<<

. 5
( 8)



>>

Categorical Imperative as the supreme principle of morality and, perhaps,
has made an error in applying it. But suppose a person, call him Stram,
has not embraced this principle. After years of long, careful, and strenuous
re¬‚ection, Stram has concluded that a version of Act Utilitarianism is the
valid moral doctrine. According to him, the supreme principle of morality is:
always do what you believe will maximize the pleasure and minimize the pain
of all sentient beings. Stram does his very best to live by this principle. He has
internalized it to such a degree that at times he is surprised to ¬nd himself
calculating the effects on sentient beings of even seemingly trivial deeds.
At one point, he ¬nds himself in a situation where he believes that lying
is demanded by the combination of his circumstances and the utilitarian
principle. After thinking through the alternatives, he decides that he must,
in his position as an accounting consultant, lie to a politician about the
county™s ability to raise funds for a proposed dam. Only by lying, he judges,
can he insure that the dam will not be built, wildlife decimated, and three
whole towns destroyed. Because he believes it to be the right thing to do,
Stram goes ahead and lies to the politician.
The key question here is not whether Stram™s action is by Kant™s standard
morally permissible; let us just assume that it is not “ that he did it on a
maxim that fails the Categorical Imperative test. The central issue is, rather,
whether the moral impermissibility of Stram™s action would, as Kant suggests,
Search for the Supreme Principle of Morality
124

be a sure sign that it was lacking in moral worth. What if, as a sincere Act
Utilitarian, Stram has used all of his powers in a struggle to determine what is
right? What if, in Kant™s terms, he did not merely wish but willed to discover
the correct course of action? In this case, it appears that his failure might
have stemmed not from his inclinations but rather from factors beyond his
control: factors that should not, in the spirit of Kant™s own view, matter in a
determination of whether his action could have had moral worth.17
Notice that the cases of Colonel Mikavitch and Stram illustrate two ways
an agent™s action might fail to be morally permissible, yet have moral worth.
In the ¬rst case, an agent has (in Kant™s view) embraced the correct moral
standard, but has failed to apply it accurately. This case illustrates an error
in principle application. In the second case, an agent has (in Kant™s view)
embraced an incorrect moral standard and applied it correctly. This case
illustrates an error in principle choice.18
Of course, examples like that of Stram and Colonel Mikavitch have an arti-
¬cial ring. There may not be many colonels or accounting consultants who
have explicitly embraced the Kantian, Act Utilitarian, or, for that matter,
any one “supreme principle of morality.” Nevertheless, there are, I ven-
ture, plenty of people who throughout their lives have done their best to
determine what is right but, measured by the standard of the Categorical
Imperative, have failed. In the determination of an action™s possible moral
worth, Kant discounts the effects an agent™s action actually has in the world,
apparently on the grounds that these effects are beyond her control.
Nevertheless, at least in the Groundwork, he does not discount mistaken
moral judgments, even though, when an agent makes her best effort to get
it right, these also seem to be beyond her control. It is this questionable
asymmetry that the cases of Stram and the colonel were designed to bring
into focus.
This asymmetry in Kant™s view stems from his limitation of actions done
from duty to ones that, by the standard of the Categorical Imperative, are
morally permissible. I suggest that we reject this limitation “ that we acknowl-
edge that a morally impermissible action can be done from duty and thus
can have moral worth.


6.7 Moral Permissibility and Moral Worth in the Metaphysics of Morals
In fairness to Kant, we should note that in the Metaphysics of Morals he moves
toward, though he does not explicitly embrace, the possibility of morally
impermissible actions having moral worth. Evidence that he makes this move
emerges in his discussion of conscience.19 For Kant, conscience is practical
reason™s capacity to hold a person™s duty before him, judge him on whether
he has abided by it, and even punish him for his failures to do so. Conscience
is an internal court where a person™s practical reason, in its capacity as judge,
Duty and Moral Worth 125

renders a verdict on her deeds. Practical reason renders this verdict on the
basis of “the law,” namely the Categorical Imperative (MS 438). According
to Kant, everyone has a conscience. When we refer to someone™s having
none, what we mean (or should mean) is that this person never heeds his
conscience (MS 400). Kant goes on to say:

[W]hile I can indeed be mistaken at times in my objective judgment as to whether
something is a duty or not, I cannot be mistaken in my subjective judgment as to
whether I have submitted it to my practical reason (here in its role as judge) for
such a judgment. . . .[I]f someone is aware that he has acted in accordance with his
conscience, then as far as guilt or innocence is concerned nothing more can be
required of him. It is incumbent on him only to enlighten his understanding in the
matter of what is or is not duty. (MS 401)

Here Kant makes an acknowledgment that, in my view, would have been
welcome in the Groundwork: without being led astray by his inclinations, an
agent can make an error in determining what his duty is. Kant does not tell
us explicitly how such an error occurs. It is evident, nevertheless, that the
mistake gets made at the level of applying the fundamental standard of moral
judgment (i.e., “the law”) rather than at the level of determining what this
standard might be. For Kant, the law on the basis of which each person™s
conscience reaches its verdict just is the Categorical Imperative.
Not only does Kant here acknowledge that an agent™s acting contrary to
duty might stem simply from an error in principle application, but he also
suggests that when it does, the agent is not morally blamable. When an agent
is heeding his conscience, that is, doing what he believes the Categorical
Imperative to prescribe, he is not morally at fault for acting contrary to
duty. In the Groundwork, Kant makes no such statement, nor is it clear that
he would be amenable to it. As we have seen, Kant is there at pains to
emphasize the ease with which each agent can determine what he (morally)
ought to do. This passage from the Metaphysics of Morals indicates a change
in Kant™s tone, and it seems to mark a shift in his doctrine as well.
But how great a shift? That a conscience-abiding agent incurs no moral
guilt in performing an undutiful action does not in itself entail that this
action can have moral worth. Kant neither states nor plainly implies that an
action done contrary to duty can have moral worth. Nevertheless, at least
on a charitable interpretation, he seems to be leaning in this direction. It
is hard to see what plausible grounds Kant could offer for granting that a
person is not morally blamable for a conscience-abiding, undutiful action,
yet denying that such an action could have moral worth. It is one thing to hold,
as Kant might in the Groundwork, that we are always morally accountable
for acting in a morally impermissible way and, on this basis, to conclude
that morally impermissible actions cannot be from duty and thus cannot
have moral worth. It would be quite another to acknowledge that we are
Search for the Supreme Principle of Morality
126

sometimes not at all accountable for acting contrary to duty, yet to cling
to a view that, in effect, rules out the possibility of an action contrary to
duty having moral worth. Recall that in his discussion of moral worth in the
Groundwork Kant appeals to our intuition that an agent™s action is not to be
disquali¬ed from having moral worth by anything we take to be outside of
his control. By claiming in the Metaphysics of Morals that a person is not always
morally blamable for acting contrary to duty, he is, in effect, admitting that
whether an agent acts contrary to duty can be determined by factors outside
of his control “ for example, how adept he is at applying the Categorical
Imperative. But now suppose that an individual not only acts in accordance
with his conscience but does so for its own sake. He obeys his conscience
simply because it is the right thing to do. It seems clear not only that this
person might end up acting contrary to duty but that her doing so could
stem from conditions over which she has no real power. In this case, the very
intuition to which Kant appeals in the Groundwork would direct him not to
disqualify the agent™s action from having moral worth. By acknowledging
that when he abides by his conscience, an agent can blamelessly violate his
duty, Kant sets himself on a path toward the view that morally impermissible
actions can have moral worth.
This discussion does not, however, allow us to conclude that the Kant of
the Metaphysics of Morals would agree entirely with my description of cases
where moral worth is at issue. I suspect that he would concur that even if
Colonel Mikavitch acted contrary to duty, her action could have moral worth.
According to the example, she had the Categorical Imperative in view, and
if she acted contrary to it, it was owing solely to a mistake in applying it.
But Stram is a different matter. Recall that Kant conceives of conscience as
an internal court where a judge renders a verdict on each person™s deeds
on the basis of “the law” “ that is, the Categorical Imperative. According
to my description, however, the judge presiding over Stram™s internal court
seems to have based his decisions not on the Categorical Imperative but
rather on a principle of utility. In the spirit of his Groundwork contention
that ordinary human reason always has the Categorical Imperative in view, I
believe that Kant would reject this description. He would, I think, insist that
in cases where an agent purposefully acts in accordance with a rival practical
principle, but contrary to Kantian duty, he has failed to heed his conscience.
And Kant, of course, would not excuse such a lapse. On his view, conscience
simply is the court of the Categorical Imperative.
Kant moves toward recognizing the moral worth of some morally imper-
missible actions, namely those which stem from errors in applying the Cate-
gorical Imperative. Nevertheless, he remains steadfast in his denial of moral
worth to actions whose moral impermissibility would seem to stem from
errors in choosing a moral standard.20 I believe this denial to be contrary
to ordinary (and better) moral judgment. Actions stemming from moral
Duty and Moral Worth 127

standards other than Kant™s can have moral worth, and can thus express
what, intuitively speaking, we might call a good will.


6.8 The (Alleged) Transparency of Moral Requirements
Let me now turn to two objections to my claim that Kant should acknowl-
edge that some actions contrary to duty can be done from duty, and thus
that among actions that have moral worth, we might ¬nd some that clash
with moral requirements. First, whatever intuitive force the claim has de-
rives largely from the following notion: however hard an agent tries to do
what is right, she might actually end up doing something that con¬‚icts with
Kantian duty. But as we reminded ourselves, Kant rejects this notion. In the
Groundwork, he says:

[W]e have arrived, within the moral cognition of common human reason, at its
principle, which it admittedly does not think so abstractly in a universal form but
which it actually has always before its eyes and uses as the norm for its appraisals.
Here it would be easy to show how common human reason, with this compass in
hand, knows very well how to distinguish in every case that comes up what is good
and what is evil, what is in conformity with duty or contrary to duty, if, without in the
least teaching it anything new, we only, as did Socrates, make it attentive to its own
principle; and that there is, accordingly, no need of science and philosophy to know
what one has to do in order to be honest and good, and even wise and virtuous. We
might even have assumed in advance that cognizance of what it is incumbent upon
everyone to do, and so also to know, would be the affair of every human being, even
the most common. (GMS 403“404)

Here Kant implies that he would deny the possibility of an agent™s trying
her best (without succumbing to inclination) to do what is right, yet erring
either in choice or application of moral standard. Of course, the principle
of “the moral cognition of common human reason” to which Kant refers is
the Categorical Imperative. Ordinary reason, he believes, has (a version of)
the Categorical Imperative always in view. In light of this belief, it is hard
to see how, for Kant, a completely sincere and dedicated inquirer could
embrace any moral standard other than this imperative. Kant also here
seems to reject the idea that someone who had embraced the Categorical
Imperative could, his best efforts notwithstanding, misapply this principle.
With the compass of the Categorical Imperative in hand, says Kant, ordinary
reason “knows very well how to distinguish in every case that comes up . . .
what is in conformity with duty or contrary to duty.”
In response, suppose for a moment Kant is right in claiming that ordinary
reason uses the Categorical Imperative (perhaps in a folksy form) as the
standard of moral judgment. It is, nevertheless, a strain to deny that even
with the best of intention and effort, we might fail to apply this standard
Search for the Supreme Principle of Morality
128

correctly. After all, haven™t we behind us two hundred years of scholarly
disagreement on how to employ the Categorical Imperative test?21 Second, if
ordinary reason always employed the Categorical Imperative as the standard
of moral judgment, then Kant might have grounds for insisting that anyone
who sincerely tried to determine what was right would not embrace or act
on any other principle. But it is easy to be skeptical as to whether ordinary
reason does employ exclusively something like the Categorical Imperative as
the standard for moral judgment. Granted, in contemplating whether to do
something, we sometimes ask ourselves the roughly Kantian question, “What
if everyone did that?” But we also sometimes pose the roughly utilitarian one:
“If I did this, how would it affect the well-being of those I care about?” The
passage in question does little to undermine the possibility that a sincere and
strong-willed moral inquirer might, either by misapplying the Categorical
Imperative or by correctly applying some other principle, act in a way that
con¬‚icts with Kantian duty.
At this point one might object that I have not really focused on the crux
of Kant™s notion that, unless he is swayed by his inclinations, an agent would
not embrace a moral standard other than the Categorical Imperative. Kant
holds that this imperative is valid (unconditionally binding on all of us)
and that it has its source in reason alone.22 If he is right, goes the objec-
tion, then the Categorical Imperative would obviously present itself as the
standard of moral judgment to every being who possesses reason. Every such
being would legislate this imperative to herself by virtue of the very cognitive
equipment she possessed. She couldn™t help but embrace it as the standard
of moral judgment.
In this objection we ¬nd the beginnings of an explanation of Kant™s view
that no agent could in a sincere quest to discover his duty embrace a moral
standard other than the Categorical Imperative. However, we ¬nd no gen-
uine justi¬cation of the view. Granted, if the Categorical Imperative were
valid and had its source in reason, we might have license to conclude that it
would be recognized by all of us as the standard of moral judgment.23 (I say
“might” because there seems to be no guarantee that reason is transparent
in the requisite sense. That an agent is obligated by her own reason to obey
the Categorical Imperative would not in itself entail that she would realize
that she is. An agent could conceivably fail to discern what her own rea-
son demands. Why should the transparency of practical reason be taken for
granted?) At any rate, the truth of the claim that if the Categorical Imper-
ative were valid and had its source in reason, it would be recognized by all
sincere inquirers as the standard of moral judgment fails to justify the view
that, actually, all sincere inquirers do recognize it as this standard. In the
second Critique, Kant himself seems to acknowledge that he does not prove
the Categorical Imperative to be valid and to have its source in reason (see
section I.3). Moreover, consider a parallel. Suppose that the following claim
is true. If an Act Utilitarian principle were valid and had its source in human
Duty and Moral Worth 129

reason and intuition, it would be recognized by all sincere inquirers as the
standard of moral judgment. The truth of this claim would fail to justify
the view that, actually, all sincere inquirers do recognize an Act Utilitarian
principle as this standard. What we might be able to conclude if we had a
proof of the validity and origins of a moral principle has little relevance to
what we have grounds to believe now, in the absence of such a proof.


6.9 Odious Actions and Moral Worth
Let me now turn to a second objection to my suggestion that, even if the Cat-
egorical Imperative is indeed the supreme principle of morality, the moral
worth of an action should not turn on its Kantian moral permissibility. If we
conclude that Colonel Mikavitch™s and Stram™s actions have moral worth,
goes the objection, then we are rationally compelled to admit that any action
that could be done from duty can have such worth. But there are actions
done from duty that are so odious that we are unwilling to grant them any
moral value.24 In Eichmann in Jerusalem, Hannah Arendt writes of the Cate-
gorical Imperative in the Third Reich, a principle that was apparently known
to some Nazis: “Act in such a way that the Fuhrer, if he knew your action,
¨
25
would approve it.” If we remove moral permissibility as a condition for
moral worth, then we are forced to conclude that an agent™s acting from duty
on an imperative such as this would have moral worth. Kant™s Groundwork
account enables us to avoid this unwelcome and disturbing conclusion.
In response, if I am correct in ¬nding an untenable asymmetry in Kant™s
view of what affects an action™s moral worth, then Kant™s account does not
really give us any philosophically plausible means for avoiding this conclu-
sion at all. However, the question is whether Kant™s account, revised in the
way I have suggested, can meet the objection. To a large extent, I think
it can.
The key to seeing this is to understand what in Kant™s view it means to
act from duty. In the preceding chapter (5.3) we found that an action is
done from duty only if two conditions are met. First, the agent™s incentive
for acting must stem from the notion that the principle (represented by
the agent as a law) requires the action. Second, the agent™s notion that the
action is morally required must itself provide suf¬cient incentive for him to
perform it. In sum, the agent™s (in itself suf¬cient) incentive for acting must
stem from the notion that a principle (represented by the agent as a law)
requires the action.
The discussion in this chapter enables us, I believe, to make explicit two
further conditions on acting from duty. First, as we found in section 6.6, an
agent must do his best to realize the end of his action. Expressing a good
will through acting from duty involves “the summoning of all the means in
our power” to realize our aim. If an agent holds breaking promises to be
forbidden by a principle she represents as a law, she would not count as
Search for the Supreme Principle of Morality
130

willing, from duty, to keep a promise unless she made every (in her view
morally permissible) effort to do so. Kant ¬nds this condition in ordinary
moral reason.
I think we also ¬nd in ordinary moral reason an additional condition
on acting from duty, namely that an agent must make a genuine effort to
determine what her duty is. At least part of why we think an agent™s making
a halfhearted attempt to attain her end to be inconsistent with its being
from duty (and thus having moral worth) is that her action betrays a lack
of commitment to doing what is morally required in the case at hand. An
agent™s failure to make a genuine effort to determine just what her moral
duty is betrays a similar lack of commitment. If someone is really interested
in doing what is morally required, then she must take an active interest in
¬nding out just what is morally required. She need not delve into casuistry
before every action, but she needs to act against the background of re¬‚ection
on the moral status of her action, that is, act against the background of what
I call conscientious re¬‚ection. Since, I venture, it belongs to our everyday
concept of an action done from duty that it express a commitment to doing
what is morally required (at least in the case at hand), we hold that no action
that fails to express such a commitment can be done from duty. We do not
allow factors that, intuitively speaking, we hold to be beyond an agent™s
control to preclude her action from having moral worth. Factors within her
control, however, are a different matter. And among these factors we ¬nd
not only the agent™s effort to realize the ends of morally required actions
but also her effort to determine just which actions these are. In the spirit
of Kant™s view, if an action ful¬lls each of the four conditions we have just
sketched, then it has been done from duty and thus has moral worth.
Returning to the objection, I doubt very much whether someone acting
in accordance with the Nazi perversion of the Categorical Imperative would
ful¬ll all four conditions we have isolated for acting from duty. In particu-
lar, I ¬nd it far more likely that slovenliness, rather than sincere effort at
re¬‚ection, would result in a person™s embracing this principle. Moreover, I
doubt very much that the agent™s (in itself suf¬cient) incentive for acting
would really lie in the notion that this principle, represented by him as a
law, required the action. It is, I think, much more likely that greed or am-
bition would constitute the grounds of his action. In the case of Eichmann,
these doubts seem to be con¬rmed. However, I cannot prove it to be im-
possible that in performing an odious action, someone might ful¬ll each of
the conditions in question, thereby giving his action moral worth. Acknowl-
edging the possibility of odious actions having moral worth is painful. Yet
I see no way of avoiding it while, at the same time, defending a coherent
reconstruction of Kant™s views.
At this point, someone might object that I have overlooked a very Kantian
way of avoiding this disturbing conclusion. In the Religion within the Limits
of Reason Alone, Kant considers the case of a religious inquisitor “who clings
Duty and Moral Worth 131

fast to the uniqueness of his statutory faith even to the point of [imposing]
martyrdom, and who has to pass judgment upon a so-called heretic
(otherwise a good citizen) charged with unbelief” (Rel 186, English ed.
174). Suppose that the inquisitor condemned the heretic to death and that
this action was, by Kant™s standard, morally impermissible. Here, it might
seem, we have a case in which a morally impermissible action could have
moral worth. For it appears that the inquisitor™s action might meet each of
the Kantian conditions we have discussed: he might have arrived through
sincere moral re¬‚ection at his action of condemning the heretic, he might
have had as his incentive for condemning him the notion that doing so was
required by a universally binding principle, and so forth. In short we seem
to have just the sort of case the objection worries about “ an odious action
that (if my view is correct) we must acknowledge to have moral worth.
One might think that Kant™s own reaction in the Religion to a case such as
this provides us with a way of avoiding this acknowledgment. Kant suggests
that the inquisitor™s action could not meet these conditions. In Kant™s view,
sincere moral re¬‚ection leads an agent to the view that he must be sure
(gewiß ) that an action he proposes to perform is right before he performs
it. Kant calls this view a “postulate of conscience” (Postulat des Gewissens ; Rel
186, English ed. 174). However, the inquisitor cannot, Kant argues, sincerely
reach the conclusion that he is sure of the rightness of his condemnation.
Apparently, Kant holds that earnest re¬‚ection would lead the inquisitor to
the conclusion that it is wrong, based on a man™s religious faith, to deprive
him of his life “ unless the divine will has ordered it (Rel 186“187, English
ed. 175). If the inquisitor believed that the divine will had indeed made such
an order, his belief would be based either on what he took to be his personal
communication with God or on divine doctrine revealed to someone else.
In either case, Kant argues, the inquisitor could not sincerely come to the
conclusion that he was sure that the condemnation was ordered by God
and thus right.26 Therefore, the inquisitor™s condemnation of the heretic to
death could not, in Kant™s view, be the result of sincere moral re¬‚ection, nor,
it seems, could the inquisitor be motivated by the notion that his action was
morally required. If Kant™s views regarding this “postulate of conscience”
are correct, then we can say that, appearances to the contrary, the inquisitor
is really not acting from duty.
Although I believe that actions like the inquisitor™s would almost always
fail to meet the Kantian conditions for moral worth, I am not convinced
that Kant™s considerations here prove that they could never meet them. First,
although I think it highly unlikely that the inquisitor would sincerely con-
clude that he was sure of the condemnation™s rightness, I do not think it
to be impossible that he would. We cannot totally discount the possibility
that, even after earnest re¬‚ection, he takes it to be certain that the condem-
nation was commanded by God. Second, I do not believe it to be obvious
that all sincere, morally re¬‚ective agents would embrace Kant™s postulate
Search for the Supreme Principle of Morality
132

of conscience. Serious moral re¬‚ection might lead an agent to the view
that, ideally, one would be sure that an action is right before he did it, but
that, sometimes, given the complexity of the moral landscape, one is forced to
choose between actions none of which one holds with certainty to be right.27
Appeal to Kant™s “postulate of conscience” would not enable us to avoid the
painful conclusion that being faithful to the spirit of Kantianism “ and, I
believe, of ordinary views regarding moral worth “ requires us to admit the
possibility (though by no means the likelihood) that some terribly wrong
actions have moral worth.


6.10 Sympathy and Moral Worth
Against Kant™s of¬cial view (at the very least his view in the Groundwork), I
have argued that some actions not in accordance with duty can be done from
duty. The logic of Kant™s own position, I have contended, requires him to
acknowledge this. Kant™s claim that all actions from duty have moral worth
must in the end be understood to allow that some morally impermissible
actions can have such worth. I have tried to defend Kant™s claim understood
in this way.
But there is a far more familiar criticism of Kant™s views regarding moral
worth that warrants attention. Kant claims not only that all actions from
duty have such worth but that only such actions have it. The criticism is that
actions from other motives, typically from sympathy, compassion, and the
like, have moral worth. A full treatment of the moral worth (or lack thereof)
of acting from such motives is beyond the scope of my project. However, as I
try to explain regarding the motive of sympathy, some speci¬cations of this
objection seem to have force, whereas others do not.
To begin, we need a rough idea of what critics of Kant mean by acting
from sympathy. On one critic™s account, namely that of Lawrence Blum,
acting from sympathy amounts to acting from an emotion that has three
elements.28 First, it has a cognitive element. Sympathy is intentional; it is
directed at another™s weal or woe. If an agent has sympathy for another,
then he believes that she is in a certain state (e.g., one of suffering). Second
(and obviously), sympathy involves feeling. To have sympathy for a person,
an agent must at least sometimes be in a certain affective state regarding
her (e.g., pained at her suffering). Third, sympathy has a conative element.
If an agent has sympathy for another, then he wants to help the person for
her own sake. To act from sympathy is to act from this emotion. It typically
involves thinking that another is suffering, feeling distress at this suffering,
and trying to help the other for her own sake. Acting from sympathy alone
does not involve any re¬‚ection on the moral status “ for example, the moral
permissibility or even virtuousness “ of one™s action.
In his well-known Groundwork I discussion of the sympathetically attuned
person, Kant does not offer a precise account of what is involved in acting
Duty and Moral Worth 133

from sympathy. It might seem clear that what he does say is incompatible
with that suggested by his critics. Kant holds acting from sympathy to be a
kind of acting from inclination (GMS 398). If my reading of acting from
inclination is correct (see sections 1.6“8), his holding this entails that, in his
view, all of an agent™s acting from sympathy is conditional on his belief that
it will have some hedonic payoff for the agent, for example, in relieving his
pain at seeing another suffer. Some might think that no action thus condi-
tioned is really done for the sake of another person. If you genuinely act
for another™s sake, the idea goes, then your expectation of a hedonic payoff
for yourself does not enter into your motivation. But I am not convinced.
Sometimes, at least, a particular action can be conditional yet done for the
sake of another. Suppose for example that given my current ¬nancial goals,
I decide that my birthday gift to a friend must meet a certain condition: it
must cost less than ¬fty dollars. It seems that, nevertheless, I might buy the
present for the friend™s sake. Analogously, my helping a stranger might be
conditional on my expectation of hedonic bene¬t for myself “ for example,
the disappearance of my pain of seeing him suffer “ yet its being so seems
compatible with my helping the stranger for his own sake. For it does not
seem to prevent me from having as one of my ultimate ends to improve the
stranger™s condition.29 Since Kant does not construct a detailed account of
what it means to act from sympathy, it is hard to determine the extent to
which his basic concept of such action diverges from that of his critics. It
seems to me, however, that Kant™s account is compatible with the notion
that an action from sympathy is done for another™s sake.
Kant offers several arguments against the view that acting from sympathy
has moral worth. According to one speci¬cation, this view is very straight-
forward. If an action is done from sympathy, then it has moral worth; an
action™s being done from this motive is a suf¬cient condition for its being
morally good. Critics of the Kantian view have not, as a rule, held this view.
One critic, for example, denies that it is morally good for a bystander to have
sympathy for a corporate criminal™s hiding his face from cameras as he is
being led to prison. The critic would presumably also hold that it would be
devoid of moral worth to act from this sympathy, for example, by trying to
block the criminal from the cameras™ view.30 It is easy to generate other cases
in which many of us would refuse to grant moral worth to an action done
from sympathy. Two members of a band of “ethnic cleansers” are plundering
the house of a hated minority. One soldier sees another struggling long and
hard to open a glass display cabinet full of delicate antique dolls that the
other wants to steal for his girlfriend. From sympathy for the other soldier
alone, the one picks the lock. On the face of it, some actions seem to lack
moral worth, even if they are done from sympathy. I will have more to say
regarding examples such as this. For now, moving from the concrete to the
abstract, let us examine some arguments Kant suggests against the view that
all actions done from sympathy have moral worth.
Search for the Supreme Principle of Morality
134

First, in the Groundwork Kant suggests that if an action is not done from
duty, it is done from inclination (section 1.6). Since actions from sympathy
are not done from duty, they are done from inclination. Yet there is no guar-
antee that an action from any particular inclination, including sympathy,
will actually be in accordance with duty (see section 5.5). Actions from incli-
nation, including those from sympathy, sometimes con¬‚ict with duty (GMS
390, 398; Rel 30“31, English ed. 26). Since only actions in accordance with
duty can have moral worth, it is not the case that if an action is done from
sympathy, then it has moral worth. This argument rests on the premise that
only actions in accordance with duty can have moral worth. But if I am cor-
rect, the logic of Kant™s own position compels him to reject this premise
(sections 6.5“7). That an action from sympathy con¬‚icts with duty does not
in itself give Kant legitimate grounds for denying it moral worth.
Kant might locate a second basis for rejecting an action™s being done
from sympathy as suf¬cient for its having moral worth in the conditional
nature of actions from inclination. As we noted, in Kant™s view, all of an
agent™s acting from sympathy is conditional on her expectation that it will
have some hedonic bene¬t for her. But the desire for pleasure has been
foisted upon us by nature. In acting from inclination, even from sympathy,
we are (in part) pursuing an end that we have not set ourselves. Only in
acting from duty do we manifest the independence from animality that
gives our action a special worth (see sections 5.4“5). This argument rests
on two very controversial premises. The ¬rst is that all acting from sympathy
is conditional in the way Kant holds. Kant™s critics do maintain that the
sympathetic agent acts from an emotion, one component of which is an
affective state (e.g., pain at the suffering of others). Yet they would probably
not agree that the sympathetic agent™s action is conditional on the expectation
of a hedonic bene¬t to herself (e.g., the relief of her pain at the suffering
of others). The sympathetic agent, the critics might say, would help even if
she believed that doing so would, on balance, increase her own suffering
by, for example, making her more familiar with the excruciating pain of a
burn victim. Since Kant simply sets out rather than argues for his hedonistic
account of all acting from inclination, and since this account does not seem
to be deeply entrenched in ordinary moral psychology, he would not be on
strong ground in insisting the critics are misguided. Another premise on
which Kant™s argument rests is that moral value accrues only to actions that
manifest a greater independence from our sensuous nature than any actions
from inclination (see section 5.4). Yet as Kant™s critics would surely wonder,
why should we consider an action™s manifesting this greater independence
from sensuous nature to be requisite for its having moral worth? Why place
so much importance on it? It would be one thing if Kant actually held that
actions from inclination were, from all perspectives, totally unfree. But he
does not hold that. According to him, all actions are done on maxims the
construction of which involves the spontaneity of the will (section 2.2).
Duty and Moral Worth 135

More promising in my view than the ¬rst two bases is a third one Kant sug-
gests for rejecting the notion that being done from sympathy is a suf¬cient
condition for an action™s having moral value. This basis is that an action
done from duty, but not one done purely from sympathy, re¬‚ects a commit-
ment to morality. The agent takes the action to be of a kind that is morally
required. At some point, though not necessarily at the time of the action, she
has judged that it is. She acts against the background of (what I have called)
conscientious re¬‚ection “ that is, thought regarding the moral status of her
action. Even if in acting from duty she gets things wrong and violates Kant™s
moral law, her action expresses conscientiousness “ something that cannot
be said for an action done from sympathy alone. Although both an agent
who acts from sympathy alone and one who acts from duty might actually
act contrary to what morality dictates, the latter does so against the back-
ground of concern with the moral status of what he does. To the Kantian,
such concern is a necessary ingredient in a morally valuable action.
Of course, Kant™s critics might object to this use of the notion of moral
commitment. Granted, actions from duty necessarily take place against the
background of conscientious re¬‚ection, whereas actions from sympathy
alone do not. Yet it would be question-begging simply to assume that only
actions involving conscientious re¬‚ection have moral worth. Some virtue
ethicists deny that conscientious re¬‚ection need play any role whatsoever
in a morally valuable action.31 Why should we not hold that what gives an
action moral worth is simply its being done from sympathy?32
In answer I can offer only an appeal to the view (which I take to be widely
shared) that some actions done from sympathy alone do not have moral
worth. The ethnic cleanser™s action of trying, from sympathy alone, to help
a “blood brother” steal from the home of an ethnic minority is such an
action. Yet in light of section 6.9 it might seem suspicious to appeal to such
examples here. After all, have I not defended the view that if the ethnic
cleanser™s action is done from duty, then it has moral worth? Some might
hold this view to be every bit as implausible as the view that done from
sympathy, his action has such worth. I do not believe that it is, but I must
leave it to the reader to decide.
Perhaps the following consideration can help tip the scale in favor of
Kantian conscientiousness over sympathy. Given the conditions that must
be met for an action to be done from Kantian duty, it seems unlikely that
the ethnic cleanser™s action would be. It seems not very likely, for example,
that the one soldier™s incentive for picking the lock on the doll case for his
comrade would stem from his notion that a universally and unconditionally
binding principle required this action. Yet it seems more likely that such
an action would be done from sympathy. Why would it be unusual for an
ethnic cleanser to think that his fellow soldier is suffering (he really wants
those dolls for his girlfriend), feel distress at his suffering, and try to help
him for his own sake?33 In short, there are possible cases of acting from
Search for the Supreme Principle of Morality
136

duty (in Kant™s precise sense) that make it dif¬cult for us to maintain that
all acting from duty has moral worth, and there are possible cases of acting
from sympathy that make it hard for us to hold that all acting from sympathy
has moral worth. However, I believe that possible cases of the latter sort are
much more likely to be actual.
If we hold moral commitment and the conscientious re¬‚ection that goes
along with it to be a necessary ingredient in morally valuable action, then
we must reject not only the notion that being done from sympathy alone is
a suf¬cient condition of an action™s having moral worth, but also a second,
different speci¬cation of the critics™ view. On this speci¬cation, only some
actions done from sympathy alone have moral value, namely the ones that
are actually in accordance with what we take moral requirements (or moral
virtue) to involve. On this speci¬cation, acting from sympathy alone to aid a
fellow ethnic cleanser to burn down a village mosque would presumably not
count as having moral value. But many other actions done from sympathy
alone “ for example, giving water to a thirsty old man “ would count as
having it. Kant would acknowledge that such an action “deserves praise and
encouragement” but not “esteem” (see GMS 398). For Kantians, if an action
is not done against the background of commitment to morality, then it does
not have moral worth “ regardless of whether it is in accordance with what
morality requires.
A third version of the sympathy objection poses a greater challenge to
Kant™s position. According to it, an action™s being done from sympathy does
not itself give it moral worth. Yet if, against the background of an overriding
commitment to morality, an action is done from sympathy, then it has such
worth. An agent has an overriding commitment to morality just in case he
acts against the background of conscientious re¬‚ection, and if after such
re¬‚ection he determines that an action is contrary to what he takes to be
morally required, he will for this reason refrain from performing it. A couple
of points regarding this objection warrant immediate attention. First, it does
not deny that actions from duty have moral worth. The objection does not
embrace the conclusion that moral worth is to be found only in (some)
actions from sympathy. Actually, and this is the second point, acting from
sympathy and with an overriding commitment to morality will involve the
possibility of reliance on the motive of duty. Suppose, for example, that
someone feels sympathy for a relative in need and is inclined to help him.
An overriding commitment to morality would require that if aiding him “ for
example, by falsely testifying to his whereabouts on the night of a robbery “
would be contrary to (his understanding of) duty, he must for this reason
refrain from doing so. In the absence of any inclination to refrain, he would
need to rely on the motive of duty to have suf¬cient incentive to conform
to (his understanding of)morality.
Kant himself denies moral worth to any action done from a motive other
than duty. Yet does he have good grounds for denying it to actions done from
Duty and Moral Worth 137

sympathy against the background of an overriding commitment to morality?
The two arguments of his that we have discussed do not seem to threaten this
view. Acting from sympathy against the background of such a commitment
no more contingently leads to action in accordance with duty than does
acting from duty. In both cases, an agent tries but might fail to conform to
morality™s demands. Perhaps Kant would claim that actions from sympathy
fail to express the high degree of independence from sensuous drives that
actions from duty do and, on that basis, deny the former moral worth. As we
have noted, however, this claim rests on the premises that all actions from
sympathy are conditional on the prospect of a hedonic payoff and that the
lesser independence from sensuous drives expressed in such actions itself
disquali¬es them from having moral worth. The prospect of successfully
defending either of these premises seems dim, and I will not try to do so
here. Of course, one might be able to develop other Kantian arguments
against the view that moral worth accrues to actions from sympathy done
against the background of an overriding commitment to morality. But unless
one does, it seems that Kant is left with no convincing rebuttal to this view.
I believe that many will be attracted to this view, as am I. Although the
view does not here get the detailed attention it perhaps deserves, I would
like to discuss one question regarding it. Suppose that someone, against the
background of an overriding commitment to morality, acts simply from a
desire to relax: he sees a ¬lm. At some point, the person re¬‚ected on the
moral status of this sort of action. If through this re¬‚ection he had found
that actions like it were morally impermissible, he would, motivated by this
¬nding, have refrained from seeing the ¬lm. Although the agent™s man-
ner of acting in some sense re¬‚ects a good character, it would be odd and,
I think, unacceptable to hold that it had moral worth. But is there a basis on
which a Kantian could deny that his action had moral worth, yet af¬rm that
some actions from sympathy have such worth, namely those done against
the background of the sort of commitment we have been discussing? Of
course, it would not do for a Kantian to locate the basis for this in the no-
tion that acting from a desire to relax lacks something that acting from the
motive of sympathy has, namely unconditional value. The Kantian denies
(correctly, I believe) that acting from sympathy has such value. (At issue
is the suggestion that perhaps the Kantian should, nevertheless, allow that
an action done against the background of an overriding commitment to
morality and [at the same time] from sympathy, has moral, and thus un-
conditional, worth.) There is, I think, a way for the Kantian to distinguish
those actions, done against this background, which do have moral worth,
from those, also done against it, which lack it. She might simply appeal to
ordinary rational knowledge of morals to support the notion that different
motives have different value characteristics. It is a feature of sympathy that
when an agent acts from it, as well as against the background of an overrid-
ing commitment to morality, his action has moral worth. However, it is not,
Search for the Supreme Principle of Morality
138

for example, a feature of greed that when an agent acts from it, as well as
against this background, his action has moral worth.
At this point, some further questions might come to mind. Why limit
moral worth to all actions done from duty and some actions done from
sympathy? Might not ordinary moral consciousness take it as a feature of
other motives (e.g., love of God) that when an agent acts from them, as
well as against the background of an overriding commitment to morality,
his action has moral worth? I would like to acknowledge this possibility. A
detailed exploration of re¬‚ective moral common sense would be necessary
to develop a full list of those motives that, against the requisite background,
would produce actions that have moral worth. I do not try to construct such
a list here.


6.11 Summary
Kant claims that an action has moral worth if and only if it is done from duty.
The logic of Kant™s own position, I have found, compels him to af¬rm that
some actions contrary to duty can be from duty, and can thus have moral
worth. Against the background of this ¬nding, I have defended one-half
of Kant™s claim, namely that being done from duty is a suf¬cient condition
for an action™s having moral worth. (An agent acts from duty just in case
her incentive for acting stems from the notion that a principle, represented
by her as a law, requires the action; this notion itself provides suf¬cient
incentive for her acting; she acts against the background of conscientious
re¬‚ection; and she does her best to realize her action™s end.) The other half
of Kant™s claim I ¬nd far less compelling. In my view, Kant does not establish
that being done from duty is a necessary condition for an action™s having
moral worth. Although an action™s being performed against the background
of a commitment to morality is requisite for its having such worth, its being
done from duty might not be. Kant does not successfully rule out the view
that actions from sympathy, when performed against this background, also
have such worth.
Since he does not, it will not be open to us to appeal in the derivation
of the Categorical Imperative to the notion that only actions from duty
have moral worth. That it will not be might seem to place the derivation
in jeopardy, since this notion is so entrenched as a central Kantian dictum.
As I try to show in the following chapters, however, rejecting this notion
lessens little if at all the force of the derivation. Key to the derivation is not
the notion that only actions from duty have moral worth but that all such
actions have it.
7

Eliminating Rivals to the Categorical Imperative




7.1 Aims of the Discussion
On the criterial reading, Kant™s derivation of the Formula of Universal Law
has three main steps. First, Kant tries to pinpoint the features that we, on
re¬‚ection, believe that the supreme principle of morality must possess. Next,
Kant attempts to establish that no possible rival to the Formula of Universal
Law ful¬lls all of these criteria. Third, Kant tries to demonstrate that the
Formula of Universal Law remains as a viable candidate for a principle that
ful¬lls all of them. The third step is discussed in Chapter 8.
The current chapter concentrates on the second: how does (or might)
Kant try to eliminate all possible rivals to the Formula of Universal Law?
To succeed, Kant would need to prove that no possible rival possesses all
of the necessary features of the supreme principle of morality that he has
identi¬ed. It is doubtful that Kant could do so de¬nitively, for it is hard to see
how he could demonstrate that he had actually considered every alternative
to the Formula of Universal Law. Nevertheless, if we accept Kant™s view of the
features that the supreme principle of morality would have to possess, then
his argument by elimination has some force. For it does show that certain
rivals to the Formula of Universal Law fail to be viable candidates for the
supreme principle of morality.
Since in trying to eliminate rivals we will be appealing to criteria Kant
develops for the supreme principle of morality, it is helpful to have the
criteria in view. According to Kant™s basic concept, the supreme principle of
morality would have to be (i) practical, (ii) absolutely necessary, (iii) binding
on all rational agents, and (iv) the supreme norm for the moral evaluation of
action. Moreover, (taking into account the modi¬cations we made to Kant™s
further criteria in Chapter 6) this principle must be such that: (v) every
case of willing to conform to it because the principle requires it has moral
worth; (vi) the moral worth of willing to conform to the principle because the
principle requires it stems from its motive, not from its effects; (vii) an agent™s

139
Search for the Supreme Principle of Morality
140

representing the principle as a law “ that is, a universally and unconditionally
binding principle “ provides him with suf¬cient incentive to conform to it;
and, ¬nally, (viii) a plausible set of duties (relative to ordinary rational moral
cognition) can be derived from the principle. Regarding criterion v, let me
make explicit that, on my understanding, willing to conform to a principle
because the principle requires it amounts to willing, from duty, to conform to
it. It amounts to ful¬lling each of the four conditions speci¬ed in Chapter 6
for acting from duty (see section 6.9).
As it happens, the rivals to the Formula of Universal Law we discuss
(e.g., utilitarian principles) are also rivals to the Formula of Humanity.
Even though, as I argued in Chapter 3, Kant™s derivation of the Formula of
Humanity is unsuccessful, we need not give up on this formula as a candidate
for the supreme principle of morality. Kant conducts a criterial derivation
of the Formula of Universal Law. But the same technique might be used
to good effect with regard to the Formula of Humanity. In this chapter I
use the term “the Categorical Imperative” loosely to refer to either formula
(even though I do not hold the two to be equivalent), since no differences
between the two formulas will come into play.
Among the Categorical Imperative™s most pressing contemporary rivals
are consequentialist principles. I try to show that some of the criteria for the
supreme principle of morality that we have discussed at length serve as the
basis for premises in a Kantian argument against consequentialist rivals.1
This argument applies not only to familiar utilitarian forms of consequen-
tialism (sections 7.3“5), but to less familiar forms of it including “Aristotelian
perfectionism” (7.6) and “Kantian consequentialism” (7.7). Moreover, if we
share some of Kant™s views regarding the moral worth of actions, the Kantian
argument against certain consequentialist principles succeeds.
Of course, not all challengers to the Categorical Imperative as the supreme
principle of morality are consequentialist principles. With no pretension to
exhaustiveness, I consider three that are not: a principle (somewhat similar
to the Ten Commandments) in which several prescriptions are conjoined
(7.8), and two variations on the Formula of Universal Law (7.9). I argue that
if we accept Kant™s criteria for the supreme principle of morality, then his
arguments against these principles also have considerable force.
Before turning to arguments aimed at speci¬c rivals to the Categorical
Imperative, however, we consider an argument through which, with one
broad stroke, Kant tries to eliminate all rivals. As I try to show in section 7.2,
this sweeping argument is a failure.


7.2 A Sweeping Argument against All Rivals
Kant suggests this sweeping argument in both the Groundwork and the second
Critique (GMS 444, KpV 21“41), albeit in slightly different terminology. First,
he contends that all candidates for the supreme principle of morality are
material, except for the Categorical Imperative, which is formal. Second, he
Rivals to the Categorical Imperative 141

claims that no material practical principle could be the supreme principle
of morality. Therefore, he concludes, the only viable candidate remaining
is the Categorical Imperative. I begin by considering Kant™s argument for
the second premise, then move to the ¬rst.
We have already explored Kant™s basis for the second premise of his argu-
ment, that is, for the claim that no material principle could be the supreme
principle of morality. A material principle, let us recall, is one that an agent
has suf¬cient motive to conform to only if he expects that doing so will re-
sult in the realization of some object he desires and that realizing this object
will have a hedonic payoff for himself (section 1.8). A formal principle is
(in one sense) a principle such that an agent™s representing it as a law itself
gives him suf¬cient motive to conform to it (section 4.7). Kant contends
that a viable candidate for the supreme principle of morality must be a for-
mal principle. To establish this, Kant must obviously show that the supreme
principle of morality could not be a material principle. He tries to do so
with the arguments we examined in section 5.7.
Although Kant has a (at least moderately) hedonistic view of material
practical principles (section 1.8), we can summarize one of these arguments
without appealing to this view. In the summary (which is based on GMS
444), we can see that another way Kant has of putting the claim that the
supreme principle of morality must not be a material principle is to say that
it must not be a heteronomous one.2 The starting point of this argument
is familiar “ namely the notion that, according to the supreme principle of
morality™s basic concept, it must be unconditionally binding on all of us. It
must be a categorical imperative for all rational beings who, like us but unlike
perfectly rational beings such as God, do not necessarily conform to it. A
material principle is a rule such that an agent has suf¬cient motive to adhere
to it only on condition that, in her view, doing so will enable her to realize
some object she desires. (For present purposes let us stop there, without
invoking Kant™s notion that such principles are hedonically conditioned.)
In order for a material principle to be a categorical imperative, each agent
must have suf¬cient motive available to her to abide by it. If each did not,
then in some circumstances it would be impossible for her to abide by it,
and, therefore, in Kant™s view, it would not be binding on her. (An agent
cannot have a duty to do something that it is impossible for her to do.)
But now suppose that a material principle speci¬es the means to realize
some object that a particular agent does not desire, for example, greater
perfection de¬ned as the development of physical and rational capacities.
(Maybe the agent thinks that she is mentally and physically ¬t enough to
lead a rewarding life.) In this case, the agent might ¬nd herself without
suf¬cient motive available to her to conform to the principle and thus un-
able to conform to it. If she did, then the principle would not be binding
on her. But principles that, in light of a particular agent™s desires, might
not be binding on her (i.e., material principles) are obviously not categor-
ical imperatives. Since they are not, they are not viable candidates for the
Search for the Supreme Principle of Morality
142

supreme principle of morality. In the Groundwork, Kant crystallizes this ar-
gument thus: “Whenever an object of the will has to be laid down as the
basis for prescribing the rule that determines the will, there the rule is none
other than heteronomy; the imperative is conditional, namely: if or because
one wills this object, one ought to act in such or such a way; hence it can
never command morally, that is categorically” (GMS 444). Material practical
principles are heteronomous in the sense that their motivational force stems
from something outside of the will, something that each rational being does
not necessarily have: a desire for some particular object.
However compelling this argument may be, it is not enough to insure the
success of Kant™s attempt to sweep away rivals to the Categorical Imperative.
Even if Kant establishes that the supreme principle of morality must not be
material (i.e., the second premise in this attempt), he needs to convince us
of the ¬rst, namely that all rivals to the Categorical Imperative are indeed
material principles. Unfortunately, he does not do so.
In the second Critique (KpV 40) Kant sets out a table in which he cat-
egorizes rivals to the Categorical Imperative. He distinguishes “subjective”
from “objective” principles, and “internal” from “external” ones. Subjec-
tive principles are empirical; their content stems from experience. Among
such principles Kant mentions that “of education,” which, Kant asserts, was
advocated by Montaigne. Apparently, this principle derives the content of
morality solely from custom. Objective principles are based on reason, or at
least purported to be so. Wolff, for example, claimed to base his principle
of perfection not on experience but on rational concepts alone. In Kant™s
scheme, some subjective principles are internal, some external. Montaigne™s
principle is external in that education stems from outside of the agent, while
another subjective principle, that of moral feeling defended by Hutcheson,
Kant classi¬es as internal, apparently since this feeling is internal to the
agent. Kant also distinguishes between an objective principle that is exter-
nal, namely that of the will of God, and one that is internal, that of perfec-
tion. In sum, Kant categorizes six rivals to the Categorical Imperative. Of
the three whose content stems from within the agent (internal principles),
two of these are subjective, the principles of “physical feeling” and of “moral
feeling,” and one objective, the principle of “perfection.” Of the three whose
content stems from outside the agent (external principles), two of these are
subjective, the principles of “education” and “the civil constitution,” and
one objective, the principle of “the will of God.”
Referring to this table, Kant makes several claims: “all the principles ex-
hibited here are material ” (KpV 41), “they include all possible material prin-
ciples” (KpV 41), and “all possible cases are actually exhausted, except the
one formal principle” (KpV 39), namely the Categorical Imperative. Each of
these claims is controversial. Regarding the ¬rst, one might wonder whether
a theologian would or need acknowledge that a principle of obeying God™s
will is necessarily material. Why should he accept the view that an agent has
Rivals to the Categorical Imperative 143

suf¬cient motive to obey God™s will only if she thinks that doing so will en-
able her to satisfy some desire she has? Might he not instead maintain that
an agent™s notion that God wills that she do something itself can give her
suf¬cient motive to do it? Regarding Kant™s second claim, one might won-
der whether his table actually does include all possible material principles.
What about a (loosely) Nietzschean principle, something such as “In order
to ¬‚ourish, you ought to maximize your power”? Although this principle is
not among the six Kant lists, he might contend that it does ¬t into his gen-
eral schema as, for example, a subjective and internal principle. In any case,
Kant™s ¬nal claim, which amounts to the ¬rst step of his sweeping attempt
to eliminate all rivals, is the one most obviously vulnerable to criticism.
It is not hard to imagine rivals to the Categorical Imperative that, at least
on the surface, are not material principles. Suppose someone defends the
following perfectionist principle, MP™: “Develop your physical and rational
capacities.” According to the defender, MP™ commands categorically. It does
not say: develop your physical and rational capacities, if you want or given
that you want to perfect yourself or be happy or attain some other object.
It prescribes that you develop these capacities no matter what you want. In
reply, Kant might insist that though the defender does not take MP™ to be a
material principle, it is indeed one. For an agent could have suf¬cient mo-
tive to conform to MP™ only on condition that he expected doing so would
enable him to realize some object he desired. Yet, as far as I can tell, Kant of-
fers no argument for this contention. Why couldn™t an agent be motivated to
conform to MP™ simply by representing MP™ to himself as an unconditionally
and universally binding principle? If an agent™s representing the Categorical
Imperative to himself as a practical law gives him suf¬cient incentive to con-
form to this principle, why couldn™t an agent™s representing MP to himself
as a practical law give him suf¬cient incentive to conform to that principle?
In the next section, we discuss in detail how Kant might try to eliminate
consequentialist principles as candidates for the supreme principle of moral-
ity. But we can already see that it will not do to sweep them away on the basis
that they are all material principles. Take the utilitarian principle U™: “Always
perform a right action, one that yields just as great a sum total of well-being
as would any alternative action available to you.” As far as I can tell, Kant
does not establish that an advocate of this principle must acknowledge that
an agent™s having suf¬cient grounds to conform to it is conditional on her
wanting to maximize well-being or to gain pleasure for herself or anything
else. The possibility persists that she ¬nds suf¬cient grounds for complying
with U™ in the notion that doing so is morally required.3 Defenders of a
variety of candidates for the supreme principle of morality might refuse to
acknowledge their principles to be “material.” And Kant, it seems, has no
good argument with which to discredit such a refusal.
Yet perhaps we have not looked hard enough. In a chapter entitled “On
the Concept of an Object of Pure Practical Reason,” Kant describes his
Search for the Supreme Principle of Morality
144

method in the second Critique: “[I]instead of the concept of the good as an
object determining and making possible the moral law, it is on the contrary
the moral law that ¬rst determines and makes possible the concept of the
good, insofar as it deserves this name absolutely” (KpV 64). Kant claims that
other philosophers failed to adopt this method, a failure that led them
into error regarding the supreme principle of morality. They began with an
object that they considered to be good (e.g., the perfection of our capacities)
and tried to derive a practical principle from that object (e.g., the principle
that we are required to perfect our capacities). But by beginning with a
concept of the good and then trying to derive a practical principle from
it, they condemned themselves to advancing material practical principles “
ones that are not suited to be the supreme principle of morality. In Kant™s
words, other philosophers

sought an object of the will in order to make it into the matter and the ground of a
law (which was thus to be the determining ground of the will not immediately but
rather by means of that object referred to the feeling of pleasure or displeasure),
whereas they should ¬rst have searched for a law that determined the will a priori
and immediately, and only then determined the object conformable to the will.
Now, whether they placed this object of pleasure, which was to yield the supreme
concept of good, in happiness, in perfection, in moral feeling, or in the will of God,
their principle was in every case heteronomy and they had to come unavoidably
upon empirical conditions for a moral law, since they could call their object, as
the immediate determining ground of the will, good or evil only by its immediate
relation to feeling, which is always empirical. (KpV 64)

Kant suggests the following claim: if we begin in ethics with a concept of the
good and then construct a moral principle that requires the promotion of
this good, we must acknowledge that an agent will have suf¬cient grounds to
conform to the principle only if she expects doing so will have some hedonic
bene¬t. In effect, we must acknowledge that the principle is material. If Kant
successfully defended this claim, then he might indeed have good grounds
for asserting that the rival principles we mentioned earlier were material.
An advocate of MP™ or U™ would likely begin his moral theorizing with the
concept of an object as the good: in the former case, perfection; in the latter,
the general happiness. However, as the cited passage illustrates, Kant does
not really argue for the claim in question. He leaves it strikingly unclear why
a principle based on the concept of some object as the good must be such
that an agent could have suf¬cient grounds for conforming to it only if he
expected a hedonic payoff from doing so. To insist that such a principle
must have this feature seems unfounded.
Perhaps Kant is correct that no material principle could cohere with his
basic concept of the supreme principle of morality. However, this claim does
not give him a quick route to the elimination of all rivals to the Categori-
cal Imperative. For Kant does not show that all rivals actually are material
practical principles.
Rivals to the Categorical Imperative 145

7.3 The Structure of Act Utilitarianism
In light of the shortcomings of Kant™s sweeping attempt to dismiss all rivals
to the Categorical Imperative on the basis that they are material practical
principles, it makes sense to look for other arguments he might offer against
particular competitors.
Let us begin with consequentialist principles, speci¬cally utilitarian ones.
In his critical writings in ethics, Kant does not explicitly consider utilitarian-
ism. He mentions “the principle of sympathy for the happiness of others,”
(GMS 442, note), which he attributes to Hutcheson. And he discusses brie¬‚y
the possibility that the happiness of others is the object of the will of a ra-
tional being (KpV 34). So Kant does seem to entertain the notion that the
supreme principle of morality is one that requires us to promote the happi-
ness of others. Against this quasi-utilitarian notion, however, Kant employs
the suspicious argument we discussed in section 7.2, one according to which
such a principle must be material.4
Kant might argue against a utilitarian principle with the help of an appeal
to his view that only the good will is unconditionally good. Whatever the
supreme principle of morality is, he might claim, it must have something
unconditionally good as its “ground.” The utilitarian would have to take
everyone™s being happy as the unconditionally good ground of her principle.
But everyone™s being happy is not unconditionally good. Since it is not,
Kant might conclude, the utilitarian principle could not be the supreme
principle of morality. This argument does not seem promising, for Kant
fails to establish that everyone™s being happy is not unconditionally good
(see section 3.7). Does he have any better argument available to him with
which to eliminate utilitarianism as a rival?
To answer this question, it is helpful to have a particular utilitarian prin-
ciple in view.

U: An action is right if and only if it yields as great a sum total of individual
well-being as would any alternative action available.

Amartya Sen has shown that U follows from two separate views:, one is an
account of goodness; the other, an account of the connection between good-
ness and rightness. According to “Outcome Utilitarianism,” the goodness of
a state of affairs is solely a function of the sum total of individual well-being in
it. More precisely, any state of affairs is at least as good as an alternative state
of affairs if and only if the sum total of individual well-being in the one is at
least as large as the sum total of individual well-being in the other.5 According
to “Act Consequentialism,” the rightness of an action is solely a function of
the goodness of its consequences. More precisely, an action is right if and
only if the state of affairs resulting from the action is at least as good as each
of the alternative states of affairs that would have resulted respectively from
the alternative feasible acts.6 In discussing U, we assume that its defender
grounds it in Act Consequentialism and Outcome Utilitarianism.
Search for the Supreme Principle of Morality
146

Against the possibility that U could be the supreme principle of morality,
Kant has available to him a simple, straightforward argument. U runs afoul
of Kant™s basic concept of this principle. On this concept, the supreme
principle of morality would manifest itself to us (human rational agents) as
a categorical imperative. It would be absolutely necessary, prescribing that we
ought to act in a certain way, no matter what our particular inclinations might
be. However, U just tells us which actions are right. It does not prescribe to us
that we ought to do right actions. Strictly speaking, it does not prescribe how
we ought to act at all. U does not have the form of a categorical imperative. Of
course, U™s advocate might simply reject Kant™s basic concept of the supreme
principle of morality. She might insist that such a principle need not take the
form of a categorical imperative, or even that it need not be practical (i.e.,
something on account of which we can act). The utilitarian might conceive
of her principle as a fundamental description of right action and nothing
more. Doing this would, however, not threaten the Kantian claim we are
considering “ namely, that if there is a supreme principle of morality, in
the basic sense of such a principle that Kant employs, then it is the Categorical
Imperative.
Rather than responding to Kant™s argument against U by rejecting his
basic concept of the supreme principle of morality, the utilitarian can simply
give U the form of a categorical imperative:
U™: Always perform a right action, one that yields just as great a sum total
of well-being as would any alternative action available to you.
Here the utilitarian has added a further principle to the two from which U
was constructed “ namely, what we might call the principle of imperative
rightness: Always act rightly. The resulting principle U™ appears to conform
to Kant™s basic concept of the supreme principle of morality. It could be a
practical, absolutely necessary, universally binding, fundamental norm for
moral evaluation of action. How might Kant exclude the possibility that it is
the supreme principle of morality?


7.4 Against Act Utilitarianism
To eliminate rivals to the Categorical Imperative, Kant has at his disposal not
only the criteria contained in his basic concept of the supreme principle of
morality but also the further ones he develops in Groundwork I. Using some
of these further criteria, it is fairly simple to construct an argument to block
the possibility that U™ is the supreme principle of morality:

1. Whatever the supreme principle of morality is, your willing from duty
to conform to it has moral worth.
2. This moral worth does not stem from any effect of what you do, but
rather solely from your willing from duty to conform to this principle.
Rivals to the Categorical Imperative 147

3. Suppose you held that U™ were the supreme principle of morality.
4. You would then have to hold that whether your willing from duty to
conform to U™ had moral worth depended solely on its effects.

5. Since (according to 1 and 2) the moral worth of your willing from
duty to conform to the supreme principle of morality does not stem
from its effects, you must conclude that U™ cannot be this principle.

Although Kant does not make this argument explicitly, its steps are famil-
iar to us from our exploration of his criteria for the supreme principle of
morality.
The argument™s ¬rst step stems, of course, from a criterion Kant develops
in Groundwork I in his “¬rst proposition.” This principle, he claims, must be
such that all and only actions conforming to it because the principle requires
it (i.e., all and only actions done from duty) have moral worth. However,
step 1 differs from Kant™s criterion in two signi¬cant ways. Obviously, it in-
vokes not at all Kant™s view that only actions from duty have moral worth.
Moreover, embracing the premise involves rejecting the idea that only ac-
tions in conformity with duty can have moral worth. According to step 1,
whatever the supreme principle of morality is, your willing, from duty, to con-
form to it has moral worth “ even if, as it turns out, you fail to conform to it.
The second step is closely related to a further criterion for the supreme
principle of morality that Kant develops in Groundwork I (in his “second
proposition”). It follows from this criterion that we cannot af¬rm a principle
to be the supreme principle of morality unless we can hold that the moral
worth of any actions conforming to it from duty does not stem from the
actions™ effects. The main thrust of step 2 is the same as that of this criterion,
namely that the moral worth of an action does not stem from its effects or
results. However, in line with 1, step 2 does not restrict moral goodness
to actions that actually conform to the supreme principle of morality. It
(implicitly) grants that attempts to conform to the supreme principle of
morality, even if they fail, can have moral worth.
Step 4 also requires attention. In considering 4 it is important to put
ourselves in the position of someone who has, as 3 speci¬es, accepted U™
as the supreme principle of morality. We are assuming, let us recall, that
a person who accepts U™ as the supreme principle of morality grounds it
in Act Consequentialism and Outcome Utilitarianism. Such a person holds
that the goodness of a state of the universe is solely a function of the sum
total of individual well-being in it: the greater the sum, the better the state of
the universe. Now the question arises: according to a defender of U™, when
would an action done on account of U™ have moral value? A defender of
U™ sees all value (goodness) in terms of individual well-being. Therefore,
he must see the value of an action in terms of its effects on individual well-
being. It is unclear precisely where he will draw the line between an action
Search for the Supreme Principle of Morality
148

that has a positive value and one that does not. He might, for example,
claim that an action has positive value if, on balance, it raises the sum total
of individual well-being (rather than diminishing it or having no impact on
it). Or he might claim that to have positive value an action must be right “
that is, produce just as much well-being as any alternative action. Whatever
his particular view might be, for him the value (and thus moral value) of an
action is solely a function of its effects.
The proponent of U™ might respond that in his view any (positive) value of
an agent™s conforming to U™ “from duty” stems not from the effects the action
actually has but from the agent™s motive in conforming to U™. The value
derives from her willing to conform to U™ because, she believes, conforming
to it is morally required. So, for example, suppose someone tries to save a
stranger who is choking because she believes that morality (in the guise of
U™) demands it. The value of this action, the proponent of U™ might say, is
just a function of her motive in doing it, not its effects, for example, not
whether she indeed succeeds.
But this response lacks force. Granted, the proponent of U™ is not com-
mitted to holding that aiding a choking victim has moral value only if it
results in the victim™s being saved. (Although the victim might die, the ex-
ample provided to others by the attempt to save him might inspire others
to actions of the same sort, and thereby increase the sum of individual well-
being.) Yet it is not open to the proponent to derive the value of someone™s
willing to save the victim solely from his being motivated by U™ to do so.
The proponent has de¬ned the good in terms of well-being. Given that he
has, any value possessed by acting on U™ as a motive would stem from its
(somehow) promoting the general welfare. It would stem ultimately from
its effects.


7.5 Against Expectabilist Utilitarianism
Of course, even if the Kantian argument I have sketched is effective against
an act utilitarian principle such as U™, it might not work against other
varieties of utilitarianism. For example, what about an expectabilist prin-
ciple? This kind of utilitarian principle some might ¬nd most plausible.
Would Kant™s argument, if we assume that all actions from duty have moral
worth, eliminate such a principle as a candidate for the supreme principle
of morality? Consider
EU: Always perform a right action: one that you expect will yield as great
a sum total of well-being as would any alternative action available to
you.
Now let us suppose that a defender of this principle embraces it partly
because, like the defender of U™, he endorses Outcome Utilitarianism: he
holds that the goodness of a state of affairs is solely a function of the amount
Rivals to the Categorical Imperative 149

of individual well-being in it. At ¬rst it might seem that a defender of EU
could easily escape the Kantian argument. For unlike U™, it might appear
that EU would not run afoul of step 2. The defender of EU would, it seems,
not be committed to the view that the value of conforming to this principle
depended on its effects. He might coherently claim, for example, that an
agent™s action has moral value just in case she does what she expects would
maximize well-being because she takes that to be the right thing to do. To
have moral value, her action need not have the result of actually maximizing,
or even promoting, well-being. Therefore, it seems, the defender of EU is
not forced to reject step 2 and thus does not fall prey to the argument.
In response to this challenge, I want to argue that, actually, this defender
of EU cannot coherently hold that the moral value of conforming to EU
from duty does not depend on its effects. The most ef¬cient way to make this
argument is with the help of a thought experiment. Imagine an agent who
has always conformed to EU because he has taken it to be morally required
that he do so. Nevertheless, each of the agent™s actions has diminished the
sum of individual well-being, even though there have always been actions
available to him that would have promoted it. Various factors are responsi-
ble for this phenomenon. Sometimes, his best efforts notwithstanding, the
agent, who is no expert in psychology, economics, or probability theory, de-
veloped irrational expectations of the effects of a proposed course of action
on the general welfare, and, as luck would have it, things went just as an ex-
pert would predict. At other times, the agent™s expectations corresponded
with those of the experts, but the world simply failed to cooperate. He ex-
pected that praising his colleague would make him feel better, but it actually
plunged the colleague deeper into depression. He expected that his giving
to a famine relief fund would reduce the suffering caused by starvation, but
it actually ended up providing food for a paramilitary unit who ransacked a
peaceful village. Not even the example the agent set for others by his unwa-
vering conformity to EU had a positive effect on the sum of individual well-
being. Taken individually and taken as a whole, his actions neither directly
nor indirectly increased the sum total of individual well-being but actually
decreased it (though actions available to him would have increased it).
Our defender of EU as the supreme principle of morality would not be
justi¬ed in holding that the agent™s actions had moral value. The defender
embraces Outcome Utilitarianism. He holds that the goodness of a state of
affairs is solely a function of the sum of individual well-being in it. Ultimately,
his only basis for judging that an action is good is that it have a positive
effect on this sum (perhaps relative to other available actions). However,
the agent™s actions do not have a positive effect on this sum (even relative
to other available actions). Since they do not, the defender has no basis for
saying that they are good. Despite initial appearances, the defender of EU
is committed to the view that, contrary to step 2, the value (including the
moral value) of an agent™s actions does depend on their effects. Therefore,
Search for the Supreme Principle of Morality
150

as the thought experiment illustrates, the defender is committed to the view
that an agent™s action of conforming to EU because he thinks it to be the
right thing to do can fail to have moral value, thus contradicting step 1.
Some philosophers will not be satis¬ed with this response, insisting that it
neglects an important distinction between evaluation of actions and that of
states of affairs. While the defender™s embracing of Outcome Utilitarianism
requires him to judge the goodness of a state of affairs solely in terms of the
sum of individual well-being in it, his embracing of it does not require him
to judge actions solely in these terms, they will say. He is free to judge the
goodness of actions independently of their effects, in agreement with step 2
and thus ultimately with step 1. That the defender de¬nes the goodness of a
state of affairs solely in terms of well-being does not entail that he must de¬ne
the goodness of actions simply in terms of their production of well-being.
This reply does not seem convincing, as a further thought experiment
may show. Imagine two worlds. World I is that of our unfortunate agent from
the previous example “ the one who, “from duty,” always conforms to EU, but
whose actions never have a positive effect on the sum of individual well-being
(though actions available to him would have such an effect). Let us suppose
that in this world at a particular time (t), the sum of individual well-being is
ten units. In World II, the sum of individual well-being at t is also ten units.
World II is just like World I except that in it our agent™s motive for conform-
ing to EU has never been that he takes it to be morally required to do so.
Now let us return to the defender of EU as the supreme principle of
morality. The proposal on the table is that the defender hold the following.
Although the value of a state of affairs is solely a function of the sum total
of individual well-being in it (Outcome Utilitarianism), the value of actions
is not. But I do not see how the defender can coherently hold this. On
a straightforward understanding, a state of affairs is simply a state of the
universe at some particular time. The defender would have to hold that the
state of affairs (World I at t) has greater value than the state of affairs (World
II at t), since more good actions have been performed in the former than
in the latter. But in holding this, he would be betraying his commitment
to Outcome Utilitarianism. For, according to this doctrine, the value of a
state of affairs is solely a function of the sum total of individual well-being
in it. Therefore, according to Outcome Utilitarianism, the value of World
I and World II would be identical. This reply depends on the observation
that the actions that have been performed at t constitute a part of the state
of the universe at t.7 Actions that have been performed are an element in a
state of the universe. In order to rebut my reply, a philosopher would have
to deny this “ in my view, very plausible “ account of states of affairs.
With help from two of the criteria he develops for the supreme principle
of morality, Kant can construct a strong argument against one version of
expectabilist utilitarianism. But another version of expectabilism seems not
to be vulnerable to this argument.8 Suppose someone defends the principle
Rivals to the Categorical Imperative 151

EU as the supreme principle of morality but does not embrace Outcome
Utilitarianism. She holds that an agent performs a right action just in case he
does something that he expects will yield as great a sum total of well-being
as would any alternative action available to him. Moreover, the defender
af¬rms that an action is good if and only if it is right. (She acknowledges,
of course, that under certain circumstances an action that she calls good
diminishes well-being relative to other available actions.) The defender can
coherently claim that the moral worth of an action does not depend on its
(actual) effects. She can also coherently claim that each instance of willing,
from duty, to conform to EU has moral worth. In her view, an action has
moral worth just in case it conforms to EU (or, equivalently, just in case it
is right). And presumably every case of willing from duty to conform to EU
will be a case of conforming to it.9
An argument advanced in Chapter 6 supplies the basis for a Kantian re-
sponse to this version of expectabilist utilitarianism. This response, which I
merely sketch, emerges from discussion of, but does not appeal to, Kant™s
criteria for the supreme principle of morality.10 In Chapter 6, I defended
the view that Kant should acknowledge that some actions contrary to the
Categorical Imperative have moral worth. Suppose an agent has done his
best to ¬gure out what the supreme principle of morality is but has become
convinced that it is something other than the Categorical Imperative. If,
from duty, he wills to conform to this other principle but violates the Cat-
egorical Imperative, Kant should nevertheless acknowledge that his action
has moral worth. He should acknowledge this (roughly) because, intuitively
speaking, the factors that are requisite for moral worth are present. The
agent™s incentive for the action stems from the notion that it is required
by an unconditionally and universally binding principle; he holds that the
action™s being morally required itself gives him suf¬cient incentive for the
action, and so forth. The same sort of argument applies to the version of
expectabilism in question. Its defender is committed to the following view.
The only actions that are good (and thus the only ones that have moral
worth) are those that conform to EU. But having done his best to discover
the supreme principle of morality, someone might conclude that it is some-
thing other than EU. If, from duty, this person wills to conform to this other
principle but violates EU, then the defender of this version of expectabilism
must hold that the person™s action is devoid of moral worth. But, intuitively,
I think we would want to attribute moral worth to his action. And that is a
reason for rejecting this version of expectabilist utilitarianism. The supreme
principle of morality must be such that its defender can coherently claim
that all instances of willing from duty to conform to it have moral worth,
suggests Kant. The defender of this version of expectabilist utilitarianism
can coherently claim this. However, she cannot hold something that many
of us take to be intuitively clear “ namely, that some actions done from duty
do not conform to EU, and that these actions have moral worth.
Search for the Supreme Principle of Morality
152

It is, however, possible to conceive of moral theories, which some might
call utilitarian, that elude even this argument. For example, a philosopher
might defend a principle discussed earlier, U™, yet not advocate it even partly
on the basis of Outcome Utilitarianism (the doctrine according to which
goodness is solely a function of well-being). The philosopher might hold
that each agent ought always to perform a right action: one that yields just
as great a sum total of well-being as would any alternative action available
to him. Yet she might divorce the question of an action™s rightness from its
goodness. She might hold that an action has moral worth just in case an
agent does it solely because he takes it to be morally required, regardless of
whether his action is right. To rebut this sort of theory a Kantian could, of
course, claim that U™ fails to ful¬ll criterion viii for the supreme principle
of morality “ that it fails to generate a set of moral prescriptions that coheres
with ordinary moral thinking.11 But I do not defend this claim here.12


7.6 Against Perfectionism
The two preceding sections focused largely on a type of argument that ap-
peals to Kant™s notions (roughly) that all actions from duty have moral worth
and that this worth does not depend on the actions™ effects. A shortcom-
ing of this type of argument is that, as we just noted, it fails to apply to
some forms of utilitarianism (though I think Kant does have other recourse
against these forms). A strength of this type of argument is that it applies to
some nonutilitarian principles. Recall MP™, “Develop your physical and ratio-
nal capacities.” This is a principle of what Thomas Hurka calls “Aristotelian
perfectionism.”13 A proponent of MP™ as the supreme principle of morality
identi¬es human perfection as the good. He embraces what we might call
“Outcome Perfectionism,” the view that the goodness of a state of affairs
is solely a function of the sum total of individual perfection in it. To will
from duty to conform to MP™ would presumably involve trying one™s best
to develop one™s physical and rational capacities. It is easy to see how the
argument we deployed against utilitarian principles would apply to MP™:

1. Whatever the supreme principle of morality is, your willing from duty
to conform to it has moral worth.
2. This moral worth does not stem from any effect of what you do but
solely from your willing from duty to conform to this principle.
3. Suppose you held that MP™ was the supreme principle of morality.
4. You would then have to hold that whether your willing from duty to
conform to MP™ had moral worth depended solely on its effects.

5. Since (according to 1 and 2) the moral worth of your willing, from
duty, to conform to the supreme principle of morality does not stem
from its effects, you must conclude that MP™ cannot be this principle.
Rivals to the Categorical Imperative 153

In light of our exploration of Kant™s argument against utilitarian principles,
the only step we need consider here is 4. Since a proponent of MP™ as
the supreme principle of morality identi¬es the good (including the moral
good) with human perfection, he must judge an action™s moral worth to be a
function of its effects on human perfection. Yet the effects of an action (e.g.,
an increase in an agent™s physical perfection) are obviously not identical with
the action itself (e.g., an agent™s willing to develop his physical capacities).14
Although a person might will, through a strenuous exercise regimen, to
develop his physical capacities, he might injure himself in the process. His
good faith attempt to get himself in shape might do nothing but diminish his
health and vigor. In this case, a proponent of MP™ as the supreme principle of
morality would be committed to denying moral worth to the agent™s attempt.
MP™ falls prey to the same sort of Kantian argument that applies to some
utilitarian principles.
If we conceive of a consequentialist moral principle as one according to
which the moral value of an action depends on its effects, then both U™
and MP™ count as consequentialist. Moreover, it is evident how Kant might
appeal to his account of ordinary moral reasoning to argue against any such
consequentialist principle™s being the supreme principle of morality. He
would simply invoke steps 1 and 2 as they appear in the arguments against
these two principles. I call this kind of argument “valuational,” since it turns
on the question of an action™s moral value or worth.


7.7 Kantian Consequentialism?
According to David Cummiskey, Kant™s ¬rst and second propositions do not
con¬‚ict with consequentialism.15 The argument of Groundwork I does not
really threaten the notion that the supreme principle of morality is conse-
quentialist. Contrary to Cummiskey, I have found in these propositions the
basis for a Kantian argument against three forms of consequentialism: act
utilitarianism, expectabilist utilitarianism (in one version), and perfection-
ism. Cummiskey proposes a different “Kantian” form of consequentialism.
In this section, I try to show that the Kantian argument also applies to Cum-
miskey™s Kantian consequentialist candidate for the supreme principle of
morality.
Cummiskey™s detailed statement of his candidate is very lengthy. In the
end, though, he suggests that the candidate amounts to (roughly) the fol-
lowing:
KC: Maximally promote two tiers of value: rational nature and happiness,
where rational nature is lexically prior to happiness.16
In a nutshell, KC enjoins ¬rst that we must maximally promote the condi-
tions necessary for the rational choice of ends, conditions such as liberty
and life.17 The principle thus entails that if we ¬nd that the only way to save
Search for the Supreme Principle of Morality
154

two rational agents is to kill one innocent agent, then we are required to
kill him.18 Second, KC enjoins that we must maximally promote the effec-
tive realization of rationally chosen ends.19 The ¬rst requirement is lexically
prior to the second requirement in the Rawlsian sense that we are to ful¬ll
the second only if we have completely ful¬lled the ¬rst: we are to promote
happiness maximally only if we have done all we can to promote rational
nature.20 The principle thus entails that we must not kill one person in
order to make others happy.
I do not offer a thorough discussion of KC but rather focus only on fea-
tures of it that are directly relevant to my present aim. First, Cummiskey
presents KC as a categorical imperative with a scope extending to all ra-
tional agents.21 It requires all rational agents, regardless of their particular
inclinations or desires, to promote maximally two tiers of value. Second,
Cummiskey holds KC to be a consequentialist principle in the following
sense. It sets out a requirement to promote the good (the two tiers of value),
and it does not set limits on the acceptable means that an agent may em-
ploy to promote the good.22 It does not, for example, specify a duty not to
sacri¬ce one innocent person to save two others. Third, Cummiskey holds
that KC has a Kantian foundation. A proponent of KC as the supreme prin-
ciple of morality would, he suggests, defend it in part by arguing as follows.
If an agent holds there to be a categorical imperative, then he must hold
there to be something unconditionally valuable. Upon re¬‚ection, he must
¬nd that this unconditionally valuable thing is rational nature (humanity).
For he must hold rational nature to be the source (i.e., the unconditioned
condition) of value, and thus to be unconditionally valuable.23 We exam-
ined (and criticized) this argument in Chapter 3.24 Whereas Korsgaard and,
presumably, other Kantians hold that this argument supports the Formula
of Humanity (interpreted as a nonconsequentialist principle), Cummiskey
claims that it is better suited to supporting a consequentialist principle such
as KC. I do not address the issue of whether, when taken in isolation, the
argument is better suited to supporting KC. However, I defend the view that
there are Kantian grounds, manifest in Groundwork I, for rejecting KC as a
candidate for the supreme principle of morality.
KC is subject to basically the same valuational argument as the other prin-
ciples we have examined. It runs afoul of the argument™s ¬rst step, according
to which every case of willing from duty to conform to the supreme principle
of morality (whatever it turns out to be) has moral worth. A defender of KC
as the supreme principle of morality has adopted a two-tiered conception
of the good. On the higher tier is rational nature; on the lower is happi-
ness. There is, says Cummiskey, “a normative hierarchy in the theory of the
good.”25 The goodness of rational nature is such that we are never to refrain
from maximally promoting it for the sake of promoting happiness. Imagine
a scenario in which a defender of KC as the supreme principle of morality
reasonably believes that she has not done all she can to promote rational
Rivals to the Categorical Imperative 155

nature: to promote the conditions necessary for rational agency. KC, which
she considers to be the supreme principle of morality, requires that she
maximally promote these conditions. One of the conditions necessary for
rational agency is life. After much re¬‚ection, the defender concludes that
the only way to save two people is to kill one innocent person. From duty,
the defender conforms to KC and kills the innocent person. Despite the
defender™s efforts, however, the other two are also killed. And there are no
other morally relevant effects “ for example, no one, not even the defender
herself, is inspired by her action to strengthen a commitment to conforming
to KC. In this scenario, the defender would have to deny that her own action
had positive moral value (moral worth), thereby contradicting step 1. For
the action did not at all succeed in promoting the good; it did not secure
the conditions necessary for rational agency.
In response, Cummiskey would, perhaps, insist that the defender may
claim her action to have moral worth even if it does not actually secure the
conditions necessary for rational agency. She may claim that moral worth
is intrinsic to the action. But I do not see how she may do so coherently.
She has identi¬ed the good ¬rst with rational nature and second with the
realization of the objects of rational nature (i.e., happiness). Her action “
her killing from duty one innocent to save others “ has succeeded not at
all in promoting the good in either sense. It has not helped to secure the
conditions necessary for rational agency, and it has not helped to secure the
realization of rationally chosen ends. So the defender ¬nds herself with no
basis on which to conclude her action to have been good.


7.8 Against a Principle Akin to the Ten Commandments
In the preceding sections, we have explored an argument Kant might em-
ploy to eliminate consequentialist candidates for the supreme principle of
morality. Yet not all rivals to his principle are consequentialist. In this section
and the next, we examine how he might eliminate some nonconsequen-
tialist candidates. Kant™s successfully excluding these candidates would not
itself give him warrant to conclude that no nonconsequentialist rival to the
Categorical Imperative remains. In my view, Kant offers no plausible way
of guaranteeing that his arguments would be effective against all possible
nonconsequentialist principles.
Let us begin with a principle somewhat akin to the Ten Commandments.
Why, in Kant™s view, couldn™t the following conjunctive principle be the
supreme principle of morality?

TC: You ought to honor your father and mother; you ought not to kill;
you ought not to commit adultery; you ought not to steal; you ought
not to bear false witness; you ought not to covet anything that is your
neighbor™s.
Search for the Supreme Principle of Morality

<<

. 5
( 8)



>>