Moral Anti-Reductionism, Objective Moral Values, and Abstract Objects: Some Very Preliminary Thoughts

(Note: I am clearing out my backlog of draft blog posts but I am officially on a hiatus from blogging. What this means is that while you are free to comment on this post, readers should not expect engagement from me anytime soon.)

(Updated 7 January 2022)

VERY ROUGH DRAFT

Let's begin with some definitions:

moral ontology: The study of whether any moral properties and facts exist and, if so, what metaphysical status they have.

By "metaphysical status," I mean whether moral properties and facts are ontologically reducible to non-moral facts and properties. In this sense, then, there are two main theories about the metaphysical status of moral properties and facts: anti-reductionism and reductionism. Let's define those next:

moral anti-reductionism (a/k/a ‘non-naturalism’): The position that moral facts and properties are not reducible to non-moral facts and properties. The standard label for this position is the rather unfortunate and undescriptive term ‘non-naturalism.’ I have coined the new term, “moral anti-reductionism,” to be more clear. (Cf. Moore, Fales, Wielenberg)

moral reductionism: the position that moral facts and properties are reducible to non-moral facts and properties. There are two types of moral reductionism:

• reductive moral naturalism: Moral facts and properties are reducible to natural, non-moral facts and properties. (Cf. Brink, Swinburne)
• reductive moral supernaturalism: Moral facts and properties are reducible to supernatural, non-moral facts and properties. (Cf. Adams)

Here I want to explore the metaphysical status of moral facts and properties, from the perspective of moral anti-reductionism. It seems to me that if moral facts and properties exist and if moral anti-reductionism is true, then moral facts and properties plausibly exist as abstract objects. But what does that mean?

Let's start with moral facts and properties. Assuming they exist, let's plausibly stipulate that they include moral values. Turning to moral values, it's useful to start with a definition of "values" and then contrast that with "moral values." In the broad sense, "values" are the things that people care about. As Rescher pointed out in his study of value theory, there are many types of values: moral (such as honesty and fairness), economic (such as economic security and productiveness), social (such as charitableness and courtesy), political (such as freedom and justice), aesthetic (such as beauty and symmetry), and so forth. Rescher doesn't explicitly define what differentiates moral value from other kinds of value; furthermore, it's not obvious that these types of value are mutually exclusive. (For example, perhaps with certain caveats, justice is both a political and a moral value.) As a very rough first step, I suggest the following definition:

moral values: properties, character traits, or states of affairs which, by default, are worthy of praise, something we expect of others, or both. The "by default" expression is included to allow for exceptions. For example, we might say that honesty is morally valuable, but there are cases in which the moral value of honesty is outweighed by the moral value of something else, such as flourishing. Think of people in Nazi Europe who sheltered Jews and then lied to the Nazis when interrogated about whether they sheltered any Jews. The "by default" expression is designed to render such counter-examples irrelevant.

Some philosophers contrast moral values with moral disvalues. Let's define "moral disvalues" in parallel:

moral disvalues: properties, character traits, or states of affairs which, by default, are worthy of condemnation, are prohibited, or both. Again, the "by default" expression is included to address obvious counter-examples. For example, we might say that dishonesty is usually morally disvaluable, but in the case of people who sheltered Jews from Nazi persecution, such dishonesty was morally valuable.  

With "moral values" defined, let's now introduce the notion of an "objective moral value." (N.B. Depending on the day of the week, I am doubtful about whether the concept of "objective moral value" is coherent, but for this post I am going to assume that it is.) Even within the context of axiology, there is no universally agreed upon definition of "objectivity." Because this post is about moral ontology, I'm going to define objectivity in an ontological sense. Okay, but what does that mean? Well, negatively, we might say that the truth of X is ontologically objective if and only if the truth of X is mind-independent. That definition is fine as far as it goes, but it doesn't go far enough. It tells us what ontological objectivity is not, but it doesn't tell us what ontological objectivity is. So, positively, we might say that the truth of X is ontologically objective if and only if: (1) X is true in virtue of corresponding to actually existing objects or properties which function as truthmakers for the propositions in question; and (2) the truthmakers make no reference to anyone's subjective states, capacities, conventions, beliefs, attitudes, or desires.

ontologically objective moral values: propositions about moral values which (1) are true in virtue of corresponding to actually existing objects or properties which function as truthmakers for the propositions in question; and (2) the truthmakers make no reference to anyone's subjective states, capacities, conventions, beliefs, attitudes, or desires.

As an abstract definition, that definition seems okay to me, but how would this work in practice? For example, consider the following proposition:

H: Honesty is morally valuable.

If we say that a proposition like H is ontologically objective (as defined above), what functions as H's truthmaker? In other words, what makes H true?

Before I explore how a moral anti-reductionist might answer, I want to quickly summarize where we are. To sum up: we're making the following assumptions:

(1) moral facts and properties exist;

(2) moral facts and properties include moral values;

(3) the concept of "objective moral value" is coherent; and

(4) objective moral values exist, viz., moral values are ontologically objective. 

The next step is to divide ontologically objective moral values into two types: derivative and non-derivative. Non-derivative values, also known as "intrinsic values" or "ends values," are values which are not derived from anything else. In contrast, derivative values, also known as "extrinsic values" or "means values," are things whose value is derived from something else, something which might be derivatively or non-derivatively valuable. Imagine a chain of value where A is derivatively valuable because of B which is derivatively valuable because of C ... which is derivatively valuable because of X, which is non-derivatively valuable. In that example, X is a "stopping point" in the chain of value. The point I want to make here is that IF there are ontologically objective extrinsic moral values, other things serve as their truthmakers. N.B. I don't know if there are ontologically objective extrinsic moral values or even if the concept of an ontologically objective extrinsic moral value is coherent, but it doesn't matter to my point, which is simply that IF such things exist, then their value ultimately derives from other things, things which are non-derivatively valuable. So let's make an additional assumption:

(5) Ontologically objective moral values exist, and include non-derivative ontologically objective moral values.

This can be simplified to:

(5')  Non-derivative ontologically objective moral values exist.

In case this is too abstract, here's a potential example (which, again, assumes the concept of an ontologically objective derivative moral value is coherent). By definition, utilitarianism seems to entail that utility is non-derivatively morally valuable. While act utilitarians might believe that, say, honesty is morally valuable, logically consistent act utilitarians must regard honesty as derivatively morally valuable: on their view, honesty is morally valuable if and only if it maximizes utility. In other words, for them, maximizing utility is what 'makes' honesty morally valuable. But what about utility itself? On act utilitarianism, nothing else makes it morally valuable to maximize utility; maximizing utility just is valuable. 

So let's add the following to our list of assumptions:

(6) Non-derivative ontologically objective moral values exist, and include X.

X could be any candidate for non-derivative value: the utilitarian's utility; the Aristotelian's contributing to human flourishing; the normative divine command theorist's obedience to God's commands; and so forth.

It seems to me that if some instance of (6) is true, it is necessarily true, but if some instance of (6) is false, it is necessarily false. For example, if utility is non-derivatively and objectively valuable, then it is necessarily true that utility is non-derivatively and objectively valuable. Likewise, if contributing to human flourishing is non-derivatively and objectively valuable, it is necessarily true that contributing to human flourishing is non-derivatively and objectively valuable. 

This leads to:

(7) Necessarily, if non-derivative ontologically objective moral values exist, they exist necessarily.

The implication here is that, contrary to myth, if moral anti-reductionism is true, it doesn't necessarily follow that moral facts and properties are sui generis. They might be a wholly unique type of entity, but they might also simply be propositions, albeit propositions about moral facts and properties. In other words, the Platonist qua moral anti-reductionist need not view moral facts and properties as a unique type of entity over and above other kinds of entities which they recognize in their ontology; rather, the Platonist qua moral anti-reductionist might believe that propositions exist as abstract objects, and non-derivative and objective moral values "exist" as a type of proposition.

The idea of linking moral anti-reductionism and abstract objects seems plausible because of the many parallels between nonreductive moral values and abstract objects. If a non-reductive moral value or abstract object exists, it exists necessarily. If a non-reductive moral value or abstract object exists, it is causally inert (i.e., the number 2 or the moral value utility cannot cause anything).  If a non-reductive moral value or abstract object exist, it is not spatiotemporal, viz., it does not exist in space and time.

This leads to one final thesis:

(8) If non-derivative ontologically objective moral values exist, they plausibly exist as a type of abstract object, namely, propositions about non-derivative ontologically objective moral value.