<<

. 6
( 8)



>>

156

To simplify matters, let us view TC in a detheologized way, as a conjunctive
prescription “legislated” by individuals to themselves. Let us further suppose
that TC is a categorical imperative in the sense of a principle that sets out a
prescription to all rational agents regarding what they ought to do, regard-
less of what they might be inclined to do.26 Obviously, a proponent of TC
might conceive of morally permissible actions as ones that conform to TC
and morally impermissible actions as ones that do not. Moreover, a propo-
nent of TC might hold that an agent acts from duty when he wills to conform
to TC just because, in his view, TC requires that he do so. She might further
hold that any action done from duty has moral worth, regardless of its ef-
fects. (If from duty someone does his best to honor his parents, his action has
moral worth “ even if, through some unforeseen chain of events, he ends up
dishonoring them.) Mirroring Kant, the proponent of TC might conceive
of willing from duty to obey TC (in her view, a good will) to be good without
quali¬cation. This principle seems to ful¬ll much of Kant™s basic concept of
the supreme principle of morality. TC (or a principle quite like it) could be
practical, absolutely necessary, and binding on all rational agents.
Moreover, several of the further criteria Kant develops for the supreme
principle of morality do not seem to serve as a basis for rejecting TC. The
valuational argument can be successful against a particular principle only
if the principle™s advocate must hold that the moral value of willing from
duty to conform to it depends on the effects of doing so. But an advocate
of TC need not hold this. Equally unpromising as a response to TC would
be to insist that it is a material principle, and, therefore, it could not be
the supreme principle of morality. For Kant has given us no good reason
to think that TC is such a principle “ that an agent has suf¬cient grounds
to conform to it only if he expects doing so will enable him to realize some
object he desires (and/or have a hedonic payoff). One might appeal to
Kant™s notion that the supreme principle of morality must generate a set
of duties endorsed by ordinary moral reason, arguing that TC leaves some
important ones out, for example, that to promote others™ welfare. But this
tactic would be ineffective since the list of duties in TC could simply be
expanded. TC and principles like it seem to pose a particular challenge to
the possibility of a successful derivation of the Categorical Imperative.
Kant does, however, have at his disposal grounds for rejecting TC. It be-
longs to Kant™s basic concept of the supreme principle of morality that it
serve as the justi¬catory basis for all duties (section i.2). The principle must,
therefore, exhibit the reason why we have a duty to do a certain thing, yet we
do not have a duty to do something else. To some extent, TC accomplishes
this task. To the question of why x (e.g., telling the truth) is a duty, a pro-
ponent of TC can respond: because x is among the prescriptions conjoined
in the supreme principle of morality. Yet Kant would, I think, insist that this
is not enough. How, he would ask, would the proponent answer the ques-
tion of why this particular prescription is incorporated into the principle but
Rivals to the Categorical Imperative 157

another one (e.g., worship Amon) is not? Of course, from Kant™s perspective
it wouldn™t suf¬ce for her to say that the list is a product of present social
conditions “ merely the re¬‚ection of the values of a particular place and
time. To say this would be to offer an explanation but not a justi¬cation of
particular duties contained in TC. Kant implies that the supreme principle
of morality must provide a justi¬catory rationale for the duties derived from
it. As we will see in the next chapter, Kant™s own candidates do (though one
might disagree with this rationale). TC, it appears, does not really provide a
justi¬catory rationale for the duties derived from it. Therefore, TC cannot
be the supreme principle of morality.
Of course, this response would not give an answer to someone who re-
jected the notion that the supreme principle of morality need provide a
principled method (in Kant™s sense) of enumerating duties. One brand of
rational intuitionism, for example, might hold that we immediately grasp
TC, incorporating just these duties, and that is all there is to it.
There is another reason Kant might give for rejecting TC as a viable can-
didate for the supreme principle. For Kant if one can rule out the possibility
that a candidate for the supreme principle is knowable a priori, then this
candidate is not viable. As we saw earlier (section 4.10), Kant holds that only
if we can plausibly hold that a candidate is justi¬able a priori could we have
good reason to hold that it conforms to the basic concept of the supreme
principle, according to which this principle must be absolutely necessary.
Kant maintains that it is plausible to hold that the Categorical Imperative
is justi¬able a priori. In Groundwork III, he attempts to provide an a priori
justi¬cation of the Formula of Universal Law (or something quite like it),
appealing to the essential character of freedom, causality, rational willing,
and so forth, rather than to the experiences of particular individuals or cul-
tures. Kant might claim that it would be hopeless from the outset to attempt
to provide an a priori justi¬cation of TC. How, he might ask, could one
make a sincere attempt to justify TC as the supreme principle of morality
without appealing to the notion that, in the past, human beings have found
committing adultery to be wrong, honoring their parents to be required,
and so forth?
It would, I think, be misguided to react to this question by making the fol-
lowing claim: “Kant™s situation is no better, for he also relies on experience
to justify the Categorical Imperative, since for him a condition of success
for this principle™s derivation is that the duties the principle generates co-
here with ordinary moral reason.” For this claim neglects the distinction
between the derivation of the Categorical Imperative and its deduction. As
I suggested earlier (section 4.10), that a successful derivation of the Cate-
gorical Imperative must be grounded in experience does not entail that a
deduction of it must be as well.
Once again, though, an opponent might respond to Kant™s question by
asserting that we know TC through rational intuition. Our reason enables
Search for the Supreme Principle of Morality
158

us to recognize immediately that TC is the supreme principle of morality,
she might say. To me this response seems implausible, but I do not think
that Kant demonstrates it to be indefensible.


7.9 Further Nonconsequentialist Rivals
Chapter 2 focused on Henry Allison™s claim that if we grant Kant the as-
sumption that rational agents have transcendental freedom, then Kant can
offer a successful derivation of the Formula of Universal Law. Allison recon-
structs a derivation of this formula that in his view achieves its aim “ that
is, establishes that if there is a supreme principle of morality, then it is this
formula. Key to Allison™s reconstruction is the notion that only the Formula
of Universal Law (or, presumably, equivalent principles) is capable of justi-
fying the maxims of transcendentally free rational agents. I challenged this
notion, arguing that Kant fails to eliminate the possibility that some other
principle plays this role. In effect, I contended that Allison™s reconstructed
derivation does not eliminate certain rival candidates for the supreme prin-
ciple of morality. One rival was the “bizarre principle” BP: “Act only on that
maxim that you cannot, at the same time, will that it become a universal law”;
the other rival was WU: “Act only on that maxim which, when generalized,
could be a universal law.” On the criterial reading I have advocated, does
Kant have the resources to eliminate these candidates?
According to criterion viii, the supreme principle of morality must be such
that a plausible set of duties (relative to ordinary rational moral cognition)
can be derived from it. BP clearly fails to ful¬ll this criterion. According to
it, an agent™s acting on the following maxim would be morally impermissible :
“From self-love, during my free time, I will exercise in order to stay in
shape.” According to BP, let us specify, willing the universalization of a maxim
amounts to willing a world in which each agent adopts the maxim and if
the circumstances described in the maxim arise, he or she acts on it. It is
(rationally speaking) possible for an agent to act on this maxim and will its
universalization. First, there is nothing incoherent in the agent™s imagining
the world of the universalized maxim, so it is not the case that it is irrational
to will it on the grounds that it is irrational to will the impossible. Second,
in willing that each agent adopt the maxim, and if he has any free time, acts
on it, the agent would not be exhibiting the practical irrationality of under-
mining her own capacity to attain her end of staying in shape. All others™
exercising during their free time to stay in shape would not preclude her
from exercising in her free time and thereby staying in shape herself. BP
entails not only that we must not act on this (apparently innocuous) exercis-
ing maxim, but that we are forbidden from acting on a maxim such as this:
“From duty, unless I am incapacitated I will devote time and/or money to
charity work in order to better the condition of fellow human beings.” For
Rivals to the Categorical Imperative 159

this maxim also fails the test implicit in BP. Clearly, BP does not generate a
set of duties amenable to ordinary moral reason.
Although WU does not have quite the counterintuitive implications of BP,
it also seems to fail to generate a set of duties acceptable to ordinary moral
reason. WU commands that we act only on maxims that, when generalized,
can be a universal law. Consider the maxim: “In order to promote my own
happiness, I will never help a stranger or mere acquaintance in need.” This
maxim would be generalized, let us specify, if in order to promote his or her
own happiness, each agent never helped a stranger or mere acquaintance
in need. But it could be a universal law that this occur. There is nothing
incoherent or self-contradictory in imagining it. Therefore, according to
WU, it would be morally permissible to act on the maxim in question. Yet
this result seems to clash with our ordinary moral consciousness, which
embraces at least a minimal duty of bene¬cence.
Of course, we cannot take it for granted that the Categorical Imperative
itself satis¬es Kant™s eighth criterion for the supreme principle of morality.
As we will see in the next chapter, it is doubtful whether the Formula of
Universal Law generates a set of duties acceptable to commonsense morality.


7.10 Summary
Kant has the materials at hand to argue (plausibly, in my view) that cer-
tain rivals to the Categorical Imperative are not viable candidates for the
supreme principle of morality. Each of these rivals, he can show, fails to ful-
¬ll the criteria he has developed for the supreme principle. Nevertheless, the
derivation remains incomplete. First, a full derivation would require Kant
to eliminate all (possible) rivals to the Categorical Imperative. But as far as
I can tell, Kant does not provide us with an effective method for insuring
that we have considered all rivals. (As we have seen, the method Kant suggests,
namely that of categorizing all possible rivals as material principles, is not
promising.) Therefore, I do not see how even those very well disposed to
Kant™s arguments could claim that he had actually proved there to be no
rival to the Categorical Imperative that could ful¬ll each of the criteria he
develops for the supreme principle.
However, it would be no small achievement for Kant to show that, un-
like the rivals we have discussed in this chapter, the Categorical Imperative
(either in the Formula of Humanity or the Formula of Universal Law) re-
mains as a viable candidate for the supreme principle of morality. To show
this, Kant must demonstrate that his candidate could ful¬ll all of his criteria
for this principle. His attempt to do this is the topic of Chapter 8.
8

Conclusion: Kant™s Candidates for the Supreme
Principle of Morality




8.1 Kant™s Candidates and Criteria for the Supreme
Principle of Morality
Kant™s derivation as I have interpreted it is in the ¬rst instance a derivation
of the Formula of Universal Law. Yet it is open to Kant to offer a derivation
of the Formula of Humanity using the same basic steps. After all, the rivals to
the latter formula (e.g., utilitarian principles) are also rivals to the former.
If, based on an appeal to criteria he develops for the supreme principle,
Kant succeeds in disqualifying the rivals we discussed in Chapter 7 to the
Formula of Universal Law, then, in effect, he also succeeds in eliminating
rivals to the Formula of Humanity.
Now an opponent might grant that Kant, through appeals to his crite-
ria, eliminates many rivals to his candidates for the supreme principle of
morality. But, the opponent might claim, this is a Pyrrhic victory; appeals
to Kant™s criteria would also dispose of Kant™s own formulas. Does Kant
have the resources to rebut this claim? Does each of his formulas remain
a viable candidate for the supreme principle of morality? This is the ques-
tion that this chapter addresses, although it does not attempt to answer it
thoroughly.
At the outset, it is once again helpful to have in view the criteria Kant
embraces for the supreme principle of morality. There are eight main ones;
four Kant incorporates into his basic concept of the supreme principle, the
other four he develops through analysis of ordinary moral thinking. Accord-
ing 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, 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 representing the principle as a

160
Conclusion: Kant™s Candidates 161

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.
The main issue before us is whether either the Formula of Universal Law
or the Formula of Humanity remains as a viable candidate for a principle
that ful¬lls the full set of criteria. A derivation of a principle does not aim
to show that it actually ful¬lls the entire set of criteria. For showing this
would require proving that the principle ful¬lls criteria ii and iii “ that it
is absolutely necessary and binding on all rational agents. It would involve
giving a deduction of the principle, an endeavor that is not our concern
here.
Here is how the chapter unfolds. In section 8.2, I argue that each for-
mula remains a viable candidate for ful¬lling criteria i“iii, and that, if we
are willing to modify iv slightly, each one also remains a viable candidate
for ful¬lling it. Neither of the formulas fails as a candidate for the supreme
principle of morality on the grounds that it could not satisfy Kant™s basic
concept of the supreme principle of morality (if we modify this concept
a bit). The next section (8.3) attempts to show that criteria v“vii are also
unproblematic. The bulk of the chapter concerns criterion viii. Is either
the Formula of Universal Law or the Formula of Humanity such that, if
it was actually binding, from it would stem duties acceptable to ordinary
moral consciousness? Sections 8.4“6 focus mainly on the Formula of Uni-
versal Law, 8.7“9 on the Formula of Humanity. A lengthy book could eas-
ily be devoted to the question of whether, if valid, these formulas would
generate duties that square with those we take ourselves to have. As stu-
dents of Kant are well aware, each formula presents thorny dif¬culties of
interpretation. So I am not able here to answer this question thoroughly.
I argue, however, that we have good reason to doubt whether the Formula
of Universal Law ful¬lls criterion viii. Therefore, we have good reason to
doubt whether this formula remains as a viable candidate for the supreme
principle of morality. The Formula of Humanity, I suggest, seems more
promising regarding criterion viii, although it leaves us with some troubling
concerns.
The two formulas, claims Kant, are representations of “the very same
law” (GMS 436). If, as it seems reasonable to assume, this claim implies that
the two would give rise to the same moral requirements, then this chapter
offers some evidence that it is incorrect. Unfortunately for its defenders, the
Formula of Universal Law does not seem to forbid acts of violence committed
for revenge, whereas the Formula of Humanity does. It is open to a champion
of the Formula of Humanity to take the Formula of Universal Law as a rival
and to try to eliminate it as a candidate for the supreme principle of morality
on the grounds that it would clearly fail to yield duties acceptable to ordinary
moral thinking.
Search for the Supreme Principle of Morality
162

8.2 Two Formulas and the Basic Concept of the Supreme
Principle of Morality
Neither the Formula of Universal Law nor the Formula of Humanity should
be eliminated as a candidate for the supreme principle of morality on the
basis of a discernible inability to ful¬ll criteria i“iii. We could act on account
of each one “ each is practical “ though, as we have noted regarding the
Formula of Universal Law, it is harder than Kant acknowledges to determine
which actions the principles require. Each formula could also be absolutely
necessary, that is, binding on all the agents within its scope, regardless of the
agents™ particular inclinations. Moreover, the scope of each could extend to
all rational agents. Despite its use of the term “humanity,” the Formula of
Humanity is not limited in scope to human beings. For, as we have noted,
“humanity” there refers to rational nature, that is, the capacity for rational
choice, a capacity inherent in all rational agents.
There is, however, a dif¬culty that arises in connection with criterion ii.
For the sake of simplicity, in explaining it I focus on us, human agents,
bracketing other rational agents, and I use the generic term “Categorical
Imperative” to refer to both the Formula of Universal Law and the Formula
of Humanity, since no differences between the two formulas will come into
play. To say that the supreme principle of morality is absolutely necessary
is to say that without possible exception we ought to conform to it. And it
indeed does seem that the Categorical Imperative could be such that without
possible exception we ought to conform to it. The dif¬culty is not with the
possibility of the Categorical Imperative™s ful¬lling ii, but with something
that would result if it did ful¬ll ii.
The dif¬culty arises against the background of Chapter 6. There I argued
that Kant needs to acknowledge that even with the best intentions and effort
an agent might not only fail to apply the Categorical Imperative correctly
but might even embrace a rival as the supreme principle of morality. Take
someone who has done the latter, Stram the utilitarian from section 6.6.
For the Categorical Imperative to be a viable candidate for the supreme
principle of morality, it must be at least possible that Stram ought always to
abide by it, even though he often fails to do so. Yet, and here the dif¬culty
emerges, it seems that Kant must deny this. Kant embraces as an axiom that
ought implies can. According to him, I believe, this axiom entails that if an
agent is obligated to conform to a principle, then he must have an incentive
to conform to it. If an agent does not have an incentive to conform to
a principle, then she will not be able to do so, since for Kant all action
requires an incentive (Rel 35, English ed. 30).
However, it appears that at some points Stram might not have an incentive
to conform to the Categorical Imperative. He has done his best to determine
what his duty is, yet has adopted a principle that in particular cases in his
life clashes with the Categorical Imperative. In accordance with a Kantian
Conclusion: Kant™s Candidates 163

theory of agency, let us suppose that Stram always has an incentive to do
what he takes to be morally required. (Of course, his having an incentive to
do something does not, on this Kantian theory, entail that he will do it. For
no incentive can determine an agent™s will unless he has incorporated it into
his maxim. And instead of the moral incentive, he might choose to incorpo-
rate into his maxim some inclination.) Even if we suppose that Stram always
has this incentive, he would, it seems, sometimes fail to have an incentive to
do what the Categorical Imperative requires, since doing what this imper-
ative requires would sometimes amount to doing just the opposite of what
he thinks he is morally obligated to do. For example, he takes himself to
be required to lie in certain circumstances, but the Categorical Imperative
entails that he has a duty not to lie in these circumstances. If we want to
maintain that Stram nevertheless ought to (has a duty to) abide by the Cate-
gorical Imperative, then we must deny Kant™s notion that ought implies can.
In light of Chapter 6, it seems that for Kant maintaining that the Categorical
Imperative is absolutely necessary would require him to abandon a notion
he holds near and dear. What might make matters seem even worse is that,
on my reading, Kant appeals to this very notion in his defense of criterion
vii (see section 5.7).
In response, I do not see any great harm in Kant™s abandoning the notion
that if an agent has a moral duty to do something, he must have an incentive
to do it. First, it does not strike me as implausible to maintain the following.
A morally re¬‚ective agent who, since he did not believe it to be his duty
to do something, did not have an incentive to do it nevertheless morally
ought to have done it. (It might, however, be implausible to blame the
person for failing to do what he ought to have done.) Second, though it is
true that one argument for criterion vii appeals to the “ought implies can”
notion in question, Kant has another argument at his disposal that does not
(see section 5.7).1 Third, Kant™s moving away from the notion in question
would actually be a far less radical departure for him than it might seem. In
Chapter 6, we came across the following passage:
[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 seems to acknowledge that without being led astray by her in-
clinations, an agent can make an error in determining whether she has a
duty to do something. But in order for the notion of her making such a mis-
take to make sense, there must be a correct answer to the question of what
her duties are. It must be possible that though an agent does not believe
Search for the Supreme Principle of Morality
164

she has a duty to do something, she actually does. Let us suppose that the
following is such a case. An agent is convinced that being truthful to the
police and telling them the whereabouts of an innocent person they intend
to jail is morally forbidden, when it is actually morally required. Kant must
admit that the agent might not have an incentive to abide by what is morally
required, that is, to abide by her duty. After all, why should she have one?
She thinks that in this case being truthful to the police is morally forbidden.
If Kant here invoked the notion that an agent does not have a duty to do
something unless she has an incentive to do it, he would have to conclude
that, actually, the agent does not have a duty to be truthful to the police. But
this would contradict the assumption with which we began, namely that, as
a matter of fact, she does have such a duty. In short, it would be dif¬cult for
Kant to cling to the notion that if an agent has a duty to do something, he
must have an incentive to do it, all the while acknowledging, as he seems
to in the Metaphysics of Morals, that an agent can be mistaken about what
her duties really are. It appears that in making this acknowledgment, Kant
himself is, at least implicitly, moving away from the view that ought implies
can (interpreted in the particular way in question).
At any rate, returning to the essential point at hand, the Formula of
Universal Law and the Formula of Humanity remain as viable candidates
for principles that ful¬ll criteria i“iii. However, the last criterion in Kant™s
basic concept of the supreme principle of morality, iv, poses a problem. For a
principle to be in Kant™s sense the supreme norm for the moral evaluation of
action, every action™s moral permissibility, moral requiredness, and moral
worth must be de¬ned in terms of it. It is possible to de¬ne the moral
permissibility or moral requiredness of any action ultimately in terms of the
Formula of Humanity or the Formula of Universal Law. To focus on the
former, if in performing an action an agent treats humanity in herself and
others as an end, the action is morally permissible. If in refraining from
performing an available action the agent would not be treating humanity
in herself and others as an end, then the action is morally required. There
seem to be no actions the moral permissibility or requiredness of which
could not be “covered” by either one of these principles, although it might
not be a simple matter to determine how the action is covered “ that is,
whether it is permissible or required.
But what about moral worth? Granted, there is nothing within either
principle itself that would preclude our de¬ning moral worth with refer-
ence to it. We might, for example, hold that a necessary condition for an
action™s having moral worth is that it not con¬‚ict with what the Formula of
Universal Law requires. In effect, this seems to be Kant™s own position. If the
argument of Chapter 6 has been successful, however, we can see that this is
a problematic view for Kant to hold. He needs to acknowledge that some
actions not in accordance with this formula, namely those done from duty,
have moral worth. In my view, neither the Formula of Universal Law nor the
Conclusion: Kant™s Candidates 165

Formula of Humanity, nor, for that matter any other principle, is suited to
be the (sole) principle in reference to which all morally valuable action is
de¬ned. In acting from duty (and thereby ful¬lling the four Kantian condi-
tions on such action speci¬ed in section 6.9), agents can be acting on various
different principles, ones that clash with Kant™s as well as with one another.
Nevertheless, all of these actions have moral worth, or so Kant should grant.
If this is correct, then Kant must either conclude that neither of his own
candidates for the supreme principle of morality satis¬es his basic concept
of this principle, or alter his basic concept. The latter course clearly seems
preferable. From now on, we will understand Kant to hold that the supreme
principle of morality must be the supreme norm for the evaluation of the
moral permissibility and requiredness of an action, but not of its moral value.
Kant™s Formula of Universal Law and Formula of Humanity remain viable
candidates for principles that satisfy criterion iv understood in this way.
Of course, we are in no position to contend that Kant™s formulas actu-
ally do meet his basic concept of the supreme principle of morality. For we
have not shown (nor will we show) that either one is binding on all rational
agents. However, we can see that Kant™s formulas remain as viable candi-
dates for principles that realize this basic concept (if we modify the concept
slightly).


8.3 Two Formulas and Further Criteria
Do Kant™s formulas also stand as ones that we can reasonably maintain might
ful¬ll the other four criteria he develops?
According to criteria v and vi, the supreme principle of morality must
be such that each case of willing from duty to conform to it has moral
worth “ worth that does not stem from the willing™s results. The Formula
of Universal Law stands as a viable candidate for ful¬lling these criteria,
since its defender can coherently claim that each case of willing from duty
to conform to it has moral worth, regardless of its results. For the defender
of this formula obviously need not identify the good with anything external
to willing, such as the general happiness, and thus need not hold the value
of willing to depend on anything external, such as its effects on the general
happiness. In Chapter 6 I defended the claim that all acting from duty has
moral worth “ worth that does not stem from the willing™s effects. So in my
view, Kant™s Formula of Universal Law actually does ful¬ll criteria v and vi.
Whether the Formula of Humanity stands as a viable candidate for ful¬ll-
ing (let alone ful¬lls) v and vi might seem to be more questionable. Recall
that this formula reads: “So act that you treat humanity, whether in your
own person or in the person of any other, always at the same time as an end,
never merely as a means” (GMS 429, emphasis omitted). As we will discuss,
an advocate of this principle holds rational nature to be unconditionally
and incomparably valuable. He judges the moral permissibility of an agent™s
Search for the Supreme Principle of Morality
166

action in terms of whether she treats rational nature as such. One might
think that he is thereby committed to denying v and vi. After all, would he
not be required to acknowledge that some actions from duty would fail to
have moral worth, speci¬cally those that had the effect of harming rational
nature?
I do not see why the advocate of the Formula of Humanity would be
required to acknowledge this. I have defended the view that every case of
acting from duty has moral worth. If an agent™s (in itself suf¬cient) incentive
for acting stems from the notion that the Formula of Humanity requires
the action (and she meets the other requirements for acting from duty
discussed in section 6.9), her action has moral worth. That worth is not at
all a function of her action™s results or even of its actually conforming to
what the Formula of Humanity requires. There is nothing incoherent in
holding this and, at the same time, holding humanity to be unconditionally
and incomparably valuable. That one takes humanity to be unconditionally
and incomparably good does not rationally compel him to take it to be
the only thing that is unconditionally good. An advocate of the Formula of
Humanity can consistently maintain, in accordance with criteria v and vi,
that every case of willing, from duty, to conform to this principle has moral
worth, regardless of what results from it.
(In fairness to [potential] opponents of Kant, I should remark that a par-
allel point could be made with regard to maintaining everyone™s happiness
to be unconditionally valuable. In Chapter 7 we discussed the 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. Now suppose
that an advocate of U™ holds everyone™s happiness to be unconditionally
valuable. That he holds this does not itself entail that he must hold it to be
the only thing that is unconditionally valuable. Without contradiction, he
can also claim that acting from duty is unconditionally valuable. It is possible
that an advocate of U™ could coherently maintain, in accordance with v and
vi, that every case of willing, from duty, to conform to U™ has moral worth,
regardless of what results from it. What I tried to show in section 7.4 is that
a typical advocate of U™ cannot coherently maintain this; for a typical advo-
cate embraces Outcome Utilitarianism, the view [roughly] that goodness is
solely a function of well being.)
According to criterion vii, the supreme principle of morality must be such
that an agent™s representing it as a law provides him with suf¬cient incentive
to conform to it. Should the Formula of Universal Law and the Formula
of Humanity be disquali¬ed based on an inability to ful¬ll this criterion?
Opponents of Kant (e.g., Humeans) would be quick to suggest that no
principle could ful¬ll it, since the criterion itself rests on a mistaken theory
of agency. Sensuously based desire is a necessary ingredient in any incentive
for action, the opponents might say. And your representing a principle as
a law is not itself going to generate any desire to conform to the principle.
Conclusion: Kant™s Candidates 167

In this book, however, we have in effect assumed that the Kantian view of
agency is the correct one. This assumption would be question-begging if it
had been employed to eliminate non-Kantian candidates for the supreme
principle of morality. Yet it has not been used in this way. Not one rival
candidate has been dismissed on the basis of its failure to meet this criterion.
Actually, I have argued against Kant™s claim that all rival principles must be
understood to be material, and thus unable to ful¬ll (vii) (see section 7.2).
Nevertheless, the question remains as to whether Kant™s candidates could
ful¬ll this criterion. Assuming that sensuously based desire is not a necessary
ingredient in all incentives for action, I ¬nd no good reason to suppose that
Kant™s principles could not do so.2
The ¬nal criterion for the supreme principle of morality might pose
the greatest dif¬culty for Kant™s candidates. According to criterion viii, the
supreme principle must be such that, if it were binding on us, a plausi-
ble set of duties would stem from it, where “plausible” means in accord with
re¬‚ective moral common sense. Unless Kant™s formulas meet this criterion,
we must eliminate them as candidates for the supreme principle of moral-
ity. Much of the rest of the chapter is devoted to the question of whether
they do.


8.4 Two Formulas and Ordinary Moral Consciousness
A couple of observations will be helpful before we examine whether the
Formula of Universal Law and the Formula of Humanity would generate
prescriptions in accord with re¬‚ective moral common sense.
First, the project of examining whether these formulas ful¬ll criterion
viii faces a dif¬culty from the start. Contrary to what Kant implies, ordinary
moral thinking is not of a piece. Sincere, re¬‚ective people disagree about
what a person morally ought to do “ for example, when a gravely ill person™s
committing suicide would end her suffering and diminish that of her family.
My criticism of Kant™s notion that all actions from duty conform to it turns
on there being such disagreement. However, there seems to be widespread
agreement on some issues. The commonsense view, for example, seems to
be that making a false promise from the motive of ¬nancial gain is morally
wrong “ we have a duty to refrain from doing so. It is when a principle would,
if valid, fail to generate duties of this sort, ones that ordinary rational moral
cognition seems clearly to endorse, that we should reject the principle, or
at least that is how I interpret criterion viii.
Second, thorough assessment of whether Kant™s formulas would gener-
ate duties that accord with re¬‚ective moral common sense would require
thorough examination of precisely how best to interpret the formulas. The
latter task alone might call for a book-length treatment. For as anyone who
has taught the Groundwork is all too aware, Kant himself suggests various
different readings of these formulas, especially of the Formula of Universal
Search for the Supreme Principle of Morality
168

Law. In offering a brief assessment of the formulas, which does not aim to
be de¬nitive, I often rely on the interpretive work of others.


8.5 Formula of Universal Law: Practical Contradiction Interpretation
Does the Formula of Universal Law generate prescriptions acceptable to
common sense? To begin, let us suppose, as Kant quite reasonably does,
that according to common sense an agent ought not to make false promises
for his own ¬nancial gain; it is morally impermissible to do so. Since Kant
holds that all acting is acting on a maxim, if the Formula of Universal Law is
to yield results consistent with common sense, a maxim of false promising for
one™s own ¬nancial gain must fail the test contained in this formula. Kant,
of course, holds that such a maxim does fail, and thus that we have a duty
not to act on it (GMS 422).3 Philosophers have offered various accounts of
precisely how the maxim fails the test, but I explore only two of them here.
(As I indicated earlier, I simply assume that each of these two accounts is
permitted by Kant™s texts.)
According to Korsgaard, on the most philosophically plausible reading,
an agent cannot act on the sort of false promising maxim in question and
at the same time will that it become a universal law because doing so would
generate a “practical contradiction.”4 To see how it would, we need ¬rst to
note that, as Kant indicates (GMS 422), the maxim of the action would be
something like FPM, “From self-love, when I believe myself to be in need of
money I shall borrow money on a promise to repay it, even though I know
that this will never happen.” How would we describe a world in which this
maxim would be a universal law, that is, the maxim™s “universalization”? On
the interpretation Korsgaard advocates, the Practical Contradiction Inter-
pretation, we would say that in this world the following obtains: from self
love, when anyone believes himself to be in need of money, he tries to borrow
money on a promise to repay it, even though he knows that this will never
happen.5 What would be contradictory in the agent™s acting on FPM and,
at the same time, willing the world in question? Imagine that she is doing
this. First, since she is acting on FPM, the agent is trying, through the means
of making a false promise, to attain her end of getting money. Second, in
willing the world of FPM™s universalization, she is willing a world in which
taking these means will not enable her to attain her end. For if each person
in ¬nancial need tries to get money on a promise of repayment (and, if she
succeeds, does not in fact repay), then potential lenders will not lend money
simply on a promise to repay. It will not be possible using a promise alone “
in contrast, for example, to some kind of written contract “ for a person
in ¬nancial need to get money.6 So the agent is trying through a particular
means to attain an end and at the same time willing a situation in which
it is impossible through this means to attain the end. In effect, the agent
is willing that he be thwarted in attaining the end he is pursuing. Therein
Conclusion: Kant™s Candidates 169

lies a practical contradiction.7 Therefore, insofar as he is rational, an agent
cannot act on FPM and at the same time will that it become a universal law.
Kant holds that in acting on FPM an agent would be violating the For-
mula of Universal Law and that, therefore, we have a duty not to act on
this maxim. The Practical Contradiction Interpretation offers a way of un-
derstanding how, precisely, acting on FPM would violate this formula. In
short, the maxim would fail because the agent™s attaining the end it speci-
¬es (getting money) through the means it speci¬es (making a false promise)
depends on most agents™ not taking this means to the end.8 The maxim™s
effectiveness would be a function of its being exceptional. The Practical
Contradiction Interpretation allows us to see that as far as a maxim such
as FPM is concerned, the Formula of Universal Law generates results that
cohere with ordinary moral thinking.
This interpretation, however, also generates results that clash with ordi-
nary moral thinking. Following Barbara Herman, suppose that an agent acts
on the maxim: “From self-love, I will shop in this year™s after-Christmas dis-
counts for next year™s Christmas presents in order to save money.”9 If every-
one acted on this maxim (i.e., if it were universalized), then after-Christmas
discounts would disappear “ they would be too damaging to pre-Christmas
income. In willing the world of his universalized maxim, the agent would be
willing a world in which it would not be possible to save money by means of
shopping in this year™s after-Christmas discounts for next year™s Christmas
presents. It would be irrational for the agent to will this world and at the
same time act on his maxim of saving money through taking this very means.
For in willing this world, he would be willing to be thwarted in his pursuit of
his end. If the maxim of false promising generates a practical contradiction,
then so does this maxim of economical shopping. The effectiveness of acting
on either of them is a function of its being exceptional that people do so. Al-
though common sense would condemn as morally impermissible acting on
the maxim of false promising, it would not condemn as such acting on the
maxim of economical shopping. There just does not seem to be anything
morally wrong with taking advantage of after-Christmas sales in a way that is
effective only against the background of others not trying to take advantage
of them in this way. If the Formula of Universal Law says otherwise, then so
much the worse for it “ in particular for its prospects of ful¬lling criterion
viii for the supreme principle of morality.
The discussion of maxims in section 1.3 laid the groundwork for a possi-
ble reply to this objection. On the view I adopted, the maxim of an agent™s
action is the most general rule of the proper form on which he acts. If this
view is correct, it seems unlikely that the agent™s rule of economical shopping
really counts as his maxim; for this rule seems too speci¬c. Is it not likely that
the rule is ancillary to (i.e., serves as a means of executing) a maxim such
as “From self-love, I will shop at sales in order to save money”? If so, then
(arguably) no practical contradiction is generated by the agent™s acting on
Search for the Supreme Principle of Morality
170

his maxim and at the same time willing that it become a universal law. In
willing the universalization of his maxim, the agent would not (arguably)
be preventing himself from attaining his end through the means speci¬ed
in his maxim.
Is this response effective? In acting on the very speci¬c rule of shopping
at this year™s after-Christmas sales for next year™s gifts, the agent might not
actually be implementing a more general rule, which would count as her
maxim. The very speci¬c rule might just be her maxim.10 It would be strange
if it were since it is hard to see how someone would think to adopt the rule
if not in the context of carrying out some general policy of trying to save
money in her shopping. But the strange is far from the impossible. If this
very speci¬c rule were her maxim, it would, I think, be contrary to ordinary
moral consciousness to claim it to be morally impermissible for the agent
to act on it. Yet that is what a defender of the Formula of Universal Law
would have to do, at least on the interpretation of it that we have been
employing.
Moreover, not all rules that, contrary to ordinary moral reason, fail the
Formula of Universal Law test on this interpretation are so speci¬c that we
would question them as examples of maxims. Suppose that Jack, the son
of dock workers, acts on the following rule: “In order to earn a comfort-
able living, I will become a professor, rather than do physical labor.” (For
Jack making a comfortable living amounts to making enough to have his
own house, car, computer, and so forth.) There would be nothing odd if,
in Jack™s case, this rule were not ancillary to a more general one. Let us,
then, assume that the rule is his maxim.11 On the Practical Contradiction
Interpretation of the Formula of Universal Law, this maxim turns out to
be morally impermissible. In willing a world in which everyone acted on
his maxim, Jack would be willing the ineffectiveness of the means he takes
(becoming a professor) to his end (earning a comfortable living). For in
this world the institutional framework for salaried professors would, in all
likelihood, not be in place. Universities do not function without support
from people who earn a living through physical labor.12 Some maxims “
for example, Jack™s as well as the false-promising maxim (FPM) “ specify a
means that is effective for attaining their end only in a context in which it
is exceptional for agents to take this means to the end. These maxims take
advantage of predictable regularities in agents™ behavior. On the Practical
Contradiction Interpretation, the Formula of Universal Law condemns as
morally impermissible all acting on such maxims. But this condemnation
clashes with commonsense morality, according to which acting on some of
these maxims (e.g., Jack™s) is not contrary to duty.
Of course, the Practical Contradiction Interpretation is not the only
reading of how Kant envisages (or might envisage) that a maxim of false
promising would fail the Formula of Universal Law test. Perhaps there is an
Conclusion: Kant™s Candidates 171

alternative reading according to which FPM would fail, but other maxims,
ones such as Jack™s which we take to be morally permissible, would pass.


8.6 Formula of Universal Law: Universal Availability Interpretation
Thomas Pogge has presented an alternative that might seem to secure these
results.13 The main feature that distinguishes Pogge™s reading from the one
we have already considered is its account of how a maxim is to be univer-
salized. On the Practical Contradiction Interpretation, imagining a maxim
to be a universal law amounts to imagining a world just like ours except
that everyone has actually adopted the maxim (and acts on it when occa-
sions arise). On Pogge™s reading, imagining a maxim to be a universal law
amounts to imagining a world just like ours except that everyone believes
himself to be permitted (i.e., “morally” free) to adopt the maxim, and those
who are inclined to adopt it do so (and act on it when occasions arise).14 In
light of this difference between the Practical Contradiction Interpretation
and Pogge™s reading, I call the latter the Universal Availability Interpreta-
tion. According to the Universal Availability Interpretation of the Formula
of Universal Law, an agent is to ask herself whether she can act on a maxim
and at the same time will (in short) that everyone hold the maxim to be
available, in the sense of morally acceptable.
The Universal Availability Interpretation saves the Formula of Universal
Law from yielding the result so unwelcome to common sense that it is morally
impermissible to act on the maxim of earning a comfortable living by be-
coming a professor or that of economizing by shopping at this year™s sales for
next year™s gifts. For purposes of illustration, I will just consider once again
the former maxim, held by Jack: “In order to earn a comfortable living, I
will become a professor, rather than do physical labor.” Jack could act on
this maxim and at the same time will that everyone feel (morally) free to act
on it. In willing a world in which everyone did feel this way, Jack would not
be rendering ineffective the means speci¬ed in his maxim for attaining his
end of earning a comfortable living. If the moral availability of this maxim
resulted in a mass rush to graduate school of those aiming at a comfortable
living, then he would be thwarting this means. But surely such a rush would
not occur. For it is not any moral qualms about Jack™s maxim that stand in
the way of masses of people adopting it but rather things like inclinations to
take different means “ for example, ones that require less time in the library
or laboratory “ to the end of earning a comfortable living. The Universal
Availability Interpretation has the advantage over the Practical Contradic-
tion Interpretation of allowing the Formula of Universal Law to grant the
moral permissibility of some maxims that, though they depend for their
effectiveness on being exceptional, are not condemned by ordinary moral
reason.
Search for the Supreme Principle of Morality
172

Perhaps, however, this advantage comes at too high a price. For it is
questionable whether, on the Universal Availability Interpretation, false-
promising maxims turn out to be morally impermissible.
Pogge suggests that his interpretation does generate the desired results
regarding such maxims. The example of a false-promising maxim Pogge
considers is a bit more general than the one we have thus far discussed;
it is “When in need, I will make deceitful promises so as to alleviate my
dif¬culties.”15 On his reading, in the world of the universalized maxim ev-
eryone would feel (morally) free, when in need, to make deceitful promises
so as to alleviate his dif¬culties.16 According to Pogge, the false-promising
maxim would be “pointless” in this world; for acting on it would not alleviate
one™s dif¬culties.17 That it would be pointless, he continues, “leads to the
rejection of that maxim, because . . . its universal availability would block the
agent™s attainment of the material end of his conduct under the maxim.
And with the objective out of reach, the agent cannot will the maxim: If it
cannot satisfy his interest in its material end, the agent loses his only possi-
ble (heteronomous) motive for adopting it.”18 If his acting on a maxim is to
pass the Formula of Universal Law test, an agent must (rationally speaking)
be able to act on it in the world in which the maxim has been universalized,
suggests Pogge. But in the world of the universalized false-promising maxim,
the agent could not act on his maxim. For, as the agent would realize, act-
ing on it would do nothing to enable him to secure his end of getting out
of dif¬culties. Therefore, the agent (insofar as he was rational) would ¬nd
himself with no motive to adopt his maxim. In this sense, he could not act
on it. So the false-promising maxim turns out to be morally impermissible.
On the Universal Availability Interpretation, the maxim™s turning out
this way depends on its being the case that in a world where everyone felt
(morally) free to act on the maxim, it would be “pointless” for a particular
individual to act on it. But is this really the case? According to Pogge, “people
in need would (be known to) have no reason not to make deceitful promises”
and “potential promisees would (be known to) have good reason to reject
promises made by persons in need.”19 I think the ¬rst point is questionable.
Granted, in the world in which everyone feels morally free to act on the
maxim of false promising, people in need would (be known to) have no
moral reason not to make deceitful promises. However, this does not entail
that they would (be known to) have no reason at all not to make such
promises. For there are prudential reasons not to make deceitful promises,
even when one is in dif¬culties. For example, an agent might judge that
the sanctions he would incur if it were to become known that he made a
deceitful promise would be worse than his present dif¬culties and that the
chances of its becoming known are great enough to render it not worth
the risk for him to make the deceitful promise. The penalties in question
might range from prison time if, for example, the deceitful promise was
that his home remedy would cure cancer, to an inability without collateral
Conclusion: Kant™s Candidates 173

ever to again obtain money from his family if, for example, the deceitful
promise was that he would pay back a loan from his uncle. (The notion that,
if found out, the agent would incur these penalties is compatible with no
one™s holding it to be morally wrong to make deceitful promises in order
to alleviate one™s dif¬culties. Members of the agent™s family, for example,
might refuse to lend him any more money [in the absence of collateral]
not at all on the basis that his behavior was morally bankrupt but simply
because they do not want to lose any more money.) Contrary to Pogge™s
¬rst point, in the world of the universalized maxim, people in need would
sometimes have reason to refrain from making deceitful promises; it would
be prudential rather than moral reason.
If the ¬rst point is questionable, then so is the second. In the world of the
universalized maxim, would potential promisees always (or even the great
majority of the time) have good reason to reject promises made by persons
in need? Let™s say that in the imagined world someone asks you to loan him
money on the basis of a promise that he will pay it back. You would have good
reason to reject his proposal if you (reasonably) believed that, in his view,
his breaking the promise would not result in any signi¬cant penalty for him.
(You might reasonably believe this if he is a stranger who probably does not
think he will ever see you again). But you would have good reason to accept
it if you (reasonably) thought that in his view his breaking the promise would
hurt him a great deal. (You might reasonably believe this if he were a young
business associate who depended on you for his climb up the corporate
latter.) In the imagined world, it is not clear that in acting on the maxim of
deceitful promising, an agent would be employing an obviously ineffective
means to an end. Whether he would be depends (among other things) on
others™ perceptions of his prudential reasons for keeping his promise. In short,
it seems that sometimes an agent acting on the deceitful-promising maxim
could attain his end of alleviating his dif¬culties in a world in which everyone
felt morally free to act on this very maxim. So it is questionable whether on
the Universal Availability Interpretation the false-promising maxim actually
turns out to be morally impermissible.
There is another problem with the Universal Availability Interpretation.
Ordinary moral reason would, I venture, condemn as contrary to duty acting
on the maxim “If anyone commits adultery with my spouse, I will kill the
person in order to get revenge.” On the reading in question, however, the
Formula of Universal Law would not. An agent (insofar as she was rational)
could act on this maxim in a world in which everyone felt morally permitted
to do so as well. In this imagined world, perhaps those who did commit
adultery would take greater precautions than they do now to avoid contact
with the betrayed husband or wife. But that would not, for example, preclude
an agent in the imagined world from attaining her goal of getting revenge
through killing the woman who seduced her spouse; it would just make the
killing more dif¬cult.20 Perhaps it is partly because (on his interpretation)
Search for the Supreme Principle of Morality
174

the Formula of Universal Law licenses such maxims that Pogge does not
believe that on its own it constitutes a viable candidate for the supreme
principle of morality.21
In sum, on neither the Universal Availability Interpretation nor the Prac-
tical Contradiction Interpretation would the Formula of Universal Law, if it
were binding on us, generate duties that cohere with the dictates of ordi-
nary moral thinking. On the former interpretation, it would turn out that,
contrary to ordinary conviction, we have no duty to refrain from acting on
(certain) maxims of false promising and violence. On the latter, it would
turn out that, contrary to ordinary conviction, we have a duty to refrain
from acting on (certain) maxims of taking advantage of predictable regu-
larities in others™ behavior, maxims such as that of earning a comfortable
living by becoming a professor rather than by doing physical labor. At least
on two readings, the Formula of Universal Law does not ful¬ll criterion viii
for the supreme principle of morality.
It would, of course, be unwarranted to take this to show that the Formula
of Universal Law fails to ful¬ll criterion viii. Our discussion has not been
thorough enough to establish this conclusion. However, I do think that it
helps to con¬rm a suspicion expressed recently by several Kantians that
despite some ingenious efforts, no one has been able to make this formula
work.22 Perhaps someone will, but as Herman says, “past experience suggests
a permanent ¬x-it situation: the correction of one dif¬culty or apparent
oversight creates space for new problems to emerge.”23


8.7 Fundamentals of the Formula of Humanity
The prospects for the Formula of Universal Law™s generating a set of duties
acceptable to ordinary moral reason do not appear to be good. Are the
prospects for the Formula of Humanity any better? Kant himself seems to
favor the Formula of Humanity as a basis on which to derive duties. For
in the Metaphysics of Morals, Kant relies (at least implicitly) on this formula
to derive the vast majority of the ethical duties he sets out.24 But given
Kant™s suggestion that the two formulas are equivalent (GMS 436), perhaps
he favors the Formula of Humanity simply because in his view it is less
cumbersome to work with than the other formula. At any rate, I do not offer
anything approaching an exhaustive treatment of the issue of whether the
Formula of Humanity would generate a plausible set of duties relative to
ordinary moral thinking. However, I do hope to say enough to suggest that,
although the Formula of Humanity holds signi¬cant promise, defenders of
it must confront some troubling issues.
Before we can discuss the question of which duties would stem from this
formula (if it was valid), we need to understand the terms it employs. Unfor-
tunately, like the Formula of Universal Law, it is not easy to interpret. The
Formula of Humanity commands: “So act that you treat humanity, whether
Conclusion: Kant™s Candidates 175

in your own person or in the person of any other, always at the same time
as an end, never merely as a means.” An agent™s acting so that she treats
humanity (in herself or any other) as an end is a necessary and suf¬cient
condition for her conforming to the formula. It is a necessary condition
since the formula commands that an agent so act that she always treat hu-
manity as an end. It is a suf¬cient condition since even if the agent acts so
that she treats humanity in herself or another as a means, as long as she at
the same time acts so that she treats it as an end, she has conformed to the
formula.25 So, at bottom, the Formula of Humanity amounts to a command
so to act that we always treat humanity as an end.26 “Humanity,” let us recall,
does not refer to the class of human beings but rather to a set of capacities:
the capacities to set oneself ends and to adopt and act on rules, including
rules of prudence (hypothetical imperatives) and rules of morality (categor-
ical imperatives), often in pursuit of these ends. Would duties acceptable to
ordinary moral reason stem from the command always to treat humanity so
understood as an end?
An initial step toward answering this question is to examine the sense of
“end” or, equivalently, “end in itself” at work in the Formula of Humanity.27
Kant holds that humanity exists as an end in itself. But what does it mean for
humanity to exist in this way? First, as we know from our discussion of Kant™s
derivation of this formula, an end in itself is something that has absolute or
unconditional worth (GMS 428). It would be judged by an impartial rational
spectator to be good in every possible context, even in ones in which it
brought about undesired results. For Kant that an end in itself has absolute
worth implies that all rational agents must (are rationally compelled to)
value it and to act in ways that express their valuing it, regardless of whether
they are inclined to do so (section 3.2).
Second, to say that humanity exists as an end in itself is to say that it
has dignity (GMS 435; MS 434“435, 462). To have dignity, Kant suggests, is
to have “unconditional and incomparable worth” (GMS 436). We have just
noted what it means to have the ¬rst aspect of dignity, namely unconditional
worth. Kant explains the second aspect of dignity, namely incomparable
worth, by contrasting it with price: “What has a price can be replaced by
something else as its equivalent ; what on the other hand is raised above all
price and therefore admits of no equivalent has a dignity” (GMS 434; see
also MS 462). The value of something with dignity, then, is incomparable in
the sense that it has no equivalent for which it can be exchanged. As Thomas
Hill has argued, that it is seems to have two implications.28 First (and quite
clearly), something with dignity can never be legitimately sacri¬ced for or
replaced by something with price. Not even all the gold in Fort Knox would
compensate for the killing of one rational agent. Second (and not quite so
clearly), something with dignity cannot even be legitimately sacri¬ced for
or replaced by something else with dignity. Beings with dignity, says Kant,
admit of “no equivalent.” If, therefore, it is ever legitimate to kill one being
Search for the Supreme Principle of Morality
176

with dignity, thereby saving several other such beings, it will not, it seems, be
because it is legitimate to make an exchange of the (lesser) value inherent
in the former with the (greater) value inherent in the latter. An end in itself
(and thus humanity) has dignity in that it has an unconditional value that
admits of no equivalent, not in terms of price, nor, it appears, even in terms
of other beings with dignity.
We have found that to say that humanity is an end in itself is to imply that it
is something that has an absolute and incomparable worth. But presumably
if Kant calls humanity an end in itself, then in some sense he thinks of it
as an end. In what sense, precisely? This question is puzzling if one takes
as a point of departure Kant™s de¬nition of an end in the Metaphysics of
Morals. “An end,” he says, “is an object of the will [Willk¨ r] (of a rational
u
being), through the representation of which the will is determined to an
action to bring this object about” (MS 381; see also MS 384“385). In other
words, an end is a state of affairs or event such that an agent, through her
idea of it, is determined to will to realize it. An agent might, for example,
have as an end to maintain his weight under two hundred pounds for the
next six months or to win a tennis tournament. An end on this account is
a goal, aim, or target “ an object to be produced.29 Yet Kant suggests that
humanity is not an “end to be effected,” but rather an “independently existing
[selbstst¨ ndiger] end” (GMS 437; see also MS 442). So in calling humanity an
a
end in itself, he must have a broader notion of an end in view. Indeed, in the
Groundwork, immediately before his derivation of the Formula of Humanity,
Kant says that “an end is what serves the will as the objective ground of its
self-determination” (GMS 427). An end is an objective ground of an agent™s
determining his will to an action. An end is a ground in that it is a reason
that an agent has (or at least ought to have) for acting; an end is an objective
ground in that it is an object such that, through representing it to himself, the
agent gives himself (or at least ought to give himself) a reason for acting.30
On this broader conception, ends are not limited to objects to be produced.
They include any object the idea of which does (or ought to) give an agent
a reason to act in a certain way. Someone™s aim or goal of winning a tennis
tournament might count as such an object, but so might an existent object
such as her humanity. And Kant, of course, thinks that humanity, wherever
and whenever it manifests itself, counts as an end in this broad sense. It does
so by virtue of its being absolutely and incomparably valuable.
For Kant humanity exists as an end in itself, an unconditionally and in-
comparably valuable object the idea of which gives (or at least ought to give)
all rational agents a reason for acting. What does it mean to act so that one
always treats humanity as an end in itself, as the Formula of Humanity com-
mands? Presumably one acts so that one treats humanity as an end in itself
just in case what one wills to do is consistent with holding humanity to be
of absolute and incomparable value. In the Groundwork, Kant calls rational
nature (i.e., humanity) an “object of respect.” In the Metaphysics of Morals,
Conclusion: Kant™s Candidates 177

he suggests that any being with humanity must not only respect himself
but “exacts respect for himself from all other rational beings in the world”
(MS 435; see also MS 462). These comments suggest that an agent™s action
is consistent with his holding humanity to be of absolute and incomparable
value only if it manifests respect for humanity. Humanity is not an end to be
effected (produced), but it is an end to be respected. However, the question
remains: which actions are consistent with an agent™s holding humanity to
be an end in itself ? Might the Formula of Humanity be more effective than
the Formula of Universal Law at generating duties acceptable to ordinary
moral reason?


8.8 Deriving Duties from the Formula of Humanity
The Formula of Humanity would (if it was binding on us) seem to be effective
at engendering certain duties we take ourselves to have (e.g., a duty not to
kill for revenge). From the Formula of Humanity (unlike from the Formula
of Universal Law) it clearly follows that one must not act on a maxim of
killing an adulterer to get revenge; for murdering an adulterer, and thus
destroying his humanity, is obviously not consistent with respecting it as
something absolutely and incomparably valuable. In general, destroying
humanity would rarely, if ever, seem to express respect for it. And that is one
reason why philosophers ¬nd puzzling Kant™s strong advocacy of capital
punishment (see, e.g., MS 334). In any case, that killing to get revenge
is morally impermissible according to the Formula of Humanity, but not
according to the Formula of Universal Law (at least on the readings of it we
have discussed), suggests that, contrary to Kant™s view, the one formula is
not equivalent to the other.
A duty of bene¬cence would also seem to follow from the Formula of
Humanity, although its derivation is not without dif¬culties. How does Kant
arrive at this duty in the Groundwork (GMS 430)?31 Humanity as Kant un-
derstands it is a set of capacities, including the capacity to set ends and
pursue them. According to Kant, one end that each of us (i.e., each human
agent) sets and pursues is that of his own happiness.32 So, in concrete terms,
valuing our humanity (as opposed to that of other rational agents, such as
angels, who might not have their own happiness as an end) involves valuing
our capacity to pursue happiness. To conform to the Formula of Humanity,
then, an agent™s actions must be consistent with his valuing this capacity.
But it seems that one does not really value a capacity unless one values its
successful exercise. We would, for example, doubt whether someone truly
valued the capacity of an acorn to grow if she denied that, other things being
equal, it would be a good thing if it matured into an oak tree. However, this
is a tricky point. It does not seem to be self-contradictory to value a capacity
but not the “successful” exercise of it. Would there be anything irrational in
valuing an acorn™s capacity to grow but remaining indifferent as to whether it
Search for the Supreme Principle of Morality
178

just barely sprouted out of the ground or grew into a gigantic tree? Somewhat
analogously, would there be anything irrational in valuing a person™s capacity
to pursue happiness but remaining indifferent as to whether he made little
or much progress toward it?
Supposing that we grant that there would be something irrational in this,
we see that an agent™s actions are not consistent with his valuing humans™
capacity to pursue their happiness unless they are consistent with his valuing
their actually making progress toward happiness. Now an agent™s actions
are not consistent with this if he acts toward the goal of thwarting someone
else in his pursuit of happiness, if we assume the other™s pursuit is itself
consistent with the other™s appropriately valuing humanity. So an agent must
refrain from acting toward this goal. If he thus refrains, then, in Kant™s terms,
his actions express “negative agreement” with humanity as an end in itself
(GMS 430). According to Kant, however, the Formula of Humanity also
requires “positive agreement” with humanity as an end in itself. An agent
must also promote others™ happiness. The idea here is that an agent™s actions
are not really consistent with his valuing others™ progress toward happiness
unless he aids them in making it.
Of course, this account of how a duty of bene¬cence would derive from
the Formula of Humanity leaves important questions unanswered. For ex-
ample, how robust a duty of bene¬cence would follow from the formula?
Kant suggests that the formula requires that everyone try “as far as he can,
to further the ends of others. For, the ends of a subject who is an end in
itself must as far as possible be also my ends, if that representation is to have
its full effect in me” (GMS 430). However, Kant considers bene¬cence to be
an imperfect duty; and earlier in the Groundwork he characterizes a perfect
duty as “one that admits no exception in favor of inclination,” (GMS 421,
note) apparently implying that an imperfect duty does admit of such an
exception. If we must do all we can (morally permissibly do) to further the
ends of others, how can we ever justi¬ably choose to satisfy our own inclina-
tions (e.g., by watching a movie) instead of trying to promote the welfare of
others (e.g., by working a few hours at a soup kitchen)?33
Our brief discussion of the duty of bene¬cence has illustrated that deriv-
ing duties from the Formula of Humanity is not a cut-and-dried business.
Its dif¬culties will become more evident, I think, if we take a look at Kant™s
derivation of a duty of sincerity in promising.
According to Kant, the Formula of Humanity forbids an agent from mak-
ing promises that he has no intention of trying to keep. “[H]e who has it
in mind to make a false promise to others,” says Kant, “sees at once that
he wants to make use of another human being merely as a means, without
the other at the same time containing in himself the end. For, he whom I
want to use for my purposes by such a promise cannot possibly agree to my
way of behaving toward him, and so himself contain the end of this action”
(GMS 429“430). How are we to interpret this passage? According to Allen
Conclusion: Kant™s Candidates 179

Wood, Kant is arguing here that making a false promise would violate the
Formula of Humanity since it would express disrespect for rational nature.
“A false promise, because its end cannot be shared by the person to whom the promise
is made, frustrates or circumvents that person™s rational agency, and thereby
shows disrespect for it.”34 Apparently, according to Wood, when Kant says
that a promisee cannot “himself contain the end” of a false promisor™s ac-
tion, he is intimating that the latter cannot share the promisor™s end. (Here
“end” refers to an end to be produced.) That interpretation seems reason-
able enough.
But what, precisely, does it mean to say that the promisee cannot share
the promisor™s end? Wood is not very helpful on this point. The end that the
promisee cannot share is apparently the false promisor™s end to deceive him
into doing something (e.g., into giving him money), rather than the end for
the sake of which the false promisor tries to deceive him into doing this thing.
For the latter end might be that of diminishing world hunger, and there
seems to be no reason why it would be impossible for the two to share that
end. Perhaps in Kant™s view the promisee cannot share the promisor™s end of
deceiving him into doing something in the sense that it would be irrational
for him to share this end. Agents presumably share an end just in case each of
them pursues the end. But, in ordinary circumstances, it would be irrational
for the promisee to pursue the end of being deceived into doing something
such as lending someone money. For this end™s being brought about would
prevent him from attaining other ends he is pursuing “ for example, that of
eventually buying himself a car.35 The notion of irrationality at work here is
familiar to us from our discussion of the Formula of Universal Law. In effect,
if the promisee shared the false promisor™s end, then the former would be
willing that he be thwarted in attaining ends he is pursuing. In a practical
sense, he would be irrational.
We might, then, take from Kant™s discussion of false promising (GMS
429“430) that if an agent™s action involves another, it expresses disrespect
for the other™s agency (and thus violates the Formula of Humanity) unless
the other can share the agent™s end. And the other can share the agent™s
end only if the other can pursue it without practical irrationality of the kind
we have just described. To put the view brie¬‚y, a necessary condition for the
moral permissibility of actions affecting others is that they be done to attain
ends that others can share.
Unfortunately, there are dif¬culties with this view. First, suppose that Pete
acts on the maxim: “In order to be the number-one ranked men™s tennis
player of the year, I will win every major tournament I enter.” At ¬rst glance,
it does not seem to be morally impermissible to act on this maxim. However,
doing so might violate the Formula of Humanity as just interpreted. Acting
on this maxim might frustrate some rival player™s rational agency, thereby
showing disrespect for it. For presumably some rival players cannot share
Pete™s end. Imagine that Pete and Andre are competing in the ¬nal of the
Search for the Supreme Principle of Morality
180

U.S. Open and that at stake is the number-one ranking for the year, which
each player has as his goal. In pursuing the end of Pete™s being number
one “ for example, by purposefully throwing the match “ Andre would be
willing to be thwarted in attaining his end of being number one. Andre
cannot share Pete™s end in the sense that it would be practically irrational
for him to do so. In general terms, this reading of the Formula of Humanity
has the following implication. Suppose an agent is pursuing an end in a
competition. If his competitor cannot, rationally speaking, both pursue the
agent™s end and strive to secure his own end, the agent™s action is morally
impermissible.
One might respond that, although this implication initially seems to dis-
credit the Formula of Humanity, re¬‚ection reveals otherwise. Granted, it
would be worrisome if the Formula of Humanity entailed that pursuing an
end in competitive sports (or some other competitive endeavor) is always
wrong. But on the reading in question, the formula does not entail this. If
Pete™s end were not to be number one but to develop his capacities as a
tennis player, then he would not be disrespecting Andre™s agency. For this
is an end that Andre can share. (Of course, if Andre perseveres in pursuing
the end of being number one, then he is presumably violating the Formula
of Humanity by disrespecting the rational agency of some other player who
himself aims to be number one.) This reply has some force. According to
re¬‚ective moral common sense, it seems, Pete and Andre would in some
sense be more virtuous if each could share the other™s end. Many of us do
¬nd the character of competitors who each have as an end to develop their
own capacities morally more attractive than ones who each have as an end
to defeat their rivals. There is something admirable in holding that, ulti-
mately, one is “competing against” oneself. However, I think that ordinary
moral reason would ¬nd unacceptably strong the judgment that it is morally
wrong to act as Andre and Pete do in the example.
There is a second dif¬culty with the view that an agent™s doing something
to another expresses disrespect for the other (thus violating the Formula of
Humanity), unless the other could, without practical irrationality, share the
agent™s end. Suppose a police of¬cer has the end of preventing race-based
attacks on law-abiding citizens. In pursuing this end, she arrests a white
supremacist, someone she believes (correctly) to be planning an attack on a
preschool frequented by Asian Americans. The dif¬culty is that in arresting
the white supremacist she might be pursuing an end that he cannot share.
Suppose, as the of¬cer is aware, his end in planning the attack was to get
revenge on a racial group that he thinks to be inferior to whites and thus
undeserving of the rights and liberties its members possess. It would be
practically irrational for him to pursue his end of revenge and at the same
time to will the of¬cer™s end of preventing race-based attacks on law-abiding
citizens. For, in willing the former, he would be thwarting his pursuit of the
latter. Therefore, the view at issue forces us to embrace the counterintuitive
Conclusion: Kant™s Candidates 181

conclusion that, in making the arrest, the of¬cer is expressing disrespect for
the white supremacist™s rational agency and thereby acting wrongly.
A natural reply to cases such as this is to ¬ne-tune the view in question,
perhaps by claiming the following. According to the Formula of Humanity,
a person whom an agent is treating in a certain way must be able to share
the agent™s end, unless what would prevent his sharing the end is his acting
in a morally impermissible way. On this modi¬ed view, it seems that the
of¬cer™s arresting the white supremacist would conform to the Formula of
Humanity. For what would prevent the white supremacist from sharing her
end would be his (obviously immoral) attempt to get revenge.
But what is the standard by which we are supposed to determine whether
the other is acting in a morally impermissible way? Perhaps Kant would
appeal to the Formula of Universal Law, holding that a person whom an
agent is treating in a certain way must be able to share the agent™s end,
unless what would prevent his sharing the end is his acting contrary to
the Formula of Universal Law. But this appeal would be problematic. First,
if in some cases such an appeal were necessary to make the Formula of
Humanity work, then would it really be a viable candidate for the supreme
principle of morality? The supreme principle is supposed to be such that all
moral duties are derived ultimately from it, not from it in combination with
some other moral principle. Of course, this dif¬culty would dissolve if, as
Kant suggests, the two principles were equivalent. But as we have seen, it is
very doubtful whether they are. Second, and more important, the Formula
of Universal Law does not appear to be a reliable indicator of an action™s
moral permissibility. Indeed, the formula seems particularly ineffective in
generating results that cohere with the ordinary conviction that actions such
as that of the white supremacist are wrong. A maxim of attacking a racial
minority to get revenge would seem to pass the Formula of Universal Law
test. We proposed a modi¬cation in our understanding of when someone
whom an agent treats in a certain way must (morally speaking) be able to
share the agent™s end. This modi¬cation is ineffective.
Perhaps the modi¬cation we need is not in our understanding of when
someone must be able to share an end, but in our understanding of what
it would mean to share an end. We have been employing an interpretation
according to which someone shares an agent™s end just in case he (actually)
pursues the end. But this interpretation might be misguided. After all, in
the false promising example, Kant says: “For, he whom I want to use for my
purposes by such a promise cannot possibly agree to my way of behaving toward
him, and so himself contain the end of this action” (GMS 429“430, emphasis
added). Perhaps another™s containing the end of an agent™s action toward
him (i.e., sharing this end) amounts to the other™s being able to consent
to the agent™s pursuing his end in the way he does. (This strikes me as a
rather tenuous sense of sharing an end, but so be it.) On this reading, the
Formula of Humanity would escape the (in my view unwelcome) implication
Search for the Supreme Principle of Morality
182

that Pete™s action toward Andre was morally impermissible. For there is
no reason to suppose that Andre cannot consent to Pete™s pursuing the
end of being number one through beating him in the ¬nals of the U.S.
Open.
The obvious dif¬culty presented by this interpretation, however, is that of
pinpointing what it means for a person to be able to consent to being treated
in a certain way. It is clearly not the case that a necessary condition for an
agent™s being able to consent is that he would, upon re¬‚ection, consent if
given the occasion to do so. Would the white supremacist, even if queried
after calm deliberation, consent to the of¬cer™s action of trying to thwart his
plot? Being able to consent in the requisite sense to being treated in a certain
way must amount to being able, rationally speaking, to consent to it. But
what does rational consent amount to? Echoing the preceding discussion,
one might claim that a person can rationally consent to an agent™s pursuit of
his end just in case this pursuit would not in itself prevent the person from
attaining his ends. Pete™s pursuit of the number-one ranking would not itself
block Andre from gaining this ranking. Yet the of¬cer™s pursuing his aim
of preventing race-based attacks on law-abiding citizens through arresting
the white supremacist may well preclude the latter from attaining his goal of
revenge. So that strategy does not seem promising. One might instead claim
that a person can rationally consent to an agent™s pursuit of an end just in
case this pursuit is morally permissible. Once again, however, we need to
know what the standard of moral permissibility is supposed to be. If it is
the Formula of Universal Law, then familiar dif¬culties arise. First, if we
need to invoke this formula, then it is questionable whether the Formula of
Humanity is really a candidate for the supreme principle of morality. Second,
the Formula of Universal Law generates counterintuitive results. The white
supremacist™s maxim seems to pass its test, and thus we seem to arrive at the
odious conclusion that his victims can rationally consent to being attacked.
There is, I think, something philosophically attractive in the notion that
an agent does not respect another™s rational nature unless the other can
rationally consent to the way he treats him. But specifying what is meant
here by rational consent is a dif¬cult task “ one that I do not undertake
here.
In the end, perhaps we need not focus at all on some inability of the
recipient of a false promise to consent to or to share the promisor™s end to
explain why the promisor™s action violates the Formula of Humanity. Here
is a sketch of how one might proceed. To conform with this principle, the
promisor™s action must be compatible with his valuing the recipient™s hu-
manity as something absolutely and incomparably good. But in typical cases,
making a promise to another that one has no intention of keeping is not
compatible with so valuing the recipient™s humanity. To value another™s hu-
manity is to value his capacity to set and to pursue ends. But if one values his
Conclusion: Kant™s Candidates 183

capacity to pursue ends, then, other things being equal, one must also value
his capacity to pursue them successfully. (This move, which we discussed
brie¬‚y in connection with Kant™s derivation of a duty of bene¬cence, is ad-
mittedly controversial.) In a typical case, however, someone making a false
promise realizes that he will thwart (or at least runs a signi¬cant risk of
thwarting) the promisee in her pursuit of her ends. For example, the false
promisor would realize that, because he obtains a loan from a person on the
basis of a promise to her that he has no intention of keeping, the person will
not have that money at hand to do what she wills with it. In short, making
a false promise to someone expresses disrespect for the person™s rational
agency since it expresses indifference (or even contempt) for the agent™s
own projects.


8.9 Formula of Humanity: Further Challenges
The preceding section illustrated some of the challenges we face in de-
riving from the Formula of Humanity duties we take ourselves to have.36
Although the details need to be worked out, it seems that the formula is
capable of generating duties of bene¬cence and sincerity in promising. I
will close my brief discussion of the Formula of Humanity by pointing out
a few further hurdles defenders of it need to overcome if they are to es-
tablish that it ful¬lls Kant™s eighth criterion for the supreme principle of
morality.
First, a cluster of questions arise around Kant™s claim that humanity has
dignity and must be treated as such. As something with dignity, humanity has
incomparable worth. According to Kant, it appears, this implies that it is never
legitimate to treat humanity as a value to be exchanged for or compensated
by either anything with mere price or anything with dignity. One fails to treat
humanity as an end in itself in any situation in which one destroys the
humanity in one person on the grounds that doing so is necessary to secure
the “greater value” inherent in the humanity of two or more other persons.
But what if the number of these other persons is 10, 1,000, or even 1 million,
as it might be in some emergency? Even in extreme circumstances is it
morally impermissible to treat humanity as a value to be sacri¬ced in order
to secure more (even vastly more) of this very same value? It is not obvious
that we would morally condemn the leader of a counterterrorist force for
treating one innocent hostage held at the front of a plane as a value to
be sacri¬ced (along with the value inherent in the terrorist) in order to
preserve the greater value of the 350 remaining passengers and crew on
board. Some emergency situations threaten to bring out disagreement with
Kant™s apparent view that the value of the humanity is not only incomparable
to the value of things, but also to the value of other instances of humanity
itself.37
Search for the Supreme Principle of Morality
184

In my view, though admittedly not in everyone™s, it would be particularly
damaging to the Formula of Humanity if it followed from it that it would
never be permissible to kill one agent, where killing this agent would have the
effect of saving (many) others. However, that it is wrong to treat the humanity
in one individual as a value to be sacri¬ced for the sake of preserving the
“greater value” inherent in the humanity of many others does not entail that
it is morally impermissible to kill the one and thereby preserve the humanity
in the many. It simply entails that one™s grounds for killing the one cannot
be that the humanity in him does not add up to a value as great as that
of the humanity in the many others.38 But the challenge then is to locate
other grounds consistent with the Formula of Humanity™s being the supreme
principle of morality for killing in circumstances in which, according to
common sense, this is appropriate.39
Another cluster of issues that must be addressed, if we are to see that
the Formula of Humanity generates results acceptable to ordinary moral
reason, concerns the formula™s implications regarding how we must treat
existing beings who do not have humanity (e.g., animals), as well as beings
who do not yet exist but who will have humanity (future generations). This
cluster of issues concerns the scope of humanity. In the Groundwork Kant
contends that only persons, that is, beings who have humanity, are ends in
themselves, and that all other beings “have only a relative worth, as means,
and are therefore called things” (GMS 428). In the Metaphysics of Morals,
he asserts that “a human being has duties only to human beings (himself
and others), since his duty to any subject is moral constraint by that sub-
ject™s will” (MS 442). Kant goes on to make clear that the term “human
beings” here refers to “persons,” beings who have humanity, and that all
the persons of which we have experience are human beings. So, in effect,
Kant is claiming here that a person has duties only to himself and to other
persons; he has no duties to beings who do not possess the set of capaci-
ties that make up humanity. There seems to be no tension at all between
this claim and the Formula of Humanity itself. After all, this formula com-
mands us merely so to act that we treat humanity as an end in itself; it says
nothing concerning how we must act toward beings without humanity. In
light of Kant™s suggestion that beings without humanity are valuable merely
as means and that we have no duties to such beings, it might seem to fol-
low that according to the Formula of Humanity, it is morally permissible
to treat them as we will. Does the Formula of Humanity forbid, as ordinary
moral reason (arguably) does, our causing tremendous pain to animals or
to severely disabled human beings for the sake of making our lives a bit
easier?
As readers of the Metaphysics of Morals are aware, Kant argues that, al-
though we have no duty to beings without humanity, we do have duties with
regard to such beings. More precisely, we have duties to ourselves that require us
Conclusion: Kant™s Candidates 185

to treat these beings in certain ways. For example, Kant argues that “violent
and cruel treatment of animals” is opposed to “a human being™s duty to
himself . . . for it dulls his shared feeling of their suffering and so weakens
and gradually uproots a natural predisposition that is very serviceable to
morality in one™s relations with other people” (MS 443). Kant™s point here
seems to be that by treating animals (and presumably severely disabled hu-
man beings) violently and cruelly, we desensitize ourselves to the suffering
of persons, thus making it more dif¬cult for us to ful¬ll our duties to them.
As Kant himself suggests (MS 456“457), if we do not cultivate our capacity
to recognize suffering in persons, then we might be less effective than we
otherwise would be in ful¬lling our duty of bene¬cence toward them. To
help others effectively, we need to recognize when (and how) they need
help. Training ourselves to share in their feelings of suffering can aid us in
doing so. At any rate, Kant appears to base a prohibition on cruel treatment
of animals on an agent™s duties to other persons, so it is odd that he says it
is based on an agent™s duty to himself. Perhaps Kant has in mind that, by
diminishing an agent™s ability to ful¬ll his duty of bene¬cence, cruelty to
animals would hinder his ability to ful¬ll a duty he discusses just a few pages
later, namely his duty to himself to increase his moral perfection (MS 446).
This duty requires one to strive to ful¬ll all of his duties, including, of course,
that of bene¬cence.
Kant™s claim that an agent must not be cruel to beings devoid of humanity,
since doing so hinders her from ful¬lling duties to the humanity in herself,
may not satisfy re¬‚ective moral common sense. First, it seems to be a ques-
tionable thesis concerning human psychology that cruelty to animals (or, for
that matter, to the severely disabled) always “weakens and gradually uproots”
an agent™s ability to feel the suffering of other persons. Is the equestrian who
whips her horse in a competition necessarily diminishing her capacity to em-
pathize with her fellow persons? Second, and more important, some might
question whether the reason we should avoid treating nonpersons cruelly
is really that (or just that) we thereby make it more dif¬cult for us to ful¬ll
our duties to ourselves, rather than that (or also that) we make them suffer
unnecessarily. Unnecessary suffering, they might say, is a bad thing, whether
it be the suffering of a horse or of a man in the late stages of Alzheimer™s
disease. Kant writes that “agonizing physical experiments [on animals] for
the sake of mere speculation, when the end could also be achieved without
these, are to be abhorred” (MS 443). But are they to be abhorred, as he
suggests, simply because they diminish the experimenter™s capacity to ful¬ll
his duties to himself, or (also) because they cause needless pain to sentient
beings?
The passage in the Metaphysics of Morals we have brie¬‚y discussed raises
questions not only regarding the Formula of Humanity™s implications
concerning the treatment of existing beings devoid of humanity but also
Search for the Supreme Principle of Morality
186

concerning obligations we (might) take ourselves to have to future genera-
tions. Cited more fully, the passage reads:

[A] human being has duties only to human beings (himself and others), since his
duty to any subject is moral constraint by that subject™s will. Hence the constraining
(binding) subject must, ¬rst, be a person; and this person must, secondly, be given
as an object of experience, since the human being is to strive for the end of this
person™s will and this can happen only in a relation to each other of two beings that
exist (for a mere thought-entity cannot be the cause of any result in terms of ends).
(MS 442)

We might get the impression here that according to Kant an agent has duties
only to existing persons, not to any persons who will exist in the future. For
the latter beings seem merely to be “thought-entities.” This impression is
not dissipated by the Formula of Humanity itself, which seems compatible
with the view that we have no duties to future generations. For it merely
commands that we so act that we treat the humanity in ourselves and in any
other as an end in itself. But, arguably, since future generations do not yet
exist, there is no humanity in them. If Kant™s theory, speci¬cally his advocacy
of the Formula of Humanity as the supreme principle of morality, implies
that we have no duties to future generations (of persons), then it might
clash with re¬‚ective moral common sense. For many of us do hold that we
have such duties “ for example, a duty not to pollute the environment to
such an extent that our descendants (none of whom now exist) will live in
a quagmire of disease and malnutrition, and thus be unable to effectively
pursue their happiness.
A ¬rst step toward meeting this challenge, which seems to pose fewer
dif¬culties than Kant™s views toward animals, the severely disabled, and so
forth, might be to examine the dialectical context in which Kant™s remarks
occur. Kant™s suggestion that persons do not have duties to mere “thought-
entities” appears merely to be a premise in an argument he aims against
the notion that we have duties to God. Shortly after making this suggestion,
Kant claims that “we do not have before us, in [the idea of God], a given
being to whom we would be under obligation; for in that case its reality
would ¬rst have to be shown (disclosed) through experience” (MS 444).
We can have obligations only to beings belonging to a kind which is such
that we can experience its members™ reality, Kant seems to be arguing. God
is not such a being; therefore we cannot have obligations to God. Whatever
the merits of this argument, it might be compatible with the notion that we
have duties to future generations. For the future generations in question
do belong to a kind, that of persons, which is such that we can experience
its members™ reality; we experience the reality of persons every day. The
Formula of Humanity commands that we treat the humanity in ourselves and
in others as an end in itself. I see no reason why a defender of the Formula
of Humanity could not suggest that the scope of “others” include the future
generations of persons whom we can reasonably be assumed to affect.
Conclusion: Kant™s Candidates 187

In sum, to ful¬ll Kant™s eighth criterion, a candidate for the supreme
principle of morality must generate moral prescriptions that square with
those uncontroversially embraced by re¬‚ective moral common sense. The
Formula of Humanity faces some dif¬culties on this score. But the prospects
for it seem much brighter than those for the Formula of Universal Law.


8.10 Where We End Up
The argument of this book has taken shape against the background of the
traditional reading of Kant™s Groundwork derivation of the Formula of Uni-
versal Law. Kant™s derivation fails miserably, according to this reading, since
it involves a leap from a practically uninformative principle to the Formula
of Universal Law. Kant is left with embarrassingly inadequate support for
one of the foundational claims in his ethics. I have argued that we respond ef-
fectively to the traditional reading neither by appealing to the second Critique
derivation of the Formula of Universal Law, as reconstructed by Allison,
nor by focusing solely on Kant™s derivation of the Formula of Humanity,
as reconstructed by Korsgaard. Both of these reconstructed derivations suf-
fer from fundamental ¬‚aws. We should instead challenge the traditional
interpretation itself.
The central thesis of the book can be crystallized into a few sentences:
There is a textually plausible reading of Kant™s Groundwork derivation of
the Formula of Universal Law, namely the criterial reading, that shows this
argument to be far more philosophically engaging and forceful than does
the traditional reading. With the help of this argument, Kant makes a con-
vincing case against some key rivals to the Formula of Universal Law (e.g.,
some consequentialist principles). Even though in the end the Formula of
Universal Law has serious and probably fatal shortcomings as a candidate
for the supreme principle of morality, an argument of the sort Kant employs
in deriving it (a criterial argument) holds substantial promise as a way of
defending his Formula of Humanity “ a principle that many philosophers,
including me, ¬nd especially attractive.
Kant™s criteria for the supreme principle of morality have been at the
core of the discussion. By appealing to them Kant shows several rivals to his
formulas not to be viable candidates for status as the supreme principle. So it
might be helpful to summarize some main ¬ndings regarding Kant™s criteria.
Four of them belong to Kant™s basic concept of the supreme principle
of morality. According to this concept, the supreme principle of morality
must be (i) practical, (ii) absolutely necessary, (iii) binding on all rational
agents, and (iv) the supreme norm for the moral evaluation of action. These
criteria, which are all at least implicit in the Groundwork Preface, have not
received nearly as much attention as the ones Kant develops in the course of
Groundwork I. That is because the focus of the book has been the claim 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.
Search for the Supreme Principle of Morality
188

Although I do not pretend to have defended criteria i and ii here, I agree
with Kant that according to re¬‚ective moral common sense, the supreme
principle of morality would have to be both practical and absolutely nec-
essary. It would have to be something on account of which agents might
actually act “ not, for example, merely a theoretical tool to be used by ex-
perts to determine the rightness of actions after they occur. It would also
have to be a principle that each agent ought to obey no matter what she
desired.
However, I have made a couple of critical points in connection with the
criteria implicit in Kant™s basic concept of the supreme principle of moral-
ity. First, and very brie¬‚y, Kant implies that if we embrace criterion ii, then
we must embrace criterion iii. In other words, if we agree that the supreme
principle of morality must be absolutely necessary “ that is, unconditionally
binding on the agents within its scope “ then we are compelled to agree that
the principle must have a scope that extends to all rational agents, including
any nonhuman ones such as angels. Apparently, Kant believes this point to
be obvious. In section 2.4 (in connection with Henry Allison™s reconstruc-
tion of Kant™s second Critique derivation of the Formula of Universal Law)
I protested that it is not. Kant owes us an explanation of why a principle
could not be unconditionally binding on all human rational agents “ that is,
one that each of us ought to obey no matter what her inclinations might be,
yet not binding on some other type of rational agent, for example, a type
that is necessarily incapable of conforming to it. I do not have any particular
reason for rejecting iii; it is just that, contrary to Kant, I do not believe that
it follows quickly and easily from ii.

<<

. 6
( 8)



>>