In this paper I will explore the question of the existence of moral facts, paying particular attention to the arguments of two moral skeptics, Gilbert Harman and John Mackie, and to responses to their arguments. In doing so, I will argue against the existence of moral facts. Although I feel that the arguments of Harman and Mackie both ultimately fail to disprove the existence of moral facts, I believe that their arguments provide adequate grounds for skepticism about moral facts. That is, I will not rule out the possibility of the existence of moral facts, but I maintain that the moral realist's belief in moral facts is unjustified.
I do not believe that my position presents any great problem with morality. Even though I do not believe that moral facts exist, I believe that the concept of moral facts is tremendously useful. The concept of moral facts, so prevalent in our moral discourse, allows for the appearance of objectivity in ethics. I believe that it is this concept of moral facts, not the existence of the facts themselves, which enables us to have a system in which morality can be taken seriously.
I will develop my position more fully in this paper, which I have divided into sections as follows: In the first section, I will define moral facts and consider some reasons why we want moral facts. I will describe the positions of moral realists and moral skeptics. I will briefly sketch some of the main objections to moral realism and will outline some possible responses to these objections.
In the next three sections, I will turn to Gilbert Harman's ``Ethics and Observations'', in which Harman attempts to distinguish explanations in science from those in ethics in order to demonstrate a special reason for skepticism about moral facts. I will present both Harman's argument and the replies formulated by Nicholas Sturgeon and Geoffrey Sayre-McCord. In the next two sections, I will focus on John Mackie's ``The Subjectivity of Values'' and David Brink's ``Moral Realism and the Sceptical Argument'', paying particular attention to the argument(s) from moral relativity and disagreement. Having explored these five works, I will then conclude the paper by reassessing my thesis concerning moral facts.
Moral Facts, An Introduction:
Are there any moral facts? Are there any good reasons to believe in the existence of moral facts? And does it even matter if there are moral facts? These are the central questions of this paper. To understand and arrive at answers to these questions, it is necessary to consider the views of those who answer the first two questions affirmatively - moral realists - and those who do not - moral skeptics.
In brief, the thesis of moral realism is that there are moral facts. According to the moral realist, to say that a course of action is morally right or wrong is to assert that there is an objective fact that the particular course of action is, in fact, right or wrong. A particular course of action is right (or wrong) even if no one expresses their belief or thinks that it is right or wrong. That is, moral facts are independent of moral claims.
For moral realists the existence of such independent moral facts is extremely useful. Moral facts provide objective truths about moral matters. The truth or falsity of moral claims can always be determined by considering the corresponding moral facts. Additionally, because moral facts are objective, they allow us to have meaningful moral discussions in which moral statements exert obligatory force. Moral facts allow us to decide between two possible courses of action, to provide justification for our actions and, due to their universal objective nature, to judge the actions of others.
As moral facts are so useful, it is no wonder that moral realists, such as Sturgeon, Sayre-McCord, and Brink, worry about the possible dangers of a world without moral facts. Without moral facts, how can we arrive at correct moral decisions? If our moral claims are merely subjective, how can we judge the actions of others? And how can the apparent normativity of ethics and the motivating force of moral claims by explained without moral facts? We want the ability to arrive at correct moral decisions; we want to be able to judge the actions of others; we want meaningful moral discourse to be possible. And according to the moral realist, all of this requires the existence of moral facts.
However, I am not convinced that the non-existence of moral facts needs to be a problem for morality. I believe we may well be able to do without moral facts. We may even act as if there were moral facts without the actual existence of moral facts. This is a view which is shared by many, though not all, moral skeptics. According to moral skepticism, the existence of moral facts is deniable or at least doubtful. The views of Harman and Mackie, then, fall into this category. It is the analysis of their arguments against moral realism on which this paper will focus.
Having considered reasons why we might want moral facts, I turn now to arguments against the existence of moral facts. I will only enumerate and briefly discuss a few of these arguments here:
Against the second of these arguments, that there is no internalism so there are no moral facts, it can be argued that: (1) it might be the case that there really is internalism so there is no problem with the existence of moral facts, or (2) that it might be possible for moral facts to exist without requiring internalism.
Against the third of these arguments, that moral facts are queer entities, it may be argued that: (1) moral facts may be non-queer natural facts, or (2) moral facts may supervene upon natural facts.
And finally, against the fourth of these arguments, that moral disagreement is inconsistent with the existence of moral facts, any of the following can be argued: (1) that what appears to be moral disagreement is really just disagreement about underlying non-moral facts, (2) that moral disagreement is not genuine but may just be temporary and resolvable, or (3) that moral facts may exist but be unknown, so that agreement and/or disagreement is seen to be irrelevant to the existence of moral facts.
In the remainder of the paper, I will address and more fully develop these arguments, particularly the first and last. In order to do so, I will focus on the views of Harman and Mackie respectively.
Harman and Observation:
Gilbert Harman's position in ``Ethics and Observation'' is not one of general skepticism, but one of only moral skepticism. While Harman is a skeptic about ethics, he is not a skeptic about science and scientific facts. As such, he attempts to demonstrate that there is a good reason to deny the existence of moral facts and that this reason does not similarly rule out the existence of scientific facts. To do so, Harman points to a distinction between moral observation and scientific observation.
Harman begins by asking the following question: Can moral facts be tested and confirmed in the same way that scientific principles can? (Harman, 119) The answer, Harman believes, is no, and it is from this distinction between science and ethics that Harman thinks his limited skepticism follows. Central to Harman's thesis are observations and explanations. Observation, says Harman, plays a role in science that it doesn't play in ethics; assumptions about physical facts are required in order to explain scientific observations, but such assumptions are not needed in explanations of moral observations.
Harman offers an example of a scientific observation and explanation. A physicist, upon seeing a vapor trail in a cloud chamber, may say ``there goes a proton''. According to Harman, the best explanation of the physicist's thinking there is a proton is that there really is a proton (Harman, 121). Harman believes this is not so in ethics. That is, in moral observations or claims, the truth or falsity of the claims seems irrelevant to the explanations as to why they were made. The actual rightness or wrongness of an act does not have an effect on the perception of rightness or wrongness (Harman, 122).
Here Harman offers a second example, concerning someone who observes children setting a cat on fire. The observer makes the moral comment: ``That's wrong!'' (Harman, 120). Harman explains that he makes this observation not because there is a moral fact that setting cats on fire is wrong, but because he has the thought that setting cats on fire is wrong. The existence of a moral fact of wrongness seems irrelevant to the observation or thought that setting the cat on fire is wrong (Harman, 122).
I am not convinced by these two examples. First, if there is reason to be skeptical about moral facts, it seems there may also be reason to be skeptical about scientific observations and explanations. Consider the proton case: Isn't it possible that the best explanation for the physicist saying that there is a proton is simply that the physicist thinks there is a proton? The reason the physicist says there's a proton is that he accepts a correlation between vapor trail and protons. His accepted correlation may in fact be false. In such a case the explanation for the physicist's claim cannot be that there really was a proton. Additionally, Harman's claim that the actual wrongness of an act has no effect on it being thought wrong seems fishy. Couldn't it be the case that one thinks setting a cat on fire is wrong simply because it is wrong, just as we might say the physicist thinks there's a proton because there is a proton?
Of course, this question, which I will consider in the next section, only suggests that the best explanations of moral observations might appeal to moral facts, not that the best explanations do appeal to moral facts. It might be that the observer judges the cat-burning to be wrong because it is a societal convention that cat-burning is wrong and/or that he has been educated to believe this. In this case, the explanation that cat-burning is judged to be wrong is due not to the existence of moral facts, but to the moral psychology of the observer.
Let's suppose, as Harman believes, that objective moral principles or facts really are of no help in explaining why an observer thinks an act is wrong. Harman then concludes that since moral facts are of no explanatory help, they are unnecessary. Worse, the existence of moral facts would be a kind of ontological clutter. From the explanatory irrelevance of moral facts, it may be seen to follow that moral facts do not exist.
I find this argument troubling for two reasons which I will explore in the next two sections. First, I am unconvinced that moral facts are explanatorily irrelevant. Second, I am unconvinced that the explanatory relevance or irrelevance of moral facts constitutes grounds for determining whether or not they exist.
The Explanatory Role of Moral Facts:
In this section I will explore the claim that moral facts do not play an explanatory role. In doing so, I will consider the argument offered by Sturgeon in his ``Moral Explanations''.
Harman's application of the explanatory criterion - that belief that the existence of moral facts is justified if and only if moral facts provide reasonable explanations of moral observations - depends upon the denial of the claim that moral facts have explanatory power. Whereas Harman suggests that moral facts don't provide reasonable explanations and therefore are likely not to exist, Sturgeon believes moral facts do provide explanations. Sturgeon believes that often the fact that a particular act is wrong explains why we think it is wrong (Sturgeon, 239).
To illustrate this, Sturgeon offers several examples, one of which I will sketch here: We make the claim that Hitler's actions were wrong because of the actual wrongness that we perceive. It is not merely that Hitler's actions are perceived to be wrong, but that they are wrong. Similarly, one might explain that we think slavery is immoral because it is immoral. That is, the inherent wrongness of these actions forms part of the explanation for why we believe that they are wrong. (Sturgeon, 245)
I do feel, for instance, that Hitler's actions were wrong and that this view is (nearly) universally accepted. It is in making moral claims like this that we want to appeal to objective moral facts. Furthermore, it seems descriptively accurate to say that we feel Hitler's actions were wrong because they simply were wrong.
However, I am worried by explanations like this. Sure, we may want to say that Hitler's actions were objectively wrong. In fact, people do quite often appear to be appealing to moral facts in order to explain such moral claims. We appeal to moral facts because we want to be able to apply our standards of morality. But just because we believe that something is objectively wrong, does not mean it really is the case that it is objectively wrong.
How then do we know if something is objectively wrong? Do our explanations that an action, such as cat-burning, is wrong depend on moral facts? Only if, says Harman, moral facts are relevant to the explanation as to why we think the action is wrong. As a way to evaluate claims of such relevance, Sturgeon offers a counterfactual test, which I find quite compelling. For each situation in which moral judgment of the nature ``that's wrong!'' arises, e.g., cat-burning, Hitler's actions, etc., consider a counterfactual situation nearly identical to the actual one, but in which the act or acts performed are not wrong. (Sturgeon, 246-247)
Would the moral observer still think that the actions were wrong, even if they were not? That is, would the observer still judge cat-burning, for example, to be wrong even if it wasn't wrong? There are two cases: If moral facts are irrelevant to our explanations of moral observation and judgment, then the actual and counterfactual situations differ only in the existence or non-existence of a moral fact pertaining to a particular action. One would still judge the action to be wrong. That is, one who would judge cat-burning to be wrong if it were the case that cat-burning was objectively wrong would also judge cat-burning to be wrong without the existence of an objective truth about cat-burning.
Alternatively, as Sturgeon suggests, the actual and counterfactual situations may not only differ in the existence of a moral fact. Instead, the counterfactual situation may look quite different from the actual. Suppose that cat-burning is objectively wrong and you see some children on the street burning a cat and think their action is wrong. Now consider a situation as close to this one as possible, but in which their action isn't wrong. This is not to say that cat-burning is suddenly right; the moral fact that cat-burning is wrong remains. Then if the actions of the children in the counterfactual situation aren't wrong, their actions must be something quite different from cat-burning. Sturgeon concludes that moral facts are not irrelevant to our explanations of moral observations; the existence or non-existence of particular moral facts does change our observations. (Sturgeon, 249)
Sturgeon's counterfactual-based argument is supposed to suggest that there are instances in which the existence of moral facts is relevant and plays an explanatory role in our observations. However, I am not sure this follows. Instead, I think his argument suggests that if we think there are moral facts, this thought influences our moral observations and, as such, plays an explanatory role. The argument is not one about the existence of moral facts, but of the perception of existence and knowledge of particular moral facts.
Problems with the Explanatory Criterion:
Even if it is valid, does Sturgeon's counterfactual test restore the thesis of moral realism? In `Moral Theory and Explanatory Impotence'', Geoffrey Sayre-McCord says that it does successfully refute the objection that moral facts are irrelevant, but that it does not refute a related challenge: the objection from explanatory impotence. This is the charge of the moral skeptic that although moral facts may be relevant in providing an explanation for moral observation, the observation could have been otherwise explained just as well without appealing to moral facts, so moral facts are still unnecessary. And if moral facts are unnecessary, there is no reason to believe that moral facts exist. (Sayre-McCord, 272)
To illustrate this point, Sayre-McCord provides the following analogy: Suppose we have learned to formulate witch-explanations of occurrences. For example, we might suggest that the sky is blue on some days because blue is the color preferred by the witch who chose the color. On days when the sky is gray, another witch chose the color. Even as we are arriving at such witch-explanations, we need not actually believe in witches. If we can provide another equally good or better explanation as to why the sky is blue, then we need not believe that witches exist. We can create numerous (false) explanations by appeal to witches, but that does not mean that witches exist. (Sayre-McCord, 273)
As with witches, Sayre-McCord suggests that if comparable or better explanations can be formed without appeal to moral facts, moral facts are explanatorily impotent and we are not justified in believing in the existence of moral facts. This argument demonstrates, one might say, that moral facts don't exist. This suggestion is in accord with a convention in metaphysics to prefer minimal ontologies to ones containing more entities. That is, philosophers want to admit as few different kinds of entities as possible to the set of kinds of existent things.
The appeal of this argument for me is not ontological minimalism, but my idea of justified beliefs. That there might be witches or moral facts does not seem to be an adequate reason to believe that there are witches or moral facts. Instead, I believe that the skeptic's response is the appropriate one. Rather than provisionally accept the existence of moral facts without proof, I believe that the default position is to question, not accept, the existence of moral facts or any other insufficiently justified hypothesis.
As such, I will now question the explanatory criterion which forms the foundation of Harman's argument. Sayre-McCord provides two statements of the criterion. According to the strong form of the explanatory criterion, a hypothesis should be believed if and only if the hypothesis plays a role in the best explanation we have for our making the observations that we do. For moral facts, this becomes: moral facts and/or their existence should be believed if and only if moral facts and/or their existence play a role in the best explanation we have of our making the moral observations that we do. The weak form of the explanatory criterion drops the ``if'' part of the criterion, maintaining that a hypothesis should be believed only if (not if and only if) it plays such an explanatory role.
That is to say, if we accept the explanatory criterion, then there are scientific facts if they figure into the best explanation for at least some of the accurate observations that we make. Similarly, there are moral facts if they figure into the best explanation of at least some of the accurate observations that we make.
I am not at all convinced by Harman's argument from explanatory criterion. I do not think, as I will explain, that a type of entity need exist to be useful. So the fact that moral facts are useful, as I think they are, does not convince me that they exist. Nor do I think that for belief in some type of entity to be justified, and therefore for it to exist, it must play an explanatory role.
Furthermore, I believe that this argument, if it is to demonstrate the existence or non-existence of moral facts, must rely on a fallacy by confusing epistemology and metaphysics. If belief in moral facts is justified, this does not mean they exist. In science as well as ethics, people often do have justified, but false, beliefs. Similarly, if the belief in moral facts is unjustified, this does not mean that moral facts don't exist. I suspect that moral facts may exist, but be unknown to us. If we have no knowledge of moral facts, clearly the existence of moral facts cannot play an explanatory role. But surely that doesn't mean that moral facts don't exist; it only means that we don't know that they exist. That we have no knowledge of a particular entity, be it moral facts or anything else, does not imply non-existence.
However, I believe there is an important claim which can be salvaged from this fallacy. Although having no knowledge of something is not a reason to conclude that it does not exist, having no knowledge of something is a reason to seriously consider the possibility that it does not exist. The argument that moral facts may not play an explanatory role does not disprove the existence of moral facts, but it does provide reason to doubt the existence of moral facts. This more or less summarizes my response to the question of the existence of moral facts: while I do not rule out the possibility of the existence of moral facts, I am unconvinced of their existence. As I have said, I believe skepticism is the appropriate response when adequate justification is unavailable.
Finally, I will focus briefly on my claim that even if belief in moral facts is justified, this belief does not mean that moral facts exist. I believe that moral facts are extremely useful and may be the kinds of things that we want to have in a moral system, but I am not convinced that moral facts exist. Moral facts do not exist just because we want them to, nor do I believe that moral facts need to exist to be useful. It is the idea of moral facts that is useful, in much the same way that accepted conventions are useful.
There are many non-moral parallels that support my thesis that something need not exist to have useful explanatory power. For example, we refer to colors in some of our non-moral explanations. But colors don't exist, or at least not as primitives; they are reducible to particular wavelengths of light. We make reference to colors in explanations not because colors exist or because we must rely on colors for our explanation, but because appeal to colors yields simpler, better explanations.
Similarly, where we refer to water in explanations of our observations, we may instead rephrase the explanation in terms of H2O. Where we refer to pain, we may instead refer to neurophysiological brain states (Sayre-McCord, 274). But we do not because it would be tedious to do so. It is notationally and conceptually useful to us to refer to water, pain, and color. I do not believe this means that water, pain, and color exist as their own entities independently of H2O, brain states, and wavelengths; instead, there is a relation of supervenience. In much the same way, I suggest that reference to moral facts is useful in explanations even if there are no moral facts or if moral facts can be reduced to something more fundamental.
I believe, as Sayre-McCord does, that moral facts are useful in practice when explaining observations. As we may explain rainbows in terms of colors, oceans in terms of water, and symptoms in terms of pain, so also may we rely on moral facts and virtues to explain observations. Moral facts, says Sayre-McCord, allow us to explain such observations as Mother Teresa's goodness having won her a Nobel prize and the death of millions as resulting from Hitler's inhumanity (Sayre-McCord, 275). And this is, as I suggested earlier, what we want from moral facts.
However, I do not believe that moral facts need to exist to be useful in this way. My primary objection, then, is not that moral facts do not provide explanations, but to the explanatory criterion and its application here. I do not believe that the perceived usefulness of the concept of moral facts implies their existence, any more than I think the perceived uselessness of moral facts implies their non-existence. I maintain that there is no valid correlation to conclusively demonstrate the existence or non-existence of moral facts from arguments for or against explanatory relevance and potency.
Mackie and Moral Skepticism:
I will now abandon the argument from explanatory power and consider instead the views of John Mackie, as presented in his ``The Subjectivity of Values''. Here Mackie provides something of a positive account for moral skepticism against the claims of moral realists that moral facts are a necessary part of our system of ethics. Additionally, Mackie raises several objections to the existence of moral facts. These arguments include an argument from internalism, two arguments from queerness, and an argument from moral relativity and disagreement. Of these, I will focus in the next section on the argument from relativity and moral disagreement. I believe that this argument presents the greatest challenge to moral facts.
Like Harman, Mackie is a moral skeptic. His main thesis is that there are no moral facts, and he attempts to defend this thesis by raising objections to the views of moral realists. Also like Harman, Mackie's skepticism is limited to ethics alone. That is, both Harman and Mackie believe that moral facts don't exist and, at the same time, that non-moral, particularly scientific, facts do. As such, both feel that there is a fundamental difference between science and ethics. For Harman, this difference is, in part, the different roles their respective facts play in explaining scientific and moral observations. For Mackie, this difference is, in part, the differences between the possibilities for resolution of scientific and moral disagreements.
Before turning to Mackie's arguments, I would like to discuss his views on moral objectivity more generally. Although Mackie does not believe there are moral facts, he does point out that moral facts are part of our vocabulary. There is, he says, an assumption that there are objective values and moral facts. Ordinary moral judgments do include a claim to objectivity. When we say that someone should or ought to do something, we intend to communicate an objective claim. One way to understand this is to appeal to moral facts. It seems that this really is what we do, as Mackie suggests that the assumption that moral facts exist is incorporated into the very meaning of many ethical terms. (Mackie, 108)
How, then, does Mackie reconcile this apparent dependence on moral facts with his belief that moral facts don't exist? Mackie's theory takes the form of an error theory: Although most people making moral judgments claim to be pointing to something objectively prescriptive, this claim is false (Mackie, 109). There are no moral facts but, as I suggested in the beginning of this paper, we act in our everyday moral discourse as if there were. While this may seem a bit paradoxical, I do not believe there is any contradiction at all. The first part of the statement, that there are no moral facts, is a metaphysical claim about existence. The second part, that we act as if there were moral facts, is a description about our actions. Much as we may believe in something that doesn't exist, so too may we act as if something, which doesn't exist, does.
How is it that we come to have objectivity in ethics without the existence of moral facts? This is precisely the question that Mackie attempts to answer at the end of his essay. Mackie considers two different origins for such a system. The first relies on positive or natural law. According to this explanation, ethics is a system of law from which the legislator has been removed. In the case of religion and societies with religious roots, the legislator may be God. Mackie's second explanation is that such a system of ethics arises naturally from society and societal convention. (Mackie, 116)
It is this explanation which I find most compelling. According to this explanation, moral attitudes are both socially established and socially necessary. Morality is necessary in society to enable moral discourse and regulate interpersonal relations. To these ends, objective morality makes sense. When a speaker says should or ought, he is expressing demands which he makes, as a member of a society, and which he has developed by participation in that society. We want to be able to make such demands upon others. One way for this to be the case is for us to accept the same morals. We may accept the same morals because they are objective moral truths, but we may also have shared morals because we have compromised or otherwise agreed upon a certain set of moral claims as accepted.
This description fits with my suspicion that objective moral facts do not exist. It also fits with my belief that this need not present a problem with morality. We may even have good reason to appeal to moral facts to explain our shared morals even though such facts do not exist. It is not odd that we rely on non-existent moral facts; it is a natural consequence of our desire to maintain society. We accept the terminology of moral facts so that we can have meaningful and normative moral discourse.
So it may be the case that our system of morality can survive intact without the existence of moral facts. With this reassurance, many of the arguments of the moral realists fail. But the claim that we may not need moral facts does not by itself give us a particularly good reason to believe moral facts do not exist.
Mackie does offer several arguments against the existence of moral facts. I have mentioned three of these above: the argument from relativity, the argument from internalism, and the argument from queerness. I will here briefly discuss the arguments from internalism and from queerness before directing my attention in the next section to what I believe is the most successful argument against the existence of moral facts: the argument from relativity.
If moral facts exist, Mackie says, they are objectively prescriptive. Mackie maintains that the recognition of moral facts either necessarily motivates or provides reason for particular actions. That is, according to Mackie, moral realism requires internalism. Mackie is then able to argue that if there is no internalism, as may be the case, moral facts cannot exist (Brink, 113-114). I will not consider this argument in further detail, but mention it only as an example of one of several arguments.
Mackie's argument from queerness is really composed of two separate parts, one which is metaphysical and one which is epistemological. The metaphysical argument from queerness goes as follows: If moral facts exist, then they must be very strange, or queer, things. This argument raises many questions. What is a moral fact? Does it depend on other properties? Is it an entity of its own kind? The epistemological argument from queerness, similar to the metaphysical argument says: If we are aware of moral facts, then we must come to know them through some strange, or queer, faculty. Again, what is this faculty? (Mackie, 111)
I will not attempt to consider these questions here, but I will point out that I do not think this is a very strong argument. That something would be a queer kind of entity, I believe, does not imply that it doesn't exist. Nor does the fact that we don't know what kind of thing something is imply that it doesn't exist without our knowledge of its existence. However, I believe that both arguments from queerness do something important. They shift the burden of proof from moral skeptics back onto the moral realists. It is the job of the moral realist, as a defender of the existence of moral facts, to explain what kinds of things these moral facts actually are.
In the next section I will consider what I think is the strongest argument Mackie offers against moral realism: the argument from moral relativity and moral disagreement. In order to do so, I will consider Mackie's essay and an additional essay by David Brink.
Moral Relativity and Disagreement:
Finally, I turn to Mackie's argument from relativity which goes as follows: There exist and are observed well-known variations in moral codes from one society to another. How can this be if there are universal, objective facts about morality? The explanation for this variance does not seem to be that these different societies are having trouble arriving at objective moral facts. Instead, this variance seems to be explained by the hypothesis that these different sets of values represent the different ways of life of these societies. It is not that the morals of one society are superior to another, or that any one society has arrived at the true morality, but that the morals of each society are reflections of different cultures. (Mackie, 109-110)
I believe that Mackie's argument from relativity expresses precisely the kind of objection that I have against moral realism. Perhaps there are moral facts. But if there are, it seems that they are unknown to us. How else can we explain the fact that different societies really do seem to have different moral values? How else can we explain the fact that even within a particular society, different people seem to have different moral values? Again, it could be that there are moral facts to which we do not have epistemic access, but it seems far more likely to me that there really are no moral facts. This is not really a strong enough argument to rule out the possibility of the existence of moral facts. However, I believe it is strong enough to merit skepticism about the existence of moral facts. This argument raises a strong objection which the moral realist must address if the thesis of moral realism is to be taken seriously.
A similar, and I think slightly stronger, argument which often goes by the same name as the above-described argument from relativity is really an argument from moral disagreement. This argument relies on the observation of moral disagreements between and even within societies and on the question this presents to moral realism: How can genuine moral disagreement arise if there are moral facts? David Brink devotes much energy to clarifying this argument in his ``Moral Realism and the Sceptical Argument''. He states Mackie's argument as follows: There is observed inter-societal and intra-societal disagreement in ethics. The best explanation of such disagreement is that moral facts do not exist. Therefore, moral facts do not exist. (Brink, 115)
Brink is quick to point out that this is an oversimplification, and is not what Mackie, being only a moral skeptic and not a skeptic more generally, has in mind. Just as there is disagreement about facts in ethics, so there is disagreement about some facts in science. But Mackie does not want to say that if there is disagreement about some, or even all, facts in science, then scientific facts don't exist. Scientists may disagree about the nature of black holes, but this does not mean that there is no fact of the matter regarding black holes (Brink, 115). Just because scientists do not know the answers to some questions, and therefore do not come to agreement, does not mean that the correct, true answers do not exist.
What Mackie needs, explains Brink, is for ethics to be different from science in such a way that the argument from disagreement can be applied to ethics, but not to science. And Mackie maintains that ethics and science are different in just this way. Mackie maintains that ethical disagreement is different from scientific disagreement because while scientific disputes may not be resolved, they seem resolvable. In ethics though, many disputes seem not just unresolved, but unresolvable. That is, science and ethics differ in that although there are disagreements in both science and ethics, it is only in ethics that the disagreements are unresolvable. (Brink, 115)
Moral realists, such as Brink, can reply to this charge in any of a number of ways (Brink, 117). First, a moral realist may respond that the descriptive claim that there is moral disagreement is just plain false. Any observed moral disagreement is not actual, but only apparent. Second, a moral realist may respond that when a moral disagreement is perceived, there is a genuine disagreement, but it is not a disagreement within ethics. Instead, the disagreement arises from either disputes about how to apply the objective, agreed-upon moral facts or from misunderstandings or differences pertaining to relevant non-moral facts and views.
Third, a moral realist may concede that there are genuine and unresolved moral disputes, but these disputes are, in fact resolvable. These disputes are resolvable in principle; it is only that they have not been resolved yet. And finally, a moral realist may even concede that there are genuine unresolvable moral disputes without believing that the existence of such disagreement implies the non-existence of moral facts. Moral facts may exist, but may be both unknown and unknowable to us.
Any or all of these replies may undermine Mackie's objection from moral disagreement. Nonetheless, I feel that Mackie's argument is a strong one. The moral realist has many strategies for responding to Mackie, but it is up to him to develop an argument to do so in order to defend the existence of moral facts.
In summary, I see Mackie's ``The Subjectivity of Values'' as doing two things: First, Mackie acknowledges that moral judgments do include a claim to objectivity, but this, he does not believe, in any way supports moral realism. Mackie develops an explanation which supports his thesis that there can be, and is, objectivity in ethics without the existence of moral facts. Second, Mackie offers several objections to moral realism, including arguments from moral disagreement and the appearance of different moral values in different societies.
As I have stated at the outset of this paper, I do not believe there are moral facts. This should not be confused with the stronger claim of belief in the non-existence of moral facts. My point is simply that I believe the objections raised by moral skeptics provide significant grounds for doubting, not disproving, the existence of moral facts.
I have considered in detail two arguments against the existence of moral facts, namely the argument from explanatory irrelevance/impotence and the argument from moral relativity/disagreement. I have tried to be as critical as I can of these arguments in an attempt to justify the moral realist's position.
The first of these arguments, explicated by Harman, is based on two premises: (1) that moral facts do not, or need not, play a role in the explanation of our moral observations, and (2) that belief in the existence of moral facts is justified if and only if it plays an explanatory role. I have suggested that this strong form of the explanatory criterion is suspect and that further attacks on this argument are possible by demonstrating that the existence of moral facts does play such an explanatory role.
I think the second argument, attributed to Mackie, is somewhat stronger. The argument is that the existence of different cultural moral values and/or apparent moral disagreement seems inconsistent with the hypothesis that moral facts exist. Again, I have tried to argue against this objection. I have outlined several arguments based on the premise that, for various reasons, there is no genuine disagreement in morality. I have also expressed my view that this argument is only one about whether we know or can know the content of moral facts, and it in no way constitutes an attack on the existence of moral facts themselves.
Having pointed out these and other flaws in these two arguments against moral facts, I am convinced that neither provides adequate ground from which to conclude that moral facts do not exist. However, I do not believe this means that the arguments of Harman and Mackie have failed. I believe their arguments provide adequate reason to seriously question the existence of moral facts. These arguments shift the burden of proof to where I think it belongs. That is, the default position is not to accept, but to question the existence of moral facts. Moral realists must offer convincing arguments of moral facts, which are free from, or can satisfy, the objections of moral skeptics. As I do not believe moral realists have done this, I remain unconvinced of the existence of moral facts.
Finally, though, I am unconvinced that a system of ethics without moral facts, as commonly understood, is really a problem for morality. While I believe that there are many reasons we want moral facts, I believe these desires can be filled without the existence of moral facts. I believe Mackie is right that our moral judgments do include claims to objectivity, and I believe these claims are possible without moral facts. I believe that objectivity can, and does, arise in society without moral facts. I conclude that it is the language of objectivity or the concept of moral facts, not moral facts themselves, upon which we rely. Skepticism, then, is not something to be worried about; it is an appropriate response to moral realism, or to any theory which has not been adequately substantiated.