for kantian ontology

cum homo sit res limitata, non alia natura illi competit quam limitatae perfectionis.

preface

in quibus animae humanae a corpore distinctio demonstratur

Kant's theory is a strongly dualistic theory—human beings are composed of two parts, and each of these parts is responsible for only some of the total complex that is a single person. in all dualistic theories, one problem stands out: how do the two parts interact with each other? in this essay, i intend to zero in on the nexus where Kant links the two parts of a human, and show that it is vulnerable to attack. i will begin by addressing three empiricist predecessors of Kant, and three corresponding puzzles that their work raises for him.

Thomas Hobbes and the will

hominis corpus machinamentum quoddam est ex ossibus, nervis, musculis, venis, sanguine et pellibus ita aptum et compostium, ut, etiamsi nulla in eo mens existeret, eosdem tamen haberet omnes motus qui nunc in eo non ab imperio voluntatis nec proinde a mente procedunt.

Hobbes is here the progenitor of all mechanistic theories of human behavior. it is not particularly interesting to ask whether the exact Hobbesian sort of calculus of desires is correct. so here i will allow the name of Hobbes to stand in for a more general mechanistic theory—that humans operate according to a mechanistic process. as if we were machines, with various parts, very complexly connected, but still able to be broken down and analyzed, piece by piece, in order to explain all human behavior as the working out of perfectly ordinary physical laws.

now Kant, with his great (perhaps overweening) respect for cause and effect in the phenomenal world, seems to admit the entire truth of this model. for example, Kant discusses the relation between the empirical self and the rational self in part three of the groundwork: [one] can consider himself...—so far as he belongs to the sensible world—to be under laws of nature (p. 108f). again, it is... necessary that everything which takes place should be infallibly determined in accordance with the laws of nature (p. 114). whether the particular calculus of passions identified by Hobbes is correct is not the point. the point is the total determination of human behavior by mechanical processes.

the following question at once arises: given this mechanistic model for all that we observe in the behavior of others and ourselves, how could people be moral beings? in order to be subject to imperatives, we must be able to make choices. note, for example, that God is not subject to imperatives: for the divine will ... there are no imperatives...i will is already of itself necessarily in harmony with the law (p.39). if there is to be an ought there must be a possibility without a necessity. but we have heard that everything which takes place ... [is] infallibly determined (above). at what point do moral predicates attach to the workings of a machine? this is a question for which Kant has an answer. i shall return to it later, after considering two further questions that should puzzle Kant.

Francis Hutcheson and the moral sense

invenio in me facultates specialibus quibusdam modis cogitandi, puta facultates imaginandi et sentiendi.

Kant believes that Hutcheson's procedure is utterly flawed, but Kant's reasoning here is particularly bad. Kant holds that moral actions must be done from no inclination whatsoever. granting this, it does not follow that the learning of this action is moral or this action is not moral must proceed from no inclination either. one is an ontological statement: the morality of an action does not arise at all from any inclinations. the other is an epistemological statement: the learning that an action is immoral must not arise from any inclinations. the first here does not imply the second.

to show this, let me paint the picture of an odd mixture of Hutcheson and Kant—a kantesonian. our kantesonian agrees completely with Kant that the morality of actions consists entirely in one's doing them for the sake of duty. no decent kantesonian would ever suggest that a action is morally worthy if it is done solely with the intent of earning the approval of others. but kantesonians happen to find the proofs of the categorical imperative lacking. for whatever reason, Kant has failed to convince them by a priori proofs that that particular ethical command is binding.

but a good kantesonian is simply filled to the brim with reverence for the law—a good kantesonian truly wants to do morally lawful actions solely motivated by duty. alas, he has not been able to make a priori any judgments of just what duty requires. but, being so motivated by duty, he looks about for the next best thing—reasonably informed judgments about duty. to be sure, this is not as good as a certain judgment—but certainly better than throwing up one's hands and abandoning one's commitment to act from duty alone!

so the ever-religious kantesonians think it likely that God would not have created people with no ability to find out what duty requires. and since strictly a priori proof does not suffice (remember that the kantesonians for some reason just don't buy Kant's proof of the categorical imperative)—since they do not have any proof apart from experience, our kantesonians conclude (reasonably, if not certainly) that God has given them some way to make judgments to find out what duty requires. perhaps they conclude that we must have a sense to find this out—in fact, suppose they conclude that their built-in (but contingent) sense of approbation is just such a thing. (it should be clear that atheistic kantesonians might quite equally agree, for the reasons that Hutcheson and Hume both give.)

so these kantesonians use their experience to make judgments about what duty requires: do this, don't do that, and so forth. but this epistemological method does not, not even a little, taint their next step—to act solely from duty. they do not conclude that different experience would somehow change what duty is, but rather that they have a decent (if possibly imperfect) faculty to discern duties from experience.

to give an analogy: suppose i enjoy reading great novels. what makes a novel great is the words on the page—great words, great novel. now i read a review from a trusted reviewer, who says read this book—it's a great novel. i go out and spend lots of money to acquire the volume. what is the motive for my action? is it the reviewer's praise? isn't it rather my appreciation of the anticipated greatness of the novel? clearly, the reviewer does not make the book great, and even if i always buy great books only when i hear a great review, i am not buying them from the review but rather from my appreciation of great novels. Perhaps there are other people who do buy books only from the review—such people are putting on a pretense—we sneer at them, because they read only because of what people will think, and not from any appreciation of great literature in itself.

so Kant has muddled these two questions: how do i find out which actions are moral? (how do i find out which books are great?), and what in the action is motivating me to do it? (am i motivated by the praise of the reviewer or the expected greatness of the prose?) in fact, contrary to Kant, it might well be that there are no a priori proofs of any ethical predicates. i am not trying here to attack the proof of the categorical imperative, but rather to point out that it is logically possible for there to be such a thing a duty, and even actions motivated solely by reverence for law, but simultaneously no a priori demonstrations of what duty requires.

Kant denies this possibility, for example, on p. 49. we shall thus have to investigate the possibility of a categorical imperative entirely a priori, since ... we do not enjoy the advantage of having its reality given to us in experience. but the preceding section does not establish that we must look for a priori grounds for categorical imperatives. all the preceding section establishes is that any imperative gained from experience might be tainted by interest. in other words, that experience is not a perfect guide. but this should not stop us from making reasonable conclusions where we have done our best to remove interest—we might not be perfect, but we still might get somewhere.

the situation is quite like that of speculative propositions about the ordinary existence of tables and books and such. i might be deceived by my senses about whether there are books around me, but that does not imply that i must examine the possibility of the existence of books entirely a priori—and not even though there is absolutely no hope of removing all doubt about sensation. but my second puzzle here is not directly to attack this point, but rather rather i bring it up to note that Kant himself allows for a distinction between how one learns ethical predicates and the ontological basis for their truth.

Kant deals with one important subjective version of this possibility—specifically, that one might be motivated by duty alone, and yet learn ethics through examples, intuition, and empathy. his entire discussion of ethical pedagogy in the critique of practical reason assumes the possibility of acting from duty alone but not ever learning any non-contingent non-empirical proofs of what duty requires. this poses a puzzle now, our second question to consider: given the supposed link between epistemological and ontological ethical motivations, how can ordinary people ever be acting from duty alone, if they do not study and learn the a priori proofs? this second puzzle is also answered by Kant, an answer we shall consider later.

David Hume and other-directed interest

praeter dolorem et voluptatem, sentiebam etiam in me famem, sitim, alisoque eiusmodi appetitus; itemque corporeas quasdam propensiones ad hilaritatem, ad tristitam, ad iram, similesque alios affectus.

David Hume argues that each person has a direct desire for the good of others. not for the sake of self-interest, but each person desires the good of others. Hume notices that some people seek the good of others even though it is against their own interest, and concludes that we must renounce the theory, which accounts for every moral sentiment by the principle of self-love....every thing which contributes to the happiness of society recommends itself directly to our approbation and good-will (section v, part ii).

Kant does not accept the existence of such a kind of desire. just as his view of empirical human nature will is the apparently more primitive concept of Hobbes, so he has a more primitive concept of desire as well. Kant thinks that all desires are inclinations, that we have an immediate interest in satisfying for our own sake. it is meaningless to him to say that we might have an interest in others.

Kant recognizes of course that we might have a desire for the good of others, but for him when we act upon that desire we are merely sating an internal hunger. we would be helping another only for the sake of ourselves, only to help us meet some inclination or sate some internal need. kant's discussion of the honest grocer on p. 9, for example, tells us that while honest, the grocer might be acting out of self-love, and that we cannot assume him to have in addition an immediate inclination towards his customers, leading him ... to give no man preference.

Kant notices that there are people who are ... of so sympathetic a temper that, without any further motive of vanity or self-interest, they find an inner pleasure in spreading happiness around them and can take delight in the contentment of others as their own work (p. 10). but note his language carefully—these people are getting pleasure and delight for their trouble. which is to say, they are meeting desires, not just acting from inclination.

Kant links inclination and desire rather too strongly here. i submit that inclination is a contingent fact of human nature which propels us to act. desire is a sort of hunger, a thing that when sated produces some characteristic happiness or at least the absence of some unhappiness. surely Kant is right that if we act only from desire, our actions have no moral worth—we are simply acting in self-interest. but it seems to me that all inclination is not necessarily such a thing.

Kant assumes that all inclinations are desires—that they are self-interested. should it not rather be meaningful to have an inclination that is not self-interested? Hume thinks that such a thing in fact exists. for this sort of inclination to be real, it must not produce a distinct happiness, but still be a motive for action. Hume seems to present exactly the right sorts of examples to justify this approach.

this line of reasoning results in the following antinomy: suppose i have an inclination to follow the law as such? perhaps all people have such an inclination, more or less strong. following such an inclination is both from inclination and from duty. but it is not the sort of ambiguity between conformity with duty and from duty—it is truly done for the sake of duty. the hypothetical contradiction fails to apply. in fact, once we are here, we might go back and revisit the separation of desire from inclination.

does it not turn out, even if all inclinations are desires, that if a person has a desire to follow the law as such, then following such a desire is ipso facto following the law for its own sake? Kant seems to have no good answer to this question. i will sharpen it, to produce my third and final puzzle: given the direct action for the good of others that the categorical imperative requires, isn't following that law, even from a pure inclination, good enough? Kant has an answer for this puzzle, which i will elaborate shortly.

Kant and the pineal gland

ex magna luce in intellectu magna consequuta est propensio in voluntate.

do I ever, in fact, make a conscious choice to follow law or reason? it might seem i do at first, every time i choose to tell the truth instead of seeking raw self-interest, i choose to follow reason. but this is too cheap an answer. suppose i choose to lie. is that a conscious choice not to follow reason? surely not—for what i would do if challenged (and what i do inside myself even if not challenge) is to trump up a justification. i assert a principle in accord of which the general rule does not apply to me in this situation. now, very frequently, my reasoning is faulty in such cases.

but i am still appealing to reason. i submit that there is no actual alternative. i submit that i never make a choice to act contrary to reason. i do, of course, sometimes act in contrary to reason or the moral law. but i am not in such an action acting from a choice at all, and when i look at my action in retrospect as a choice, i present rationalizations—which is to say, choice and reason are intimately bound up with each other, so that i cannot imagine choice happening without some kind of reason together with it.

thus, the sharpest version of this question is: what would it look like to choose contrary to reason? Kant's reply to this question is that there is a special sort of feeling, called reverence. reverence is a feeling directed towards the law, which affects our action. this feeling of reverence answers the question here, by explaining what it would look like: a choice contrary to reason is a choice that does not have reverence as a component. rationalizations are post hoc justifications that articulate what a decision might have looked like if reverence had been present, but in which it actually was not. (also, rationalizations are thus neatly separated from mere mistakes in reasoning.)

this idea of a feeling of reverence also neatly addresses the puzzles of the three questions above. first, the Hobbes-inspired puzzle asked how moral decisions could ever attach to the workings of mechanistic phenomenal human beings. the answer is that reverence is a feeling in the phenomenal self which admires the law, perceived by reason in the noumenal self. here, then, is a way for the law perceived by the choosing self to have some relation to the calculus of feelings in the empirical self.

the second puzzle is to ask how people can be said to be acting from duty alone if they have not studied and grasped the a priori nature of the categorical imperative. the answer is that the key moral action is acting from reverence. if one is acting from reverence, and is in fact following the law, then a simplification or even an error in the reason does not affect the morality of the action, which derives from the feeling of reverence, not the accuracy of the reasoning process.

and finally, the third puzzle addressed the question of what to do with the possibility of an inclination to act in accord with law as such. is not reverence just such an inclination? Kant thinks it is analogous to inclination, but not inclination proper. no matter, i would suggest that this is only an argument about words. i would prefer to call it an inclination but not a desire, Kant calls it a feeling but not an inclination. the important thing is that it is a feeling—an empirically discernible something—which may be acted from as the basis of a morally worthy act.

the notion of reverence seems to tie off all these loose ends. neatly addresses all these puzzles as it does, does it generate a puzzle of its own? well, suppose i happen to lack the feeling of reverence. am i blameworthy? suppose i do not make a choice to lack it—i've just never had it at all.

Kant's reply is that i am able to make a free choice whether or not to have reverence. that this is precisely the key ability to choose that freedom implies. this would be an acceptable (if slightly vacuous) reply, except that Kant himself provides the nail for his own coffin. why does everyone not make the correct choice to have a adequately strong feeling of reverence? (leaving aside one question: just how strong is adequate?)

different people make different choices. Kant knows this, and spends effort in a description of pedagogy about how best to train children to be moral—in fact, Kant describes a method of moral education which is to result in children having more reverence for law—more devotion to duty itself. if this is really a meaningful effort, then there must be contingent empirical reasons why some people have more or less reverence than others—and therefore that having reverence is not a free choice, but has a strong empirical component.

reverence is one feeling among others, with empirical causes. moral blame or censure cannot relate to this feeling more than any other, if Kant is to be consistent. if the choosing self is real, Kant has not adequately explained how it has anything to do with the empirical self, and thus cannot explain how the empirical self acquires any moral blame or censure. pinocchio's strings are now cut.

sources: