MORAL DEADLOCK

RONALD D. MILO

Introduction

Very often moral disagreements can be resolved by appealing to (non- moral) factual considerations because in
these cases the parties to the dispute agree as to which factual considerations are relevant. They agree, that is, with respect
to their basic moral standards. Hence, when their disagreement about the non-moral facts is resolved, so is their moral
disagreement. But sometimes moral disagreement persists in spite of agreement on factual considerations. When this
happens, and when neither party is guilty of illogical thinking, we have a case of moral deadlock.
Some claim that moral deadlocks occur because in such cases the disputants disagree with respect to their most
basic moral standards or values. Since they do not even agree as to what factual considerations are relevant to the moral
evaluation in question, arguments proffered on either side will inevitably break down. Moral deadlock results because the
disagreement is rooted not in disagreement in beliefs, but in disagreement in attitudes, and because whether or not one is
logically constrained to accept a given moral standard depends on what one's basic attitudes and preferences happen to be.
Others argue against the possibility of such radical moral disagreement, as I shall call it, on the ground that it is laid down
in virtue of what we mean in calling an act morally right or wrong that certain facts necessarily count as evidence or
reasons for or against the moral rightness or wrongness of acts. Anyone who denies their relevance is simply refusing to
discuss the question of the morality of the act in question. One is not free, therefore, to make whatever one likes count as
evidence, just so long as one has the appropriate pro- or con-attitudes or is prepared to make the requisite universal
prescriptions.
Now, for reasons which can only be partially elaborated in this essay, I am inclined to agree with this latter point
of view. Hence, I am also inclined to deny that radical moral disagreement is possible. To claim that what I am calling
radical moral disagreement is possible is to claim that people can be said to disagree about the moral rightness or wrong-
ness of a certain course of action because they disagree about the basic criteria of relevance for determining moral
rightness and wrongness. I am inclined to deny this because I am inclined to think that in such cases we do not really have
moral disagreement at all-that in the absence of certain agreed upon standards there can be neither moral disagreement nor
moral agreement. But it is important to recognize that in saying this I am not saying that moral deadlocks cannot occur.
What I am calling a moral deadlock occurs whenever the disputants agree as to what factual considerations obtain but
nevertheless disagree in their moral evaluations, and when neither is guilty of illogical thinking. To hold that there are
certain factual considerations whose relevance to moral questions cannot be denied by anyone who wishes to consider
these questions is not to hold that there is some unique decision-procedure for resolving all moral disputes, so that in any
such dispute at least one of the parties must be mistaken. Rather, as I shall argue, moral disagreement can persist even
when there is no disagreement either about what factual considerations are relevant or about what factual considerations
obtain, and when no party to the dispute is guilty of fallacious reasoning. It is a mistake, therefore, to identify moral
deadlock with radical moral disagreement.
Indeed, as I shall try to show in this essay, not only can moral disagreement persist in spite of agreement on basic
standards of relevance, together with agreement on the facts and the absence of illogical thinking, the disagreement may
persist for a variety of reasons. The main purpose of this essay is to explore these sources of moral deadlock and to
explain how it is that moral deadlock can sometimes be the end result of internal moral disagreement i.e. disagreement
between people who agree with respect to their basic moral standards or values. Once one realizes the extent to which
moral deadlocks can occur even among people who agree with respect to their basic moral standards, and once one
realizes the seriousness and importance of some of these disagreements, much of the sting will have been removed from
the claim that radical moral disagreement is not really possible; for much of what those who wish to claim that radical
moral disagreement is possible want to say will have been granted. Nevertheless, I hope it will also be the case that once
one clearly distinguishes radical moral disagreement from the kinds of moral deadlock that are really species of internal
moral disagreement, the possibility of radical moral disagreement will seem problematic.

Internal Moral Disagreement

When moral disagreement occurs between people who agree with respect to their basic moral standards, we have
a case of internal moral disagreement. When, in addition, they agree with respect to their factual beliefs and no one is
guilty of irrational thinking, we have a case of moral deadlock. Now some may wonder how such a deadlock can be
possible. If they agree both as to what factual considerations are relevant and as to what facts actually obtain, how can
they disagree in their moral judgments without at least one of them being guilty of illogical thinking? I shall argue that
this sort of deadlock is possible-and for at least three different sorts of reasons.
1. Conflicting Moral Standards
Although persons agree with respect to their basic moral standards, they may disagree about what to say in
some case where these standards come into conflict, and their disagreement need not be attributable to any
disagreement in factual beliefs or to bad reasoning on the part of one of them. In such cases their disagreement
will be attributable to some disagreement in attitude, but, as I hope to show, this does not entail any disagreement
with respect to the basic moral standards that they accept.
Consider the following case. C is dying of some terminal illness and wishes to learn before he dies whether or
not his son is guilty, as charged, of murder. When C has only a few hours left to live he asks his friend, D,
whether his son has declared his innocence or confessed his guilt. Suppose that D knows that the son has
confessed. Now let us also suppose that, when confronted with this case, A and B disagree. A thinks that D ought
to tell C the truth about his son, and B thinks that D ought to tell a benevolent lie in order to spare C needless
mental anguish. A and B agree both that it is wrong to lie to people and that it is wrong to cause them mental
anguish; but in this case, where (they agree) these two moral considerations come into conflict, they weight them
differently.
The question now is this: Can they still be said to agree with respect to their basic moral standards, and, if so,
how can we explain their disagreement? One may be tempted to think that, since they disagree in their moral
evaluations of this case, they must disagree with respect to their basic moral standards. Given that consistency
requires one to universalize one's judgments, at least in the sense of judging similar cases similarly, doesn't this
mean that one of A's moral principles is that it is better to avoid lying than to avoid causing pain in cases like this,
whereas one of B's moral principles is that it is better to avoid causing pain in cases like this? This must be
admitted. A and B do have different moral beliefs about these kinds of cases, and these beliefs can be formulated
as general principles. But this does not mean that they accept different basic moral standards. Basic moral
standards are those that specify what kinds of factual considerations count as evidence, or reasons, for certain
kinds of moral judgments.
One might think that basic moral standards include not only criteria of relevance, but also principles of
priority for ranking these criteria. There is nothing wrong with this way of viewing things, so long as one realizes
that principles of priority are not themselves criteria of relevance and are really ancillary to the more fundamental
standards of relevance. In any case, what we are trying to see here is how people who agree on their basic
standards of relevance can nevertheless disagree in their moral evaluations, even though they do not disagree
about the facts and neither is thinking illogically. A and B agree that the fact that D's telling C that his son has
declared his innocence would be a lie is a reason for thinking that it would be wrong for D to tell C this, and they
agree that the fact that D's telling C that C's son has confessed would cause C a great deal of mental anguish is a
reason for thinking that it would be wrong to tell C that. They disagree only about which of these takes
precedence, or overrides, in certain kinds of circumstances. What, then, is the source of their disagreement about
this? There is certainly a difference of moral opinion here. But if it is due neither to their having different moral
standards nor different factual beliefs nor irrational thinking on the part of one of them, to what can it be
attributed? It seems to be due to a difference with respect to their basic values. A places a greater value on telling
the truth (or avoiding deception) in these kinds of cases, whereas B places a greater value on the avoidance of
pain. But doesn't this entail a difference with respect to their basic moral standards? The answer here is not so
obvious as it may seem to be. It may be that A places a greater value on telling the truth because he thinks it is
worse for a person to be deceived about a matter like this than to suffer the kind of mental anguish that would
result from knowing the truth. B holds the reverse to be true, thinking that a person is worse off suffering this kind
of pain. What about these more fundamental judgments of value? Do they indicate fundamental differences of
moral opinion?
It is not clear to me that these judgments about what is good or bad, or better or worse, for a person are any
more to be classified as moral judgments than are the judgments that it is worse for one to lose one's sight than to
lose one's hearing, that it is worse to be shot than to have one's money taken, or that it is worse to be lacking in
intelligence than athletic ability. These value judgments might serve as grounds for moral judgments-e.g. for
judgments that it is better (or worse) to treat someone in this way rather than that way-but they do not themselves
seem to me to be moral judgments. Such judgments about what is good or bad, beneficial or harmful, for human
beings can be used to support either prudential or moral judgments. For example, the judgment that it is bad for a
person to have decayed teeth can support both the judgment that one ought to brush one's teeth and the judgment
that parents ought to encourage their children to brush their teeth. Thus, in the case we are considering, it seems
plausible to say that, although A and B agree that what one morally ought to do (or what it is morally better to do)
 in a case like this is to do what is best for C, they disagree about what is best for C. If so, then what their making
different moral evaluations (and accepting different moral principles) with respect to cases like this shows is not
that they have different basic moral standards, but that they reach different results when they apply these
standards in conjunction with different non-moral value judgments.
In other cases the moral disagreement may stem not from their making different non-moral value judgments
but simply from their attaching different weights, or value rankings, to moral values that both accept. For
example, A and B may disagree about whether it is morally permissible to kill some innocent person when this is
the one and only way to save the lives of a hundred others. We need not suppose that either A or B denies the
relevance of the fact that the act necessary to save the one hundred lives is also a case of killing an innocent
person; nor need we suppose that either denies the moral importance of saving a hundred lives. However, A
thinks that it is worse to kill an innocent person than to fail to save a hundred lives; whereas B holds the reverse to
be true. Whereas B prefers to give priority to minimizing the number of lives lost (and perhaps, in general, gives
priority to considerations of utility), A prefers to give priority to respecting the inviolability of persons' (and, in
general, to considerations of justice).
Does this mean that A and B have different moral standards? Once again, it must be allowed that they have
different moral priorities, and hence different principles of priority for cases like this. But they do not employ
different criteria of relevance. B does not deny the relevance of not treating people merely as means, and A does
not deny the importance of minimizing the amount of harm. If either did make such a denial, we would have a
case of radical moral disagreement. But as things stand this seems to be a case of internal moral disagreement.
Indeed, there may be many cases like this where A and B would agree. B might agree with A that it would be
wrong to kill one person in order to save five others, and A might agree that if a million lives would otherwise be
lost, the one person must regrettably be killed. Nevertheless, because of the possibility of conflict between
different basic moral standards, agreement on basic moral standards cannot guarantee agreement in moral
evaluations-even when both parties are rational and fully informed.

2. Exceptions and Limitations

It is generally agreed that moral standards are not absolute but are subject to certain exceptions or limitations.
It is wrong to lie, except when doing so is necessary to avoid great harm. It is wrong to hurt another-unless, of
course, this is necessary to prevent him from killing some innocent person. It is wrong to fail to prevent serious
injury or death to someone, but not if this would involve an unreasonable sacrifice such as risking one's own life.
Moral standards are under- stood to be subject to certain exceptions for two reasons. One has to do with
minimizing possible conflicts between moral standards. By limiting moral standards in order to prevent them
from coming into conflict with other moral standards, we reduce the number of occasions on which we are faced
by moral dilemmas.
The other reason for acknowledging exceptions to moral standards has to do with the recognition that moral
standards may be limited, not only by other moral standards, but also by competing claims of self- interest. It is of
the essence of moral standards to specify certain restraints or restrictions on the pursuit of self-interest. But moral
standards are not generally thought of as completely overriding all claims of self-interest. For example, it is wrong
to hurt others; but if one can save one’s own life only by slightly hurting someone else, this would not seem to be
wrong. It is wrong to fail to save someone's life, but not at the cost of serious injury or death to oneself.
Thus, two persons may agree in accepting certain moral standards as specifying wrong-making
considerations, yet disagree about when some wrong-making consideration gives way either to some other wrong-
making consideration or to considerations of self-interest. Disagreements about when one moral consideration is
limited by, or allows an exception in favour of, some other moral consideration are disagreements about how to
resolve conflicts between moral standards. These have already been discussed.
Serious disputes may also arise about the extent to which some moral standard is limited by (or, perhaps it is
better to say, fails to override) considerations of self-interest. Most people would agree that it is permissible to kill
someone if this is the only way to prevent that person from unjustly taking one's own life. But suppose that killing
someone is the only way of preventing him from torturing oneself, or causing serious injury to oneself. In these
cases many would be inclined to think that a person is not morally required to refrain from killing an attacker
when the costs of so refraining are this great. Suppose, however, that killing someone is necessary to prevent him
from setting one's house on fire, or to keep him from stealing one's money? In these cases many would begin to
question whether the fact that complying with the moral proscription against killing would require these less
serious losses constitutes an exceptional circumstance.
There is apt to be even more disagreement with respect to the strictness of the moral requirement to prevent
harm to others. It is generally agreed that one has a duty to prevent harm to others- provided that this does not
involve an unreasonable sacrifice or excessive costs to oneself; but this agreement is compatible with considerable
disagreement about what constitutes an unreasonable sacrifice. For example, suppose that C needs a kidney
transplant, and D, his cousin, is determined to be the only suitable donor. D refuses to donate one of his kidneys
because he does not want the pain and inconvenience of the operation and also does not wish to incur the risks
that follow from being left with only one kidney. A thinks that D is wrong to refuse, but B disagrees. Here again
we need not suppose that A and B differ in any of their factual beliefs or that one of them is guilty of illogical
thinking in order to account for their disagreement. Nor need we suppose that they differ with respect to their
basic moral standards. A and B might agree that one is morally required to make certain modest sacrifices in
terms of one's own interests in order to prevent serious harm (e.g. death, injury, suffering) to others, but disagree
about what constitutes a modest sacrifice. One can view this as a disagreement about how to interpret a given
moral standard that both accept.
However, one could also say that, since they have different views on what level of sacrifice one is required to
make in order to prevent a certain degree of harm to someone else, this means that they really have different
moral standards for cases like this. In order to decide whether to describe their disagreement as a case of their
having different moral standards or as a case of their having different views on how to interpret a single moral
standard, we need to consider just where they disagree and what the source of their disagreement is. In this case
we are supposing that they agree that one is required to make some sacrifice in order to prevent harm to others,
and that they agree that there are limits to the level of sacrifice one is required to make; but they disagree about
when this limit is reached.
Let us suppose that A and B agree on how to rank levels of sacrifice on a scale from L 1 to L5. Let us also
suppose that they agree that sacrifices at L4 and L5 are more than what is required in order to prevent another's
death. (We may suppose that sacrifices at L5 include such things as risking one's own life, losing the use of one's
eyes or legs, and allowing oneself to be used as a slave, and that sacrifices at L4 include such things as contracting
some seriously debilitating but curable disease, losing a few fingers or toes, and being imprisoned for ten years.)
And let us also suppose that they agree that sacrifices at LI and L2 are required. (We may suppose that sacrifices
at LI include such things as getting one's face slapped, getting one's clothes wet or dirty, and being late for the
movies, and that sacrifices at L2 include such things as getting one's hands burned-but not seriously, having one's
property seriously but not disastrously damaged, and having to donate several days to some task one would rather
not perform.) Thus, we are now supposing that they agree on quite a lot. Where they disagree is with respect to
sacrifices at L3 (which includes such things as breaking one's leg, donating a kidney, and spending six months in
jail). A thinks that sacrifices at L3 are required; B denies this.
This disagreement has the appearance of an in-house dispute-not just because, on the present supposition,
they agree on so much, but because they agree on the basic criteria of relevance. They agree that the fact that one's
doing X would be a case of preventing serious harm to another is a morally relevant consideration. What they
disagree about is when this morally relevant consideration is overridden by or gives way to considerations of self-
interest-not in the sense that one is then justified in acting prudently rather than morally, but in the sense that one
is then not morally required to prevent the harm. We may suppose that even this disagreement is mitigated by the
fact that B agrees that sacrifices at L3 are at least morally praiseworthy. This in-house appearance of the
disagreement remains even when we suppose that the differences between A and B are wider. For example,
suppose that A thinks that sacrifices up to L4 are required, whereas B thinks that only sacrifices up to LI are
required. The disagreement between A and B approaches radical, as opposed to internal, disagreement only when
we suppose that B denies that one is required to make any sacrifice in order to prevent harm to another. I shall
consider this latter case when I discuss radical moral disagreement.
Now let us consider what the source of their disagreement might be and see whether this leads us to think
that they must be employing different basic moral standards. A and B agree on the ranking of various kinds of
sacrifices in terms of their degree of badness for the person making the sacrifice. Where they disagree, it seems, is
in their evaluation of a world in which everyone takes on the burden of making a certain level of sacrifice in
exchange for having their lives saved whenever this is possible through others making such sacrifices. A thinks
that we are better off living in a world in which the level is set at L3, whereas B thinks we are better off if it is set
at L2 or LI. A thinks that a world in which people made this kind of sacrifice (e.g. donating a kidney) in order to
prevent death to others would be a world in which people are better off than a world in which no such sacrifices
are made and such harm is allowed to occur, whereas B thinks that people would be better off living in a society
in which the reverse of this is true.
Does their making these different value judgments entail that they have different basic moral standards? Here
again I am inclined to think that what we have is a case of their interpreting differently the moral requirement to
prevent harm to others when this can be done at no great sacrifice of one's own interests, because they make
 different non-moral value judgments about when people are better, or worse, off. Compare: people are better off
living in a world in which people spend most of their time in pursuit of technological advances than they are in a
world in which people devote most of their time and energies to art and music. Or, a society which encourages
people to take a little time off now and then to enjoy pleasant pastimes is better than a society that encourages its
members to be workaholics. Or, people are better off living in a world in which natural beauty has been preserved
at the expense of unlimited industrial growth than they are in a world in which the reverse is true. I am inclined to
think that all of these are value judgments that might serve as grounds for moral judgments about how a society
ought to treat its members through its social policies, but they are not themselves moral judgments. Thus, it seems
preferable to describe A and B as employing the same basic moral standard, but interpreting it differently because
they make different non-moral value judgments.2 These different value judgments reflect different basic
preferences-a basic disagreement in attitude that may not be resolvable by any rational means. If so, the resulting
difference in moral opinion will not be resolvable by rational means either, and we shall have a case of moral
deadlock.
Thus, those who claim that when moral disagreement cannot be traced to differences in factual belief or
fallacious reasoning it must be due to basic differences in attitude seem to be correct. But this does not mean that
the parties to the dispute do not share common moral standards or criteria.

3. Questions about Applicability

People who are in agreement in accepting some moral standard may nevertheless disagree about what that
standard applies to. For example, they may agree that there is a moral prohibition against killing other human
beings, but disagree about whether this applies to a human embryo or to a foetus in the early stages of pregnancy.
People who accept this standard might also disagree about whether consistency in our thinking requires extending
the moral proscription against killing to certain higher animals as well. Again, people may agree that it is wrong
to lie or deliberately deceive others in order to take advantage of them, but disagree about whether a certain form
of advertising is a case of this. Or, they may disagree about whether consistency requires us to apply this standard
to certain cases of withholding information. These questions of interpretation seem to be different from the kinds
of questions about how to resolve conflicts between different moral standards, and about what counts as a
legitimate exception to some moral standard, that we have already considered; for here the question seems to be
whether the standard applies to the case in question in the first place, and not merely whether it is overridden by
some other consideration.
Suppose, for example, that C is a twelve-week-old foetus. A thinks that the fact that performing an abortion
would result in C's death is a wrong-making consideration (although A might admit that this consideration can be
overridden by other moral considerations), whereas B thinks that this fact is not even morally relevant (because B
does not see this as a case of killing a human being, in the relevant sense). A and B do not disagree with respect to
any non-moral facts. They agree, for example, about the physical characteristics of a twelve-week-old foetus and
about what its capacities are. They disagree only about the moral relevance of these facts.
Here it may seem as though A and B must disagree with respect to their basic moral standards, since they
disagree about whether a certain fact (causing the death of a twelve-week-old foetus) is morally relevant.
However, we can also view A and B as disagreeing about whether a moral standard that they both accept applies
to this case. They agree that the fact that an act is a case of killing a human being is a wrong- making
consideration; they disagree about whether killing a twelve- week-old foetus is a case of killing a human being.
Just what sort of disagreement is this? They do not disagree about whether the foetus has certain characteristics,
such that (both agree), if it had these characteristics, it would be (prima facie) wrong to kill it. Rather, they
disagree about which characteristics of human beings are such as to make it wrong to kill them. Although they
agree that it is wrong to kill human beings, they disagree about what it is about human beings that makes it wrong
to kill them.
Perhaps A thinks that what makes it wrong to kill a human being is that this deprives him of (or precludes his
having) certain valuable experiences-namely, those enjoyments of which all normal human beings (and perhaps
some animals) are capable and which make human lives worth living; whereas B thinks that what makes it wrong
to kill a human being is the fact that this frustrates a desire or preference to continue living. From B's point of
view, since a twelve-week-old foetus does not yet have a desire to go on living, killing it does it no harm. From
A's point of view, even though the foetus is not aware of what is happening to it and has no desire to continue
living, nevertheless, to be killed is for it to have something bad happen to it. Perhaps, then, A and B disagree
about whether the foetus' being killed constitutes a harm or evil to it. Although they agree that it is prima facie
wrong to harm any being that is capable of being harmed, they disagree about whether the painless killing of a
twelve-week-old foetus would be a case of harming it.
 One may suspect, however, that this is just a papering over of the important difference that separates A and B.
They do not just differ about whether killing a foetus is harming it; they disagree about this because they have
different conceptions of what constitutes harm. B thinks that one causes harm only by frustrating some desire or
preference. A thinks that one can also cause harm to a being by preventing it from ever having certain experiences
that it will desire to continue to have when it comes to have them. Thus, in so far as they have different criteria for
determining that an act is harmful, their agreement that it is wrong to cause harm is only superficial. Nevertheless,
this still seems to me to be a kind of in-house disagreement. In the first place, it takes place against the
background of a common assumption-namely, that there is something wrong about killing a human being.
Secondly, A agrees with B that the fact that an act frustrates a desire to go on living is a morally relevant
consideration; A merely wishes to extend this consideration in a way that B is not prepared to accept (by arguing
that the foetus will in due time come to want to continue the life of which it is now being deprived). Thirdly, A
and B are no doubt much more certain that it is wrong to kill a human being than they are about what it is about
this that makes it wrong. Their disagreement is about the rationale that explains a common moral standard that
both accept.
So far I have been concerned to show how serious and irresolvable differences are possible even among those
who agree with respect to their basic moral standards. This should not really be so surprising, when one considers
that these same sorts of disagreement can take an intrapersonal form-i.e. disagreement with oneself. One may
debate with oneself and not be able to make up one's mind about how to weight conflicting moral standards; or,
one may be perplexed about whether a moral standard that one accepts applies to a given case. Thus, the sources
of interpersonal disagreement that we have been examining are also sources of personal moral dilemmas and
individual perplexity.

Radical Moral Disagreement

It is now time to consider whether a more radical sort of moral disagreement is also possible. The most extreme
kind of moral disagreement that one can imagine-what I shall call total radical moral disagreement - would consist in a
dispute between persons who share no moral standards at all. Neither recognizes any fact as morally relevant that is also
recognized by the other (at least, not for the same reason). Hence, neither can hope rationally to convince the other of the
correctness of his opinion on some particular moral issue-should they disagree. It is not necessary for radical moral
disagreement to be as extreme as this, however. It seems that there could also be partial radical moral disagreement - i.e.
the parties to a particular moral dispute might agree on some, but not all, criteria of relevance.

Partial Radical Moral Disagreement

Total radical moral disagreement seems to be the more problematic of the two. If it exists at all, it would seem
that it must be extremely rare. Let us therefore first consider partial radical moral disagreement. It seems more familiar
and easier to imagine. Indeed, examples of it may readily suggest themselves to one. Suppose, for example, that A claims
that homosexual behaviour is morally wrong, and B denies this because, he says, such behaviour causes no harm. A
claims that it is none the less wrong because it is unnatural and contrary to the will of God. But B denies that the fact that
an act is unnatural or contrary to the will of God is, in and of itself, morally relevant.
Suppose that B asks A to explain why he thinks such considerations are morally relevant. Why does the mere fact
that an act is not the sort of act that is compatible with the genetic programming of normal human beings make that act
wrong? I suspect that A-and people like him- think that this is relevant because they think that an act's being contrary to
the normal genetic programming of human beings is proof that it is contrary to God's will (since our genetic programming
reflects God's intentions for us). Then why does A think that its being against God's will is relevant? Perhaps A, and
others, think this because their very conception of God is the conception of a being who is totally opposed to what is bad
for human beings and desires only what is beneficial for human beings, and who wills and commands accordingly. Thus,
if God wills us not to do it, it must be harmful to human beings-even if we cannot see how this is so. If this is the case,
then A and B do not really disagree with respect to their basic moral standards, since B is prepared to admit that if such
behaviour really were harmful this would be a ground for thinking it wrong.
Suppose, however, that A insists that its being somehow unbeknownst to us harmful is not what makes it wrong.
It is wrong, he claims, just because the one and only supreme being happens to dislike it. God doesn't dislike it because it
is harmful (although he does of course dislike whatever is harmful); he just happens to dislike it and wishes that it not
happen. And it is wrong to disobey the wishes of God, whatever they are.
B might persist here and ask A why he thinks it is wrong to disobey even the most gratuitous wishes of God. Is it
because we owe him a debt of gratitude for having created us and benefited us in other ways, and ought therefore to
indulge even his most capricious desires? If this is the reason, then A is appealing to a commonly accepted moral standard
that B will no doubt accept (even though he might deny the factual assumptions that such a being exists and has created
us). B will no doubt agree that one ought to be kind to and try to avoid frustrating the desires of a good person who has
greatly benefited one-although insisting, perhaps, that this consideration might be overridden by other moral
considerations. There is no radical moral disagreement here.
I would suggest, moreover, that if A continues to reject all attempts to link his alleged moral criterion to other
commonly recognized criteria, his position will eventually become unrecognizable to us as a moral position. Suppose, for
example, that he denies that God created us or has any interest at all in our welfare - being completely indifferent to this.
Now consider another example that seems to be a case of partial radical moral disagreement. Suppose that A
claims that X is wrong and cites as a reason the fact that to do X would be to tell a lie. B claims, not that it is morally
permissible to tell a lie in this case because other morally relevant considerations (the need to save a human life, for
example) outweigh this consideration, but claims rather that the fact that it would be a case of lying is not even relevant to
the question of whether it is morally wrong. In this case some might wonder whether B really understands what it is for an
act to be morally wrong. But B might convince us that he does understand what moral wrongness consists in by telling us
that he understands that if an act is a case of killing someone - or hurting someone or cheating someone or breaking a
promise or failing to prevent injury or death to someone - that would be a reason for judging it to be wrong. But he fails to
see why the fact that an act is a case of lying is, in and of itself, relevant.
Perhaps B's view is that lying is not wrong in itself, but only in so far as it is also a case of harming someone. He
may claim that the facts that he admits to be morally relevant are all cases of harming someone, but that lying is not
necessarily a case of harming someone. Being deceived is not something bad in itself; it is bad only in so far as it (almost
always) brings other bad consequences with it. Here again it is not clear that we have a genuine case of radical moral
disagreement between B and A, who believes that lying is something inherently wrong. For A and B seem to agree on a
more fundamental criterion. They agree that the fact that one is doing something that harms another is a wrong-making
consideration. They simply disagree about whether lying is necessarily harmful. Thus, this case seems no different from
certain cases of internal moral disagreement that we have already considered.
Suppose, however, that B admits that it is bad for one to be lied to, but still denies the moral relevance of the fact
that an act is a case of lying. In this case he is committed to denying in general the moral relevance of the fact that an act
is a case of harming someone. The disagreement between B and A now approaches total radical moral disagreement, for it
is hard to imagine what common court of appeals is left for them. Let us postpone the discussion of this more extreme
kind of moral disagreement and concentrate for the time being on trying to find a genuine case of partial radical moral
disagreement.
So far we have considered a case where B insists on a criterion of relevance that A rejects without necessarily
rejecting other commonly recognized criteria that A might accept, but resisting all attempts to link his criterion to these
other criteria. We have also considered a case where B denies the relevance of some commonly agreed upon criterion that
A appeals to, but without rejecting certain other generally recognized criteria. It is also possible that B might admit the
relevance of some commonly recognized standard (thus partially agreeing with A, who accepts it), but claim that this
consideration is outweighed by some other consideration whose moral relevance A denies. In order to make the
disagreement seem as radical as possible consider some of the more bizarre examples that have been suggested in the
literature. Suppose that B agrees that doing X would be a case of killing someone and that this is a reason for thinking that
X is wrong, but insists that this consideration is outweighed by the fact that doing X is necessary to bring about the
aesthetically pleasing effect of red blood on green grass. Or, perhaps B is not so bloodthirsty as this suggests, but merely
thinks that doing X is the only way in which he can avoid having to step on the lines of a pavement.
It is, of course, possible (as Philippa Foot has suggested) to under- stand B's appeal to these other considerations
as an appeal to moral considerations if one provides the proper background.4 It might be, for example, that B thinks that it
is necessary to produce the effect of red blood on green grass in order to placate the gods, who love the sight of this more
than anything else. B might think than unless this is done the gods will send a plague that will wipe out half the human
race. But if we offer this explanation of B's belief, then, once again, it is not clear that we have a case of radical moral
disagreement. For A might well agree that if B's perverse non-moral beliefs about the gods were true, and it were really
necessary to sacrifice some human lives to save a far greater number, then this would be morally justified. Or, A might
disagree about how the conflicting moral considerations in this case are to be weighted, holding that we are not justified in
sacrificing even a small number of innocent human lives in order to prevent even a large number of deaths to others. In
either case what we have is an internal moral disagreement, since in this case A and B agree on their ultimate criteria of
relevance.
Suppose, however, that B claims that an act's producing the effect of red on green (or being a case of avoiding
stepping on the lines of a pavement) is in and of itself morally relevant-that no further background information is
necessary to support this claim. At this point it seems to me that B's view becomes unrecognizable as a moral position.
Moreover, if we suppose that B has reasons for thinking that one is morally required to produce the effect of red on green
(oi to avoid stepping on the lines of a pavement), but these have no relationship to such things as avoiding harm or loss of
freedom to human beings, instead deriving solely from the consideration that red on green looks beautiful (or that
someone's stepping on the lines of a pavement is unsightly), then it seems to me that the correct description of his view is
not that he believes that it is morally permissible to kill others when this is necessary to produce the effect of red on green
(or to avoid stepping on the lines), but that he believes that in these situations aesthetic considerations ought to override
moral considerations. In that case, once again, we have no disagreement about what is morally required, but only a
disagreement about whether or not moral considerations are outweighed by other considerations. One can call this a moral
disagreement, since they are disagreeing about a matter (that both agree to be) of moral importance. But they are not in
disagreement about what is morally required in this case, although they do disagree about what one ought, all things
considered (moral as well as non-moral), to do.
Thus, cases of genuine partial radical moral disagreement that are clearly distinguishable from cases of internal
moral disagreement are not easy to construct. Although at first sight it appears that such cases not only can but do
commonly occur, a closer inspection reveals either that such disagreements are not quite so radical as they appeared to be,
or else that the real difference between the parties to the dispute is a difference of opinion about non-moral values (as in
the disagreement about the intrinsic badness of being deceived) or about how certain moral values are to be weighted
when they conflict with non-moral values (as in the disagreement about whether the beauty of red blood on green grass is
more important than the moral wrongness involved in slaughtering people in order to bring this about).
There is one case that gives me pause, however. Suppose that A takes the position that, although it is certainly
wrong to harm others, it is not wrong to fail to aid them when they are in need-no matter how serious their need and how
minimal the sacrifice required to help them; whereas B contends that both are wrong. This does seem to me to be a
genuine case of partial radical disagreement, and not just a case of internal moral disagreement in disguise. Nor does A's
position seem unintelligible as a moral viewpoint.
However, even though this seems to be a genuine case of radical moral disagreement, the disagreement is not as
radical as it might be-just because it is not total radical moral disagreement. Thus, because A agrees with B that it is
wrong to cause harm to others, there is a common court of appeals here. B might argue that A is being inconsistent in
admitting that it is wrong to cause harm while denying that it is wrong to fail to prevent it; or B might contend that the
distinction between these is not so clear-cut as A imagines and will not bear the weight of his moral position. However, I
am sceptical about the success of any such arguments. And if such arguments prove to be unsound, I will have to admit
that partial radical moral disagreement is possible- even if it is not as common as one might at first suspect.
Notice, however, that partial radical moral disagreement is still a kind of in-house dispute, because of the
common commitment to at least some moral standards to which both sides appeal. This makes certain kinds of arguments-
namely, those appealing to the requirements of coherence and consistency-possible. It is often difficult to assess the
cogency of such arguments; and therefore it is difficult to be sure that one of the disputants is not being illogical in his
thinking. When this observation is considered together with the difficulty we have encountered in trying to construct
examples of genuine partial radical moral disagreement, such disagreement remains elusive and problematic.

Total Radical Moral Disagreement

The most extreme kind of moral disagreement that one can imagine is a case where the disputants do not agree on
any criteria of relevance. There are no basic moral principles that they share. They disagree about whether some act is
morally wrong and yet neither side accepts any facts as relevant to this question that the other also accepts. In this kind of
case one might wonder whether they are really disagreeing about anything. For, if they are not disagreeing about whether
the act in question satisfies certain criteria (or about how to interpret such criteria), what are they disagreeing about? Can
they even be said to be talking about the same thing when they say that an act is morally wrong?
There are those who argue that they can-that the fact that they employ radically different criteria to govern their
applications of the terms 'right' and 'wrong' does not entail that they cannot mean the same thing by these terms. It is
sufficient that each uses these terms to express his universalizable and overriding prescriptions (or permissions). So long
as one does this one is free to employ any criteria one chooses, since no one is committed to employing certain criteria
just by considering the question of the moral wrongness of an act. One can be committed to the acceptance of certain
criteria only by one's own basic attitudes or preferences. And if one's basic attitudes and preferences are perverse enough,
one will be able to accept or reject any criterion of relevance.
On this kind of view, B can disagree with A not only about the relevance of the fact that an act is a case of lying;
B can refuse to acknowledge that the fact that an act is a case of harming or violating the interests of others is, in itself, a
wrong-making consideration. Whereas A holds that one is morally required to respect the interest of others, B can deny
this and hold instead that one is morally required to maximize one's own interests, regardless of the consequences for
others. Whereas A holds that one is morally required to make small sacrifices of one's own interests when this is
necessary to prevent serious injury or death to someone else, B might claim that, unless such behaviour can somehow be
shown to be in one's own long-range interest, it is not only foolish but immoral.
 This kind of extreme radical moral disagreement makes perfectly good sense on the sort of view we are now
discussing. But does it really make perfectly good sense? Does it really make sense to suppose that someone might think
that, if one possessed Gyges' ring, one would be morally required to rob and cheat and kill those who stand in one's way?
Even if we accept Plato's contention that such behaviour can never really be in one's long-range interest, does it make
sense to suppose that someone who mistakenly believed that it was could believe that such behaviour is morally required?
This is not to deny that it makes perfectly good sense for someone to claim that such behaviour is rationally required, that
it would be foolish for a wearer of Gyges' ring to fail to advance his own interests in these ways. But it is one thing to
claim that a wearer of Gyges' ring ought not to do what he is morally required to do; it is quite another thing to claim that
a wearer of this ring is not morally required to refrain from trampling on the interests of others in the pursuit of his own.
We can allow that A and B might disagree about whether or not one ought to act morally when it is not in one's overall
interest to do so. This would certainly be a very radical sort of disagreement. However, it would not be a disagreement
about what one morally ought to do.
Thus, I am inclined to think that total radical moral disagreement is not really possible, because the acceptance of
certain criteria of relevance is a necessary condition of considering the question of whether an act is morally right or
wrong. This is not to deny that serious moral disagreement is possible, and indeed that moral deadlocks can frequently
occur-especially about such issues as abortion. As we have seen, irresolvable differences in basic attitudes or preferences,
often reflected in different non-moral value judgments, may lead people to apply these criteria in different ways so as to
arrive at opposing moral beliefs. Very deep and fundamental moral disagreements are certainly possible, and these are
often irresolvable by any rational means of persuasion. But to admit this is not to allow that there are no ground rules at all
for moral discussions-to allow that a person is free to make whatever he likes count as evidence for moral judgments,
provided of course that he is able to accept the appropriate universal prescriptions. What criteria of relevance one is
committed to employing in answering a moral question is determined by the nature of the question, and not by the
attitudes and preferences of the speaker.

Conclusion
I have tried to accomplish two things in this paper. First, without denying the possibility of genuine moral
deadlock, I have tried to raise serious doubts about the possibility of truly radical moral disagreement. I have argued that
disagreement is recognizable as moral disagreement only in so far as both parties to the dispute accept certain criteria of
relevance. Second, I have tried to show that the acceptance of these criteria is nevertheless compatible with serious and
considerable disagreement about how such criteria are to be interpreted and about how they are to be weighted when they
come into conflict either with each other or with non-moral considerations. Such disagreements are rooted in differences
in basic attitudes and preferences, and for this reason they are often irreconcilable by any rational means.
It is important to realize, however, that we do not have to deny the objectivity of moral judgments in order to
account for the possibility of moral deadlock. To hold that moral judgments are objective one need not claim that there is
some decision-procedure for resolving all moral questions, such that all who agree on the facts and reason correctly must
arrive at the same answer. All that one need maintain is that there are certain moral standards (certain criteria of
relevance) that must be employed by anyone who wishes to make moral judgments. Indeed, I have argued that moral
disagreement is possible only between those who agree (at least, substantially) with respect to their basic criteria of
relevance, since without this agreement on what counts as evidence for moral judgments there can be neither moral
agreement nor disagreement.