1

Overridingness, Moral
Joshua Gert
Morality is one normative domain among many. That is, not only can we assess
actions as morally right or wrong, or as morally good or bad, but we can also assess
them as legal or illegal, prudent or imprudent, or as prohibited or required by the
rules of etiquette. It is quite obvious that these assessments need not always favor
and disfavor precisely the same actions. For example, it might be morally required in
some cases to perform an action that is prohibited by law, or some might suggest
it might be imprudent to do something morally good. The thesis of moral
overridingness is the thesis that moral verdicts are always in some sense supreme
whenever they come into conflict with the verdicts of a distinct normative domain.
This general thesis can be understood in a number of ways. To begin with, we need
to offer an interpretation of the claim that one verdict overrides another. One might
interpret this claim as entailing that, in conflicts of verdicts, it is irrational to go
against the moral verdict (see rationality). Or, one might interpret it as the claim
that the totality of practical reasons, taken together, always uniquely favors acting in
line with a moral verdict, even though it may not always be irrational to act in ways
that are not uniquely favored in this way (see reasons; satisficing). In making the
overridingness thesis clear, one also needs to specify which verdicts are relevant.
Very often, the thesis is understood as restricted to claims about conflicts of
requirements. This is an important restriction, since morality can yield the verdict
that an action would be morally good without being morally required. For example,
in many views, we are not morally required to volunteer our free time doing
charitable activity. In such views, morality favors such action, but does not require it
(see supererogation). On a common interpretation of the overridingness thesis,
the fact that morality favors such action does not imply that there would be anything
problematic in deciding against such a morally good action, and performing some
other action instead: say, an action that would do a better job of promoting ones
own interests.
Something similar to the thesis that morality is overriding appears as early
as the first extended philosophical discussions of virtue and justice. Platos
Republic (359d360b) famously contests the view that profitable injustice is to be
preferred when it can be done with impunity. Hobbes (1994 [1651]: Book 1, Chs.
13, 14), similarly, tried to show that injustice was always a bad idea from the
perspective of self-interest. And Kant (1993 [1785]: Ch. 2) argued that the demands
of morality are categorical in a way that no other demands can be. Sidgwick
(1981 [1907]: 508), on the other hand, had grave doubts about the possibility of
demonstrating, in a non-question-begging way, that the impartial perspective

The International Encyclopedia of Ethics. Edited by Hugh LaFollette, print pages 37643770.
2013 Blackwell Publishing Ltd. Published 2013 by Blackwell Publishing Ltd.
DOI: 10.1002/ 9781444367072.wbiee010

characteristic of morality was to be preferred even to a strictly egoistic perspective. And utilitarians such as Mill (2001 [1861]: 313) were positively clear that
the morally correct action might be very different from the action that would
produce the greatest benefit to the agent. However, Mill also held the view that
we humans are psychologically determined to pursue only what we perceive to
be best for us individually.
Contemporary discussion of moral overridingness might be said to begin with
Philippa Foots Morality as a System of Hypothetical Imperatives (1972). A
common way of attempting to show that morality is in some sense a supreme set of
norms is to point out that its commands are categorical, rather than hypothetical:
that is, that their applicability does not depend on our contingent ends or desires.
Foot raises important problems for this strategy, however. When it is spelled out
more explicitly, the thesis of categoricity seems either false, mysterious, or insufficient to support moralitys claim to supremacy. If categorical simply means not
dependent on our contingent ends, then we can truly say that moral verdicts are
categorical imperatives. After all, we do not withdraw the claim that our friend
should not cheat on her taxes just because we are reliably informed that she does not
have any ends that would be furthered by filling out her tax forms honestly. However,
this kind of categoricity seems purely grammatical. The requirements of etiquette
are similar in this respect; whether or not it furthers any of ones goals, the done
thing is to answer invitations that are written in the third person by accepting in
thethird person.
Foots second interpretation of categorical is guaranteed to be backed by decisive
reasons. However, since, according to Foot, reasons stem only from our objective
interests and subjective desires, there simply is no guarantee that a morally required
act always is supported by decisive reasons. One way of blocking this argument is to
extend the notion of a practical reason beyond the realm of interests and subjective
desires, to include, for instance, the interests of others. This strategy encounters other
problems, not specifically addressed by Foot.
Finally, there remains the thought that if one is morally required to perform some
action, then one just plain ought to do it, in a way that is false of the requirements
of etiquette, and that does not depend on the action furthering ones interests or
desires. However, says Foot, this just plain ought is an incoherent fiction, and our
temptation to appeal to it is best explained by an upbringing that, unsurprisingly,
instilled in us a sense of the great importance of morality.
Rather than vainly try to defend the categorical nature of morality in any of the
preceding ways, Foots suggestion is that we simply abandon that ambition, and
recognize that the demands of morality will necessarily be backed by decisive reason
only for those who have a sufficiently strong contingent commitment to its ends to
the welfare of other people, their fair treatment, and so on. However, this need
not undermine the importance of morality, especially to those who have this
commitment. This argument by Foot touched off a wave of attempts to show that
morality is in fact overriding in some significant way.

In arguing for the overridingness of moral verdicts, it is important to distinguish

overridingness from inescapability. To say that morality is inescapable is to distinguish it from, say, the rules of a club, from which one can escape by resigning,or
from the laws of a particular country from which one can escape by emigrating.
However, even if there is no escape from morality in this sense, it still might be that,
on some occasions, one has sufficient reasons to justify violating its requirements
just as one might have such reasons to violate the laws of a country in which one is
living, or a club to which one still belongs. On such occasions, one would still be
doing something morally wrong one would therefore not have escaped morality
but one might nevertheless be justified in a more comprehensive way: for example,
one might be either rationally required or rationally justified in acting in a way that
morality condemns.
Some have attempted to defend a weakened version of the overridingness thesis,
according to which the reasons we have do depend on our contingent commitments,
but according to which any genuine commitment to morality entails regarding it as
ones most important commitment (Phillips 1977; Shiffrin 1999), at least in cases of
conflict between its verdicts and those of some other domain. Against this strategy,
one might point out that there seem to be religious people who are honestly
committed to morality, but for whom religious requirements trump all other
concerns. Hence, this strategy, which is in some sense a psychological hypothesis,
seems falsified by empirical evidence.
The central conflict that most theorists have in mind when discussing the
overridingness of morality is the conflict between morality and self-interest. If this
is what one is primarily thinking about, then it may seem that one can defend the
overridingness of morality by noting that while verdicts regarding self-interest do
not take moral considerations into account, moral verdicts do take considerations
regarding self-interest into account. For example, it can count as a moral justification for the violation of a promise, that keeping the promise would involve significant hardship that was unforeseen at the time the promise was initially made.
Similarly, considerations of self-defense can sometimes morally justify the injury of
other people. It can thus appear that moral verdicts take other things into account,
and can be viewed as all-things-considered verdicts (Becker 1973). And, from this
conclusion, it is a short step to the further conclusion that what one ought to do
full stop is what morality requires. However, there are two worries about this line
of thought. The first is that morality may simply not take everything into account. It
is a common claim that certain considerations ones that might in fact have a
psychological pull on us are morally irrelevant. If any of these claims are true, this
means that morality does not in fact take them into account. It would take much
more than a simple appeal to a handful of cases in which self-interest is morally
relevant to establish that none of the considerations that are morally irrelevant are
ever themselves genuine practical reasons. A second problem for the strategy of
assuming that morality is all-inclusive is that, even if it is true that moral verdicts
take all the relevant practical reasons into account, those verdicts might not take

them into account in the same way in which, say, verdicts of prudence or self-interest
do. That is, to say that morality takes considerations of self-interest into account as
the preceding examples show is sometimes true is not yet to say that those very
same self-interested considerations might not justify, in some more comprehensive
way, acting against moral requirements. This would happen if morality took
self-interest into account, but discounted its importance to some degree.
One common way of thinking of the overridingness thesis is in terms of an overarching
normative system, which we might call rationality. This system takes as input the overall verdicts of special domains such as morality, prudence, and etiquette, and combines
them to yield an all-things-considered verdict. The overridingness thesis is that whenever one of these input verdicts is morally required, then the all-things-considered
verdict the verdict of rationality favors that action. What is distinctive about this
interpretation is that the overall verdicts of each special normative domain are treated as
distinct reasons, each quite independent of the other (see Haji 1998 and McLeod 2001).
One apparent problem with the view that the overarching domain of rationality
takes the verdicts of the special domains as inputs is that those verdicts are presumably
backed by reasons. To take those verdicts themselves as reasons seems to involve an
objectionable double-counting. For example, some plausible reasons why murder is
wrong are that it deprives someone of some part of the life he or she would otherwise
have lived, and that it causes suffering to the friends and family of the person who is
killed. These are reasons against murder reasons that presumably should be taken
into account by any verdict issued by the overarching domain of rationality. However,
if the moral wrongness of the murder also counts as a reason relevant to an overall
assessment, these reasons seem to be counted twice.
The preceding problem with the verdict-based interpretation of overridingness
leads to another way of thinking of overridingness, and of thinking of rationality as
an overarching normative system. This interpretation begins with the idea that there
is a domain of generic practical reasons reasons that have some power to require
us to act, and some power to justify us in acting against other such reasons. It also
holds that the rational status of an action is determined by some function of these
reasons. If one takes this view, then the bare verdicts of certain normative domains
might be shown not to entail the existence of any reason whatsoever. For example, it
may be that a requirement of etiquette just by itself provides no such reason. Of
course, this is consistent with the fact that one may have instrumental reasons to act
in line with such requirements: say, if one wishes to placate a powerful person who
takes such requirements very seriously. On the reasons-based approach to moral
overridingness, what one would need to show, in order to demonstrate that morality
is overriding, is that whenever an action is morally required, that same action is
favored by reasons that somehow outweigh or decisively oppose in some other
way, such as by undermining the reasons favoring any contrary action. This would
be enough to show that moral verdicts defeat the conflicting verdicts of any other
normative domain, except the overarching domain of rationality that gives sense
to the overridingness thesis. Nor could there ever be a conflict between morality
and rationality on this view. It should be noted that this interpretation of the

overridingness thesis is very closely related to the thesis of moral rationalism (see
rationalism in ethics). This is the thesis that moral requirements are somehow
the product of rationality (Nagel 1970: 3; Korsgaard 1986: 5).
However, the idea that overridingness should be understood simply in terms of
the weight of all relevant practical reasons also faces troubles at least for those who
advocate the overridingness thesis. Sarah Stroud illustrates this problem by asking us
to consider a series of moral views. The first such view takes only other-regarding
reasons into account, and requires us to act in the ways maximally favored by
suchreasons. The problem that this view presents for the overridingness thesis is
that the reasons that are excluded by this moral theory return to claim their due
when we turn to consider what action is favored by the totality of reasons. And there
is no reason to believe that they cannot yield a verdict that is opposed to the moral
verdict. Suppose then that we allow self-interested reasons some weight in the moral
theory but not the full weight they receive from the general perspective on reasons
that is unattached to any special normative domain. The problem then remains the
same: from the general perspective, these discounted reasons become undiscounted, and may yield a verdict at odds with the moral one. Even a purely consequentialist moral theory, which considers the agents interests as equal to the
interests of other people, may create problems for the defender of moral overridingness
(see consequentialism). At least this is true if, as Stroud believes, all else being
equal, a persons reasons for pursuing her own aims are stronger than her reasons to
advance someone elses (Stroud 1998: 183).
Stroud argues persuasively that overridingness, interpreted in terms of reasons, is
incompatible with various moral theories. However, her point can be carried further
than she carries it. For if the claim of moral overridingness is simply the claim that a
morally required action is always uniquely favored by the totality of relevant reasons,
then the only kind of moral theory that could possibly vindicate overridingness is a
theory on which moral requirements are simply a subset (perhaps a maximal subset)
of those actions uniquely favored by reasons. Consider, for example, a contractualist
theory that explains moral requirements in terms of a set of rules that rational people
would agree on in certain hypothetical circumstances (see contractualism). There
is absolutely no reason to suppose that an action required by such a set of rules will
also always be maximally favored by the available reasons in the actual circumstances
in which an agent must decide how to act, for the choice of the set of rules is not even
the choice of an action it is the choice of a comprehensive system. The reasons that
favor choosing such a system under certain hypothetical circumstances will derive,
at least in part, from the consequences of that system being generally known and
accepted. These reasons are simply very different from the reasons of relevance to
the decision of a specific agent regarding the choice of a particular action in particular
circumstances. And there is no reason to think that the action favored by the favored
system will also independently be uniquely favored by the available reasons in those
particular circumstances. Of course, the incompatibility of the overridingness thesis
with a contractualist theory by itself does not indicate which of the two ought to be
abandoned or modified.

A weaker hypothesis, closely related to overridingness, is that if an action is

morally required, then the available reasons never uniquely favor acting contrary to
that moral requirement. One way of defending this position would be to appeal to
incommensurability of reasons (see incommensurability (and incomparability)).
If the reasons that favor a morally required action are always incommensurable with
the reasons that stand behind other reason-backed verdicts such as the verdicts of
prudence or self-interest, then those latter reasons will not be stronger than the
moral reasons, and therefore the totality of relevant reasons will not uniquely favor
acting contrary to morality. A related suggestion is that generic practical reasons
play two roles: requiring action, and justifying action that the agent would otherwise
be required not to perform. If this is true, and if the reasons that stand behind moral
requirements always have sufficient strength in the justifying role, then it is possible
that one might never be required to act contrary to morality.
See also: consequentialism; contractualism; incommensurability
(andincomparability); rationalism in ethics; rationality; reasons;
satisficing; supererogation
REFERENCES
Becker, Lawrence 1973. The Finality of Moral Judgments: A Reply to Mrs. Foot, Philosophical
Review, vol. 82, pp. 36470.
Foot, Philippa 1972. Morality as a System of Hypothetical Imperatives, Philosophical Review,
vol. 81, pp. 30516.
Haji, Ishtiyaque 1998. On Moralitys Dethronement, Philosophical Papers, vol. 27, pp. 16180.
Hobbes, Thomas 1994 [1651]. Leviathan, ed. Edwin Curley. Indianapolis: Hackett.
Kant, Immanuel 1993 [1785]. Grounding for the Metaphysics of Morals, ed. James Ellington.
Indianapolis: Hackett.
Korsgaard, Christine 1986. Skepticism about Practical Reason, Journal of Philosophy,
vol.83, pp. 525.
McLeod, Owen 2001. Just Plain Ought, Journal of Ethics, vol. 5, pp. 26991.
Mill, John Stuart 2001 [1861]. Utilitarianism, ed. George Sher. Indianapolis: Hackett.
Nagel, Thomas 1970. The Possibility of Altruism. Princeton: Princeton University Press.
Phillips, D. Z. 1977. In Search of the Moral Must: Mrs Foots Fugitive Thought, Philosophical
Quarterly, vol. 27, pp. 14057.
Shiffrin, Seana 1999. Moral Overridingness and Moral Subjectivism, Ethics, vol. 109,
pp. 77294.
Sidgwick, Henry 1981 [1907]. The Methods of Ethics. Indianapolis: Hackett.
Stroud, Sarah 1998. Moral Overridingness and Moral Theory, Pacific Philosophical
Quarterly, vol. 79, pp. 17089.

FURTHER READINGS
Copp, David 1997. The Ring of Gyges: Overridingness and the Unity of Reason, Social
Philosophy and Policy, vol. 14, pp. 86106.

7
Isaacs, Tracy, and Diane Jeske 1997. Moral Deliberation, Nonmoral Ends, and the Virtuous
Agent, Ethics, vol. 107, pp. 486500.
McDowell, John 1978. Are Moral Requirements Hypothetical Imperatives? Proceedings of
the Aristotelian Society, suppl. vol. 52, pp. 1329.
Nagel, Thomas 1986. The View from Nowhere. New York: Oxford University Press.
Taylor, Paul 1978. On Taking the Moral Point of View, Midwest Studies in Philosophy, vol. 3,
pp. 3561.
Scheffler, Samuel 1992. Human Morality. New York: Oxford University Press.
Sinnott-Armstrong, Walter 1987. Moral Dilemmas and Ought and Ought Not, Canadian
Journal of Philosophy, vol. 17, pp. 12739.
Terrell, Huntington 1969. Are Moral Considerations Always Overriding? Australian Journal
of Philosophy, vol. 47, pp. 5160.
Tiberius, Valerie 2005. Value Commitments and the Balanced Life, Utilitas, vol. 17,
pp. 2445.
Timmermann, Jens 2006. Kantian Duties to the Self, Explained and Defended, Philosophy,
vol. 81, pp. 50530.
Wolf, Susan 1982. Moral Saints, Journal of Philosophy, vol. 79, pp. 41939.