An Argument for Moral Anti-Realism

It's been more than two years since I've last made a post, but here I am again. Truth be told, my original posts (which ran about a year from 2014-2015), fleshed out what already was my most well developed thoughts in philosophy. There just wasn't anything more I wanted to say in the format of an essay.

With the exception of one provocative question:

Do moral statements have truth value?

In my view, this is the principal question in meta-ethics. Without an affirmative answer to this question other normative projects in philosophy, such as practical ethics, cannot get of the ground. Answering this question is absolutely crucial for making sense of moral discourse.

Given the ramifications of answering this question, I have been hesitant to put out a definitive statement on my views. Moreover, the topic has and continues to vex me. What I seek to do in this essay, then, is to articulate an argument that, as far as I could tell, is original to me (albeit with heavy indebtedness to Kant and G.E. Moore) in favor of answering this question in the negative. In other words, I defend a position known as "moral anti-realism". After flushing out what I believe to be a strong argument in favor of moral anti-realism, I add some thoughts as to how moral realism can be salvaged.

To begin, what is meant by "truth value"? A statement has truth value, if it could be true or false. For example, "2+2=4" has truth value since it could be said that this is a true statement. Moreover, "The sun is the size of a jelly doughnut" also has truth value, but in this case, the statement is false. These two statements run in contrast to statements of taste, such as "Wendy's breakfast menu is disgusting." Though the sentence structure is similar, the sentence does not have truth value, unless it is modified to be "Wendy's breakfast menu is disgusting to me".

In contrast to the aforementioned statements, there is great puzzlement on the status of moral statements in regards to truth value. What exactly is meant by a statement like "It is wrong to profit off the sickness of others."? Is it declaring a fact? A mere feeling? Various schools of thought have developed on both sides of the realism/anti-realism divide. This post, however, will focus on the divide itself, given its critical importance.

Before delving into the argument, however, I would like to briefly address the temptation to cast the "truth value problem" as a conflict between "objective" and "subjective" morality. Someone who believes in "subjective morality" may claim that yes, there are moral truths, but they are true only for the person who affirms them. Essentially, moral truths are statements of taste. My thought on this position is that it does incredible violence to the concept of morality. Moral statements contain a certain force, a force of obligation, that gets lost when statements are reduced to statements of taste. In my opinion, either moral truths are objective (are not matters of taste) or they do not exist at all.

Now onto to the argument. If it can be shown that moral statements do not have truth value, then moral anti-realism will be affirmed. The following is my argument for moral anti-realism:
1. Moral statements are not analytic statements.
(i.e. they are not synthetic statements).
2. If moral statements have truth value, then moral statements must be analytic statements.
3. Therefore, moral statements do not have truth value.

The argument is logically valid and therefore if both premises are true, then the conclusion logically follows. Before the premises can be defended, however, a brief exposition on the nature of analytic and synthetic statements is required.

Kant classified factual statements into to two basic categories "analytic" and "synthetic". An analytic statement is a statement that is true in virtue of the fact that it's subject is contained in the predicate and vice versa. Take the example, "All thinking men think". This is a logical truth as "think" is found in the subject "thinking men". This is also known as a tautology.

Synthetic statements are more substantive. In contrast to analytic statements, synthetic statements do not feature the subject as being a part of the predicate conceptually. Take the previous example, "The Sun is the size of a jelly doughnut". This is not a logical truth, but instead one established by empirical evidence.  With this classification as a basis, Kant launched his famous epistemological project in the Critique of Pure Reason: How are synthetic, a-priori(before evidence) statements possible? We do not need to get into this project here, but the distinction is crucial for the argument, and for that, the argument is indebted to Kant.(Ironically so, as Kant was the ultimate moral realist!).

Now on to the  defense of the premises; naturally I shall start with premise one. The best demonstration of premise one's truth is G.E. Moore's "Open Question Argument" (OPA). Through the argument, Moore shows that "goodness" cannot be said to be analytically identical with some non-moral property. The reasoning is as follows: If property X were to be identical to goodness, then the question "Is property X good?" would be a closed one. To explain, it would be like asking "Is a bachelor unmarried?". The question is analytically (logically) "Yes". However, for any given X "Is property X good?" remains an open one. Take the example of pleasure. It may seem reasonable to equate pleasure with goodness but it is also reasonable to deny this (consider the dystopian world of "Brave New World"). It cannot be said, however, that it is reasonable to answer "No" to the question "Is a bachelor unmarried?". This openness demonstrates that no property has intrinsic moral content, which means moral statements are not analytic.

Onto Premise Two. To begin the defense, I contend that because of this "openness" any postulating of a property as the essence of goodness, will suffer from an element of arbitrariness, a potentially fatal problem for moral realism if unresolved. Without a solution, it cannot be definitely affirmed, even in principle, that either a certain property (ex. happiness) or type of action (ex. an action done out of duty) constitutes the essence of goodness. The arbitrariness that pervades theories of goodness enables the proliferation of contradicting positions. Different intuitions may point to one theory over an other, but the arbitrariness problem seems to indicate that intuitions are all there is. But no person worthy of being called a moral realist would claim this. It thus seems that the arbitrariness problem must be solved, if moral realism is to be rescued.

Now to the point of the premise: there would be no "openness" and thus no problem of arbitrariness, if moral statements were analytic. If properties such as love or happiness had intrinsic moral content, then questions such as "Is it good to love your neighbor" would be as closed as the question "Are bachelors unmarried?". It seems, though, that since such statements are not analytic, they could only be synthetic, which makes the arbitrariness problem more difficult to solve.

How could moral truths be synthetic? Could moral truths could be grounded on something "out there", as an empirical truth is grounded in what we call the "world". Moral facts however could not be empirical in the traditional sense. One may conduct studies, for example, on the negative impact of theft or what causes it to rise or fall.  But one does not conduct a study to find out if theft is wrong (unless one presupposes that other things are wrong (ex. suffering)).

There is, however, a live philosophical option that moral truths are "out there" and that is the position of Platonism. I will divide the Platonic take on moral ontology into two types: 1. Classic Platonism 2. Theistic Platonism. I will show that neither of these options are viable for moral realism.

Both Platonic views contend that moral truth is grounded in a transcendent reality. In my opinion, the best version of Theistic Platonism is one that makes the "character" of God the reference point for moral facts (see "A Response to Dr. Craig on Beauty"). Things are good to the extent they reflect God's character. Classical Platonism postulates that "the Good" is an abstract object which exists in the same way that numbers do in Platonic metaphysics. Things are good to the extent they reflect the nature of this abstract object.

For the purpose of this essay, the crucial problem with Theistic Platonism is that it does not meet the challenge posed by the Open Question. The ancient Euthyphro problem (found in Plato's Republic) demonstrates why. For those unfamiliar, let us grant Theistic Platonism and suppose that God's character is "loving" (thus making love "good" on this view). We can ask the following question: Is love good because it reflects the character of God or does God have a loving character because love is good?

To be committed to Theistic Platonism necessitates giving the first answer, as the later brings us back to the original starting point of looking for a ground for moral truths. Unfortunately,  the openness/arbitrariness problem rears its ugly head again. It does seem to be an open question as to whether God's character is good. (Think of His controversial behavior in the Bible). On this view, no matter what character trait God exemplifies, such a trait is by definition good. This is clearly arbitrary.

Perhaps Classic Platonism can triumph where Theistic Platonism failed. On the face of it, CP seems to do better in terms of meeting the arbitrariness challenge. It seems to be a closed question as to whether "the Good" is good. However, can an abstract object be the essence of goodness?

The belief in the existence of abstract objects(which is metaphysical view that goes beyond morality) suffers from conceptual difficulties beyond the prima facie strangeness. An abstract object is supposed to be a timeless/space less object, truly transcendent. But all instantiations of goodness are temporal. How can such an object capture whatever property that is related to goodness, such as happiness, love, or honesty?  The Theistic version is superior in this regard, as it seems to make sense that a person (God) can inhabit such concepts.

In addition to conceptual issues, CP does not map on to moral epistemology (or how we come to know moral truths), which as a disclosure is not a point original to me. Though we can and do reason about morality, at the foundations of moral views are assumptions that are based on intuitions as to what constitutes right or wrong. If "the Good" is an abstract object, then if we are to have valid knowledge as to what constitutes goodness, there must be a connection between our intuitions and the object. However, abstract objects are supposed to be causally closed (they don't effect the physical world) and thus there would be no way to connect our intuitions to the source of Goodness. This makes it possible that all our intuitions are systematically wrong, or even beliefs that are the opposite of what is the case. Given that our intuitions are the reason as to why we believe in moral truth, this position is not tenable.

Thus, CP has both severe conceptual and epistemological flaws. This stands in contrast to what would be the case if moral statements were analytic. If this were the case, for moral statements to have truth value, it would not be necessary to postulate a metaphysics with strange implications, but rather all that would be needed is the veracity of logic.

If the analysis is sound, the only way for moral statements to have truth value that both solves the arbitrariness problem and avoids conceptual and epistemological difficulties is if moral statements are analytic. But as we have seen with the exploration of the OPA, they are not. It is thus difficult to escape the argument's conclusion that moral statements do not have truth value which is the thesis of moral anti-realism.

Now that this argument has been flushed out, I want to disclose that I am more than sympathetic to the position of moral realism. I would like to say unequivocally that some things are right or wrong, and quite frankly I don't think I'm going to stop speaking in moral terms. I thus, would like to briefly comment on how my own argument could be rebutted.

The weakest premise is clearly premise two. It may be denied that moral statements need to be analytic, as it could be said that the arbitrariness problem does not need to be fully solved to affirm moral realism. Maybe the most fundamental moral facts are brute, but nonetheless true. It could be said there has to be a stopping point for an explanation somewhere.

Allowing this, one account of morality that seems plausibly true and seemingly affirms moral realism, is a kind of spin-off of John Rawls' Theory of Justice, using the idea of the "Veil of Ignorance". This is the hypothetical scenario where people choose the nature of the basic structures of their society without knowing their own characteristics (gender, race, ethnicity etc.). Instead of describing societal principles, however, the account would deal with actions on an individual level. This thought is not original to me and I won't be arguing the idea here in depth, but once I became aware of the "Veil of Ignorance" it struck me as a plausible ground for right and wrong as it captures the intuition that morality is impartial as evinced by maxims such as "the Golden Rule". The anti-realist rejoinder, however, could be that there is nothing incoherent with "egoism" an account of morality that places self-interest as the ultimate value. Interestingly, unless adjusted, the account itself can also be viewed as anti-realist as it doesn't outline an objective account of "the Good" independent of individual's conception of it.

Less plausibly, premise two may come under attack by a clever philosopher who could make the case that some property (perhaps happiness) really does have intrinsic moral content if you understand the concept deeply enough (undermining premise two). I doubt, however, such a strategy would be convincing to me.

In my opinion the argument I have outlined here for moral anti-realism gets to the heart of the problem for moral realism: It isn't clear as to what it even means for a moral statement to be true or false. Providing an answer that overcomes arbitrariness and avoids problematic metaphysical claims is the key to defending moral realism in a most convincing manner.