[This is an undergraduate paper from circa 1992. –mh]
1. What is the issue
The present essay is a defense of a view called moral objectivism and attack on its opposite, subjectivism or moral relativism. Moral relativism is probably the subject concerning which more nonsense has been written and said in modern times than any other in moral philosophy. I suspect this is partly because people wish to provide arguments in favor of relativism without first having a clear idea of what their thesis is; partly because the authors’ arguments are mostly rationalizations; and partly because the authors have a poor grasp of moral concepts. There is little I can do about the second and third problems, but I will try to help the first here. In this section I define “objectivism” and important related terms and delineate several views that might be called subjectivism, which I contend are all demonstrably false.
I am not chiefly concerned herein to defend any particular moral claims, although I shall mention some uncontroversial moral truths for illustrative purposes. Rather, my concern is to show that questions of value have objective, rational answers but not to provide those answers. The latter is a task for another time.
1.1. “Objectivism” and “relativism”
“Objectivism” denotes the thesis that morality is objective. Subjectivism holds that morality is subjective. Relativism holds that morality is relative. In the sequel, I am interested in distinguishing moral objectivism from its denial; therefore, I assume that “relative” and “subjective” both mean “non-objective”. If they do not already mean this, then I stipulate that meaning hereby. There are a number of people who believe moral relativism so defined.
1.2. What is ‘morality’
I want to make two points about what morality is as I understand it.
First, the term “morality” is subject to the same ambiguity as most other names for fields of study, which we might call the subjective/objective ambiguity. By this I don’t mean to imply that using the word one way commits one to objectivism and using it the other way implies subjectivism or anything like that; there simply are two different legitimate definitions of “morality”. On the ‘objective’ interpretation, “morality” refers to such situations as something’s being right, evil, just, or the like. On the ‘subjective’ interpretation, “morality” refers to theories about or the study of rightness, evil, justice, and the like. For example, “People must not use violence against one another” is a claim about morality in the objective sense – that is, it is a value judgement. It seeks to say what is right, wrong, or the like. On the other hand, “In Xanadu, the use of violence is strongly condemned” is not a value judgement; it can be verified or refuted purely by anthropological observation. It is a statement about morality in the subjective sense. It seeks to say what people consider right, wrong, or the like. I am not going to discuss which of these two interpretations is ‘better’. I simply point out the distinction. An analogous distinction applies to many other words, such as “chemistry”, “psychology”, “zoology”, “mathematics”, etc. If there were no people, would there still be chemistry? Well, chemistry in the objective sense would exist, but chemistry in the subjective sense would not: i.e., there would still presumably be chemicals with certain properties behaving in certain ways; but there would be no study of chemistry and no theories thereabout.
Second, in this paper it will be convenient for me to use “morality” in a very broad sense. I shall call “morality” (in the objective sense) all facts, if there are any such facts, about what is wrong, good, bad, evil, ill-advised, just, beautiful, or preferable, or any other evaluative property. Anything that is a value judgement will count as part of a morality in the subjective sense. For instance, the fact that Aristotle is a great thinker is a moral fact in the broad sense, because it requires a value judgement to appreciate; so is the fact that it’s best to eat when one is hungry, because stating it gives a prescription for action; so is the fact that the world would be better off without tyrants, because it requires a value judgement to observe (calling something “better” as well as calling someone “a tyrant” are value judgements). In particular, I stress that I do not wish to presuppose any particular theory about how people should behave nor any particular reasons why they should so behave. Most people appear to restrict the application of the term “morality” to prohibitions on actions satisfying desires. I disregard this convention. If desires must be held in check, then that will be a moral fact; and equally, if desires need not be checked but provide appropriate and rational reasons for acting (I don’t mean merely that they make one want to act, which is a purely descriptive fact and not an evaluation, but that acting in accord with them is a good thing) then that will be a moral fact. In other words, my defense of objectivism, while it says that there is at least sometimes a way one should behave, does not actually recommend anything in particular.
1.3. “Values are subjective” = “All values are subjective”
We want to know whether there are objective values (which I take for the same question as whether morality is objective). It may be asked, what shall we say if it turns out that some values are objective and some are not? The answer I give, by stipulation, is that in that case objectivism is true and subjectivism is false; that is, I interpret “morality is objective” as “some values are objective”. I might have made the opposite stipulation – viz. that “morality is objective” = “all values are objective” – but that would be less interesting since, at least on the most obvious interpretation, this would make objectivism into a doctrine that no one holds. For instance, I don’t think the value ‘the right to punish slaves for disobedience’ is objective because I don’t think there is any such right. Similarly, any number of values could be enumerated that any given person would declare to be utterly non-existent and thus not objective.
1.4. Three ways of being non-‘objective’
Suppose I offer the opinion, “Colors are objective.” What then is it that I am saying about colors? What I am saying, I think, is that colors are ‘in the object.’ In what object? In colored objects. What does “in” mean here? It means that a color – redness, say – is a property of the objects that are said to be red, that is, that the nature of those objects themselves and not anything else determines whether they are red or not. Hence, to say that morality is objective is to say that whether an action is right depends on the nature of that action; whether a person is good depends on the nature of that person; etc. Well, that just sounds trivial. How could anything not be objective?
So far as I can see, there are three and only three ways for some thing, x, to fail to be objective, for instance for values or colors to not be objective:
1. If everything is non-x; e.g., nothing has value or nothing is red. Goodness is not in the object if there isn’t anything good. Redness is not in the object if everything colored is some color other than red. For example, if someone asked whether witchhood is objective, I might answer no, because nothing is a witch.
2. If some things are x, but whether a thing is x depends not just on that thing’s intrinsic nature but on facts about the subject, i.e., the person who says or observes that the thing is x, as well. Redness is not objective if whether a thing is red ‘for some observer’ (if that makes sense) depends on the nature of the observer and not just on the nature of the object. Notice that if and only if a quality is relative does it make sense to append “for some observer” in sentences ascribing that quality; and in that case such sentences do not make sense without the addition. For example, word meanings are not objective; they are relative. A word must mean something only ‘for some speaker or listener’ and what it means depends on facts about that speaker/listener – roughly, what he has in mind when he utters/hears the word. It does not matter what properties the sequence of phonemes has, beyond pronouncibility.
3. If it is neither true nor false that something is x. Someone who thinks values are subjective in this sense would say that value judgements can be neither true nor false. Well, that sounds almost incoherent: how is it possible for a statement to be neither true nor false? Doesn’t that violate basic logic? So far as I can see there are just two ways this is possible. First, if saying something is x is not a genuine assertion, then it is neither true nor false. I can’t think of any examples of an x for which this is true, but there are numerous utterances that do not assert anything, such as, “Congratulations on your Nobel Prize” or “What time is it?” “Congratulations on your Nobel Prize” is neither true nor false. It just expresses a certain sentiment. Second, if an assertion involves a false presupposition, then it may be said to be neither true nor false. For instance, “The king of France is bald” is neither true nor false because it contains a false presupposition that there is a king of France. Some people at any rate have argued that.
Another way of stating the thesis that morality is objective is to say that values are ‘part of the fabric of reality;’ that is, there is some actual state of the world that corresponds to a value judgement. Again, that sounds trivial; how could any statement fail to correspond to some state of the world? There are the same three ways in which this could happen: if the statement is false; if it is true, but it corresponds to some state of the subject who observes it and not to the (external) world; or if it is neither true nor false.
1.5. Several relativist theories
Here are a few different things one could believe in order to be a moral relativist:
1. Moral judgements are simply universally in error; i.e., contrary to appearances, nothing is good, right, evil, just, etc. These are concepts without any application.
2. Moral ‘judgements’ are not genuine assertions. They don’t actually claim anything about the world. Instead, they are mere expressions of emotion, as “Hurray” is an expression of emotion.
3. “x is good” means “I like x.”
4. “x is good” means “x is ordained by my society.”
5. What people do when they make a moral judgement is to project their subjective mental state out into the world. They confuse their emotions with some object in the world and mistakenly take the feeling in them to be some property of the object. This is the most psychologically sophisticated version of relativism. An analogous theory might be held about colors: that when people see one of the objects we call “red,” we have a certain characteristic sensation, which sensation we confuse with some property of the object that causes it and call the property of being red.
6. Morals (in the objective sense) are established by convention; i.e., in the same sense in which a society may establish a convention such that certain kinds of pieces of paper are money, or establish conventions such that certain activities constitute marriage, and so on, just so, a society may establish conventions such that certain things are good. Things become good or bad in virtue of conventions.
1.6. What the issue is not
Some people argue about whether morality or anything else can be ‘absolute.’ “Absolute” might mean “certain”, it might mean “exceptionless”, it might mean “objective”, it might mean “universal” in some sense, or it might mean something else. I don’t know what it means in the context “There is an absolute morality;” therefore, I will not use the term. I am not interested here in whether morality is ‘absolute’ in any of the other senses than “objective”.
I am not concerned with whether there are some exceptionless rules for judging moral issues – whether there is an algorithm for computing morality. My own opinion happens to be that there is not, but that has nothing to do with the present issue.
I am also not arguing that there is a universal morality in the sense of a moral code that everybody either does or would accept.
I am not arguing that we can know moral truths with absolute precision or certainty.
I am not considering the issue of whether one should be tolerant of people with differing practices or differing views. That one should be tolerant or that one should be intolerant are particular moral conclusions that are each equally consistent with objectivism.
I am not interested in the question of whether at any given juncture there may perhaps be several distinct, equally right actions available rather than only one.
None of those things is the issue. The issue is only, as I have said, whether moral properties are in the object. Perhaps I shall take up the other issues in other essays, but not now.
By clarifying the theses of objectivism and subjectivism, I may have just drastically reduced the number of opponents I have, for many readers may have simply dropped out of the relativist camp by reason of hearing what exactly relativism is. Indeed, I suspect confusion with other issues may be relativism’s strongest means of gaining support. Nonetheless, I have no doubt there is still a substantial number of people who endorse relativism as I have defined it.
2. The consequences of relativism
The concern of this section is slightly off topic. Here I will argue that, unsurprisingly, moral relativism undermines morality and leads to nihilism because it has the consequence that any morality in the subjective sense is both arbitrary and irrational. I say this is off topic because this particular thesis does not show that moral relativism is true or that it is false; however, the issue seems important enough and enough subjectivists fail to understand it, leading them to hold inconsistent positions, for it to be worth addressing.
The argument is extremely simple. Since rational judgement presupposes some ground apart from the judgement on which for it to be based, the denial of objectivism implies the intrinsic impossibility of rational moral judgement, since said denial means that moral values cannot have any independent existence apart from the mind. To put it another way, in order for a judgement to be rational, one must make the judgement because it is true or at least because evidence indicates it is true. That means that the thing must already be true, or already be supported by the evidence. Since according to subjectivism, quite to the contrary, evaluative propositions can never be true prior to being judged correct since moral values can not exist independent of such judgements, it follows that it is impossible to make a rational moral judgement: i.e., rationality requires that a judgement be validated before it is accepted, but relativism implies that it must be accepted before it is valid if it can ever be valid at all (one version of relativism saying that such judgements can not ever be valid – but every version implies that they can not be valid prior to their being accepted), so relativism implies rational moral judgement is impossible.
Relativism makes moral judgement not merely non-rational but positively irrational, insofar as it implies that moral judgement must always proceed according to a manner which is directly contrary to what reason demands – must always occur without basis, that is.
Every action and every moral judgement is, if subjectivism is true, arbitrary – that is, groundless – because any ground for some thing must by definition be prior to that thing and, since (a) the notion of a ground or reason is normative (it implies ‘justification’) and further (b) in this case the ground in question has to be the truth of the proposition judged, relativism states that no such things as grounds can exist prior to the making of moral judgements. One version of relativism (see above, section 1.4) implies that whatever moral values we adopt are ipso facto right, but that means that a decision about which values to adopt must be arbitrary since anything we picked would be right. Another version implies that whatever values we adopt are wrong since value judgements are always false, which means that we can have no valid values, which means that every decision must be arbitrary.
What this shows is that if one knows moral relativism to be true, then one cannot rationally believe any moral judgement. One cannot do so because in order to rationally believe something, the proposition must first be justified, and as a moral relativist you know that no moral proposition is true before you believe it, so you would not have any justification for accepting it.
I should note that several influential relativists would presumably deny my analysis. J.L. Mackie, for one, claims that his view that moral values are not “part of the fabric of the world” is consistent with any moral views – i.e., he can still make ordinary moral judgements that this or that is good and so on.(1) Subjectivist philosophers, including Mackie, standardly draw a distinction between first- and second-order moral views and hope by this to show that they can maintain their ‘second-order’ view without giving up any of their first-order moral views. A ‘first-order’ moral view is a claim about what is good or bad, right or wrong; while a second-order moral view is about the nature of first-order moral views (e.g., what it is for something to be good or bad or right or wrong). The argument, presumably, is that since first- and second-order views are about different things, a second-order view cannot be in conflict with a first-order one, so we won’t have to reject any first-order moral views as a result of accepting moral relativism. Relativism is, as it is said, not an ethical theory but a meta-ethical theory.
Frankly, I find that argument preposterous. If your ‘meta-theory’ consists in the denial of the existence of any subject matter for your theory, how can you continue to have a theory? Suppose that it were claimed that chemicals have no objective existence: can anybody imagine that the adoption of this belief would have no effect on the science of chemistry? Obviously, the entire science would be undermined. By analogy, if someone says that values have no objective existence, moral philosophy is undermined since it has no subject matter. It is then comparable to the study of unicorns. Nothing positive you say about unicorns can be true since there aren’t any unicorns. And it makes no sense to say, “Well, I agree that unicorns are not real, but I still think this is a unicorn.” How is it any different to say, “Well, I agree that values aren’t real, but I still think this is a value”?
3. Arguments for subjectivism
Here I will attack the four main arguments for subjectivism that I know of, each of which is a very bad argument.
3.1. Cultural variance of moral codes
First, it is pointed out that there is wide variation in moral codes from one society to another and from one time period to another. Even people in the same place and time, as in our society, seem to have great difficulty in agreeing on moral issues. Moreover, there does not seem to be any decisive way of resolving disagreements. The best explanation for this situation, says the relativist, is that there are no facts there to determine or to agree to to begin with.
I think the level of disagreement is exaggerated. I think it would be widely agreed that courage, honesty, and kindness are virtues; that life and happiness are good; &c. The argument basically takes the most extreme and atypical examples to make its point.
But more importantly, one glance would show the absurdity of the logical extension of this argument. Disagreements in questions of history or biology or cosmology do not show that there are no facts about these subjects. In fact, the situation described above, wherein people disagree widely and there doesn’t seem to be any way of resolving their disputes, is characteristic of all of philosophy. And it is characteristic of every field that is important to people – religion, history, law, politics, metaphysics, ethics, cosmology, etc. People seem incapable of agreeing on whether God exists or not, and there are no arguments or observations that can resolve the dispute to everybody’s satisfaction. Does this show that there is no objective fact of whether He exists?
Why is it that people argue interminably about religion but not about mathematics? It is not because numbers are objective and the existence of God and similar issues are subjective. It is, mainly, because ordinary people do not care about the properties of numbers. But they do care immensely about God, life after death, and the like. And they care at least as much about morality and values. And when people care very much about something, and have a vested interest in the answer, they are likely to develop strong, dogmatic opinions and to allow their emotions to prejudice their judgement.
Additionally, as Aristotle pointed out long ago in a remark that richly deserves to be listened to but has not been,
[I]t is the mark of an educated man to look for precision in each discipline just so far as the nature of the subject admits; it is evidently equally foolish to accept probable reasoning from a mathematician and to demand from a rhetorician scientific proofs.(2)
Seemingly contrary to popular opinion, there are plenty of perfectly legitimate fields of study that are not exact sciences.
Second, it has been argued from time to time that moral relativism presents a simpler picture of the universe than objectivism. Objectivism postulates these entities, objective moral values, that we could explain the world just as easily if not more easily without. Therefore, the burden is on the objectivist to prove the existence of these things.
I think this argument is insincere; that is, nobody ever became a relativist because of this. It was invented after the fact to confuse objectivists.
The argument is exactly analogous to the following argument for mathematical relativism: Objectivism postulates these entities, objective numbers and numerical relationships, that we could explain the world just as easily if not more easily without. Therefore, the burden is on the objectivist to prove the existence of these things. Since he cannot do so, I conclude that all mathematical statements are arbitrary and subjective.
The flaw is that saying that ethical (or mathematical) statements are objectively true or false does not imply that there exist some supernatural, ethereal substances that are values (or numbers). It just implies that some things have quantities (for mathematics) or some things are good or bad (for ethics). I am not postulating the existence of any new substances. I am only judging that some things are good, and goodness is a quality, not a substance or object. Since, presumably, if objectivism is true then it is necessarily true, and since it is a conceptual and not an empirical issue, the question of simplicity or ontological economy does not arise.
If anything, we should say that the burden of proof is on the moral relativist, for advancing a claim contrary to common sense.
3.3. Where does moral knowledge come from?
Third, the relativist asks, by what faculty does one come to know about moral truths? If one cannot explain how one knows about something, then it is not plausible for one to make claims about it. Is there some special faculty comparable to perception?
The answer is that one figures out prescriptions on the basis of descriptive facts. For instance, one finds out that something is good by rationally drawing this conclusion on the basis of its other properties. It is an old platitude in moral philosophy that you cannot derive an ‘ought’ from an ‘is’, so it is supposed that what I have just enunciated is impossible. Well, in one sense, you cannot derive an ought from an is – in the sense that the prescription will not follow analytically, or just in virtue of the definitions of terms. But in another sense, you can derive an ought from an is – i.e., it will follow necessarily and a priori. Many, if not most, philosophers seem to find this kind of cognition incomprehensible, probably because of a confusion of the notions of necessity, analyticity, and a prioricity. However, examination of just about any mathematical proposition would reveal this mode of cognition – you cannot derive most theorems solely on the basis of definitions. You must also have some intuitive judgements, usually made explicit in the form of axioms. Likewise, you cannot derive substantive moral judgements solely on the basis of definitions (G.E. Moore showed that in his discussion of the naturalistic fallacy), but you can make intuitive judgements.
Moral intuition is not comparable to a special faculty of perception, because moral judgements are supposed to be necessary (given the other, descriptive facts) and not empirical. So moral intuition is just the general faculty of reason applied to a particular subject matter, viz., values, just as mathematical intuition is not a separate quasi-perceptual faculty but rather the faculty of reason applied to numbers.
That we in fact derive moral judgements from descriptive judgements all the time can be exemplified by just about any argument about moral or political subjects. For instance, it may be argued that communism is a bad system of government on the basis that it has caused tens of millions of deaths, that it impoverishes the country in which it is adopted, and that it greatly restricts people’s freedom. I think that is a good argument. It certainly is not some kind of simple logical fallacy, as the concept of ‘the naturalistic fallacy’ would presumably imply, since I am deriving a moral judgement from other, non-moral judgements.
It might be, and standardly is, replied that the argument presupposes certain implicit moral judgements, that life, prosperity, and freedom are good. I think there is something wrong with this, but it would take us too far afield to consider. Suffice it to say that if that is the case, then these suppressed premises that life, prosperity, and freedom are good are simply immediate intuitions. There is no difficulty in this proposal, since there are numerous examples outside ethics of synthetic, a priori judgements apprehended by intuition.
3.4. The political argument
Perhaps the main motivation for relativism among contemporary intellectuals is the appeal to the virtue of tolerance. In essence, the argument is this: objectivism leads to intolerance because it makes us think that we are right and other people who disagree with us are wrong. This causes conflict, chauvinism, and subjugation of some people by others, which is bad. The only way to ensure a desirable attitude of toleration on our part is to posit relativism as a moral postulate, which will reconcile us to the equal legitimacy (or illegitimacy) of all value systems and thereby enable people with different values to live in harmony, provided they accept the postulate.
The first obvious reply to this political argument is that it is a non sequitur – that is, even if true, all it shows is that it would be advantageous to somehow convince people to believe relativism; but it does not show that relativism is actually true.
Second, since this kind of argument would only move people who believe in the value of toleration anyway, it would seem at least as reasonable to simply postulate tolerance as an objective value, as to postulate general subjectivism, if we are interested in promoting tolerance.
Third, there are both theoretical and empirical grounds for believing that the opposite relation between objectivism and toleration from the one urged would exist – that is to say, it is objectivism that leads to toleration and subjectivism that leads to intolerance – for my view encourages an objective and rational attitude towards public policy and other moral questions (Cf. above, section 3.3), whereas subjectivism naturally tends towards an unreasoned and arbitrary approach (Cf. section 2), and it certainly seems that reason would counsel us to avoid destructive conflicts and respect the rights of others, whereas, for example, a purely emotional value system might lead, as it usually has in the past, to fanaticism, xenophobia, etc. If only we could get warring peoples around the world to listen to reason, one is inclined to hope, perhaps they could be convinced to resolve their disputes through negotiation rather than violence – but not if they are convinced that rational argumentation about whatever issues they disagree about is inherently futile.
The connection I suggest is supported by examples: John Locke’s political theories, which have probably led more than any others to democracy and respect for universal human rights, are a good example of the kind of conclusions that a serious attempt to identify objective moral values usually leads to. In contrast, the ideologies associated with the two major forms of tyranny of the twentieth century – namely, communism and fascism – have hardly exemplified objectivism. Orthodox Marxism holds that moral values are not objective but are mere fictions invented by the ruling class to further its class interests (much like religion). The German Nazis held that all values are determined by one’s race, that the right was just what accorded with the will of the people, and that moral values thus had no objectivity. It scarcely need be pointed out that the subjectivism that these ideologies embraced did not induce toleration on the part of their followers. Instead, it carried the implication that since reason was inapplicable to moral questions, conflicts of values could not be resolved except by the conflicting groups fighting it out.
4. Several versions of relativism refuted
In section 1.4 I delineated three ways in which relativism might be true, and in 1.5 I listed six versions of relativism (each presumably satisfying at least one of those three ways) (see above). In this section I try to drive relativism into a dilemma or series of dilemmas. I will show in turn that each of the possible versions of relativism is false, for different reasons. First, then, consider the six versions listed ….
4.1. Value judgements as universally false
This theory is really quite outrageous. It implies, among other things, that it is not the case that people generally ought to eat when hungry; that Hitler was not a bad person; that happiness is not good; and so on. I submit that this is simply absurd. I feel much more confidence in those denied judgements, as I think nearly everybody does, than I can imagine feeling in any philosophical arguments for relativism. At least, I think it would take an extremely strong argument to shake my confidence that happiness is preferable to misery, or the like. And there does not seem to be any argument at all with that import. It is hard to see how there could be.
This discussion makes me feel like G.E. Moore, who refuted skepticism about the existence of external objects by making a certain gesture and observing, “Here is one hand,” and, making another gesture, “and here is another.” For just as Moore pointed out that no premises of any philosophical argument could possibly strike him as more obvious and certain than the proposition that “Here is a hand,” I find it inconceivable how any philosophical premises could be more obvious and certain than the judgement that happiness is desirable, or numerous other similar value judgements I might make.
I doubt that anybody actually holds this view. Even arch-subjectivist David Hume remarked that “those who have denied the reality of moral distinctions, may be ranked among the disingenuous disputants.”(3)(4)
4.2. Moral judgements as expressions of sentiment
Sometimes Hume talks as if he thought moral statements were expressions not of judgements but of emotions. On this view, “x is good” is comparable to “Congratulations,” “Hurray,” “Ouch,” and other non-assertive utterances. That this is false can be seen readily from four considerations.
Call the first the appeal to introspection. The making of a normative judgement is experienced as just that – making a judgement: i.e., as a matter of good phenomenology, when one considers a moral issue, it seems clear, one is engaged in that mental process known as judgement; one is not primarily engaged in imagination or memory or perception or feeling (though those may accompany the process of judgement, of course). And I think that everybody can see this if they think about it – that is why moral judgements are, after all, called “judgements”.
If someone reports that when he introspects he does not ever detect a process of judgement going on where morals or practical matters (meaning matters of what to do) are concerned, then in the first place, I won’t believe it, and in the second place, if I did then I would conclude that the unfortunate fellow is simply unable to grasp moral concepts and is therefore unable to think about them – I would conclude that he is moved by emotions and instincts rather than reason and morality. Moral concepts and arguments are as a rule highly abstract, and the existence of such people as cannot understand them is certainly not inconceivable. Animals are most likely all in that position.
One point of distinction between judgement and feeling is of activity versus passivity – that is to say, judging is something one does, whereas having a feeling is something that happens to one. Therefore, I am saying that deciding, e.g., what is right, is something that one does (as deciding always is) and not something that one just undergoes. I think the merest introspective reflection will bear me out on this.
Second, moral judgements can properly be called “true” or “false”. If somebody says something that is not an assertion – such as “Ouch!”, then you cannot ‘disagree’ – that makes no sense. But if someone says “We should do such-and-such,” you can disagree. You can call someone’s value judgements true or false in the way you cannot call “ouch!” true or false, which shows that there must be some proposition the value judgements express.
Third, it’s pretty obvious that, linguistically, prescriptions take the form of statements, and we all recognize them as such. They use the indicative mood, containing a subject and predicate, &c. And I don’t see any special reason for thinking that there is something deceptive about our language (and presumably virtually all others).
Fourth, normative judgements can stand in logical relations to other propositions. For instance, the statement, “I should return this book to the library” straightforwardly entails the admittedly objective statements
I can return this book to the library.
I borrowed this book from the library.
This book exists.
I have not returned this book to the library.
That these descriptive judgements follow from the normative judgement should be uncontroversial. Additionally, the statement, “I should return this book to the library” is correctly said to be in contradiction with the statement, “I should not return this book to the library.” But logical entailment and contradiction are relations between propositions. If moral judgements did not assert anything, then one certainly could not deduce anything from them or call them “contradictory” to anything. Therefore, moral judgements express propositional contents.
4.3. “x is good” as synonymous with “I like x”
This theory would have to be expanded to include (re-)definitions of all other evaluative terms as well, of course; however, we can refute this line of approach already.
It makes sense to say, “I like it, but is it really good?” but it does not make sense to say “I like it, but do I like it?” nor “It’s good, but is it really good?”
One often thinks that one likes something because it is good, but one does not think it is good because one likes it (unless one is very egocentric).
Therefore, some thing’s being good must be different from its being liked. More simply, though, this should be immediately evident, since the statement that any given person has any given psychological state is a descriptive statement, whereas the statement that some thing is good is, of course, normative. The former denotes an empirical matter of psychology. The latter expresses a value judgement. In short, this theory is a simple instance of the naturalistic fallacy. I would lump together with this view any view that identifies good, virtue, and other moral qualities with the tendency to cause some psychological state.
4.4. “x is good” as a synonym for “x is ordained by my society”
The same thing might be said about this theory: namely, to call something good is to express a value judgement, but to say something is ordained by society is to offer a descriptive judgement of anthropology which could be confirmed or refuted purely by observation. This is another case of the naturalistic fallacy.
It is possible to doubt whether what society ordains is good but it is unintelligible to doubt whether what is good is good or whether society ordains what it ordains. Therefore, ‘the good’ must be something different from ‘what society ordains.’
It is also common for society to ordain something because it is good. That makes perfect sense. But it does not make sense to say that society ordains something because it ordains it, or that something is good because it’s good.
4.5. Moral judgements as projection
To begin with, it strikes me that confusing one’s emotions with physical objects is an extremely childish error to be accusing virtually all humans, including some of the profoundest philosophers, of committing. Strangely, though, it is an error from which, seemingly, animals, small children, and dimwitted and uneducated people are exempt, inasmuch as, I believe it is commonly agreed upon, they fail to use concepts of morality, although they too experience emotion. If this be errancy, it is a form of errancy that is most curiously correlated with intelligence and education.
Second, this kind of theory could be proposed for any quality. That is, for any property that we seem to sense in objects in the world, it could always be asserted that we are projecting our subjective mental state out into the world, and it would be difficult or impossible to refute the assertion. It could, for example, be claimed that colors don’t really exist and we merely confuse our subjective sensations with external objects. Arguably, Bishop Berkeley proposed this theory for all physical objects. But it seems to me that if someone is going to propose a theory in this general vein, which implies that people are constantly falling prey to a simple error, then the burden is upon him to produce some powerful evidences to prove his theory.
Third, the theory is highly implausible purely on phenomenological grounds. There isn’t anything like a single feeling I have when I contemplate each of the things I consider to be good, as the theory would appear to predict. I think Newton’s work on the calculus is extremely good, but I don’t feel emotional about it at all. It appears to me that I make evaluations on intellectual grounds. Of course, it is possible to make them on emotional grounds, but then it is possible to believe in God, in the afterlife, and any number of other things on emotional grounds, without that rendering the issues thus treated intrinsically subjective. Moral evaluations are subject to rational argument. And if someone asks me why some course of action ought to be taken and I report that I have a certain feeling, I think everybody, myself included, will find my answer quite inadequate and irrelevant.
Fourth, if this theory is true, then why doesn’t everybody wind up with a moral code that says he may do whatever he feels like and other people may only do things that he likes – or rather, at least, one that picks out the same things as being good as happen to be liked by that individual? The fact is, we don’t have theories like that. Quite to the contrary – a great many people, one might say even a vast majority of people, have moral codes that frequently pick out as wrong things that they would otherwise enjoy (fornication is the most obvious example of such a thing).
Fifth, it is usual for a person to have a positive sentiment towards something because he believes it to be right or to have a negative sentiment because he thinks it is wrong. That is the way we normally seem to experience the connection between evaluations and emotions. That is why a psychologist would attempt to eliminate a patient’s guilt by means convincing him that he is not a bad person, and not the other way around. The theory in question reverses the causal direction.
Finally, the acceptance of this theory would presumably cause us to lose the inclination to moralize, for once we see the truth of it, we would see that all moral statements are intrinsically confused and, therefore, false or unintelligible. But I have said above (section 4.1) that the denial of all moral judgements is absurd and that I do not see how any philosophical premises that could be used to justify the theory in question could be more evident than certain value judgements (indeed, more probable than the disjunction of all possible value judgements). The claim must be either that when we judge something good we are attributing our emotions to it and therefore attributing consciousness to it, which is patently false – I said that Newton’s work on the calculus is good, but I certainly don’t think it has any feelings – or that somehow there is no intelligible thing that we are attributing. I just don’t believe the latter. I think that the concept of a thing’s being good makes perfect sense.
4.6. Morals as a matter of convention
It is crucial to note here that the theory I am considering under this heading says that morals in the objective sense are a matter of convention. I think it is perfectly possible for morals in the subjective sense to be established by convention. But it makes no sense to speak of establishing morals in the objective sense by convention. A couple of hypothetical questions should demonstrate this.
The existence of money and what counts as currency are literally established by convention. Imagine a situation in which the United States government changes our currency. It begins to print money with new kinds of pictures on it to replace the old money. A law is passed saying that the old money is no longer legal tender, and the citizens go along with it. We all start using the new money and nobody uses the old ex-money anymore. Now in that situation, would these green pieces of paper I have in my wallet with pictures of dead presidents still have monetary value? The answer is no. They would literally cease to be money in virtue of the conventions we established.
If right and wrong were established by convention, then we should be able to say something similar about them. We should be able to imagine a situation in which our society establishes different conventions and, in virtue of that fact, things that are presently right cease to be right and things that are presently wrong cease to be wrong. What would that be like? Suppose Americans were to decide that the communists were right after all and start electing socialists to government offices. Suppose that we adopt new laws and change the Constitution. The government turns socialist and, of course, becomes repressive, executes dissenters, and starts to drive us into poverty. But most everybody goes along with it. Suppose that there is a general consensus on the desirability of the new form of government. Now what I want to ask about this situation is, would communism be a good form of government, or would it still be bad? Surely this would be a case of establishing conventions according to which communism is good, if there were any such thing? (I could have imagined society turning genocidal or Nazi, etc. The point would be the same.) Yet here communism would still be just as bad as it always was. The fact that something is generally practiced, obviously, does not make it right; that is why it always makes sense to doubt whether current practices are right. It always makes sense to try to establish better conventions, to find conventions good or bad, and so on, which it could not make sense if there were no possible standard of value independent of the conventions themselves.
Note the contrast: because what counts as money is a matter of convention, a change of how we behave will make things that are presently money cease to be such; but a change of how we behave will not make what is wrong cease to be wrong. Therefore, what is wrong cannot be established by convention.
This theory, furthermore, is simply another instance of the naturalistic fallacy. That something is good is a value judgement, but that any given society performs any given set of practices and has any given set of conventions is a purely descriptive (and empirical, anthropological) judgement. They can’t be the same.
4.7. The argument generalized
The six versions of relativism I have just considered may not be the only ones. New relativist theories are constantly springing up. However, all relativist theories must fall into one of three categories, as laid out in section 1.4. Since objectivism states that moral judgements correspond to facts about the objects to which the judgements are applied, subjectivism must say (1) that moral judgements are not judgements at all and do not have propositional contents (that is, don’t represent genuine claims) or, if they do, (2) what they claim is always false, or, if it is true, (3) it represents something about the subject making the statement rather than the object the statement purports to be about. That these are the only three alternatives possible can be demonstrated from two trivial axioms, namely, the law of excluded middle and the correspondence theory of truth. For if moral judgements represent claims, then we know from the law of excluded middle that they must be either true or false. That is just basic logic. And if they are true, then we know from the correspondence theory that that means they correspond to reality. And, finally, if they correspond to reality but they don’t correspond to the nature of the object then they must correspond to the nature of the subject.
But each of these three views is surely false. The first has been refuted above in section 4.2, in which I presented four arguments to the effect that a moral statement is a proposition. The second runs contrary to patent observations that virtually everyone can see, such as the preferability of happiness to misery, the impermissibility of murder, etc. And the third view, which usually leads to commission of the naturalistic fallacy, can always be refuted by simple thought experiments, the general point of which is to hold the nature of the object constant and vary assumptions about the nature of the subject, and notice that the moral qualities remain unchanged. For instance, supposing that we all liked Nazism, yet all the same, it wouldn’t make Nazism right; supposing that we had certain emotions, it would not justify genocide; et cetera.
Most versions of relativism involve a reinterpretation or redefinition of moral judgements. What is common to all of the redefinitions of moral concepts is that they leave out everything moral. The very essence of the concept of rightness is that something’s being right is a reason to do it. But to say that I like something is not to give a reason for doing it – if somebody said, “Why should we do A?” and I said, “Because I like it,” this would not give him a reason. To say that my society approves of something is not yet to give a reason for it either. To express one’s emotions does not give anyone a reason for action. Et cetera. But something’s being good or right is a reason for doing it (indeed, in the latter case, an absolutely compelling reason).
I have tried to show that, like most false philosophical theories, moral relativism dissolves under clarification. In order to evaluate relativism and objectivism in ethics, we must first give each of these theses a clear meaning. I have defined objectivism as the view that some moral properties appertain to objects in virtue of the nature of those objects. This confined subjectivism, defined as the denial of objectivism, to three possible circumstances under which it could be true: (1) if there were no moral propositions, (2) if moral propositions were universally false, or (3) if the truth of moral propositions depended on the subject who judges them. Although moral subjectivists are usually equivocal vis-a-vis which of these alternatives they mean to assert, we have found that the positing of each of them is flawed in its own way, leaving the relativist no logical space in which to stand. In understanding the issue, it is thus essential to distinguish the different sub-alternatives discussed and pin any given version of relativism down to one of them.
I have also considered some arguments that relativists advance. Whereas one might initially have thought that relativism, being by no means intuitively obvious, would require some pretty compelling arguments to have so firmly convinced such a large majority of the intellectuals of our society, the forthcoming arguments are typically disappointing. The existence of disagreements, by no means unique to ethics, does not imply the absence of facts about which to disagree, and I have explained this disagreement otherwise. The epistemological problem about ethics is paralleled by epistemological problems that could be raised about mathematics, metaphysics, or any other a priori discipline, and should be resolved in the same way, by appeal to the general intuitive cognitive faculty that we humans seem to have. And the appeal to the virtue of toleration, we found, constitutes a better argument for objectivism than for subjectivism. These relativist arguments must be admitted to be at best inconclusive, if not clearly unsound.
Finally, recall that I argued that the acceptance of relativism would undermine all morality. Although the apparent undesirability of this consequence does not prove the theory to be false, if our initial, intuitive confidence in our moral theories is greater than the prima facie plausibility of the arguments offered on behalf of relativism, as certainly seems to be the case, then it would be irrational to reject to former in deference to the latter.
For all of these reasons, I conclude that relativism is both pernicious and logically untenable.
1. Mackie, J.L. Ethics: Inventing Right and Wrong (New York: Viking Penguin Inc., 1977) pp. 16, 106.
2. Nicomachean Ethics I.3, 1094b24-28.
3. An Enquiry Concerning the Principles of Morals, section I.
4. An Enquiry Concerning the Principles of Morals, section I.