Why Moral Judgments Aren't True or False and Why They Don't Have to Be

One big debate in metaethics today is whether or not we can argue for the truth or falsity of value claims. Is it true that kicking babies is bad? Or eradicating ethnic groups (genocide)? Is it true that keeping promises is good? That being kind to animals is? That being one's brother's keeper is? That we should treat others as we would wish to be treated? So-called moral realists argue for the necessity of being able to claim that such statements, and a great many others, are either true or false for, if it turns out we can't, then it looks as if we have lost the possibility of believing in the rightness or wrongness of these and similar behaviors. And this seems to undermine the moral project (the possibility of selectively differentiating between a certain class of actions). Without the ability to grant validity to our moral claims, we seem to be at a loss to tell others to do or not do a whole class of actions which seem to demand just such differentiating capability from us if we are to get on, in a satisfying way, with our lives. Societies and communities stand on the capacity to make such differentiations and to defend them when we do so, i.e., to believe in the truth of our differentiations.

No one doubts that one can judge the prudential value or its lack in certain behaviors, of course. We can know if it's true that we should keep our promise to another if we come to believe that doing so will prompt some good result for ourselves, e.g., that they will keep promises in turn to us (something we wish or need them to do) or if keeping promises will yield other rewards for us. But if personal rewards are the name of the game, then it's not keeping promises that's good (the right thing to do) per se, but getting the reward. And doing something to get a reward for doing it seems to vitiate the claim of moral goodness because it allows us, when the desired reward may not be available, to disregard practices like promise keeping, indeed to pursue a policy of deliberately performing their opposites.

Moral realists want to say that moral value claims carry their own criteria for being true or false, criteria which are not the same as the benefits an agent may gain, or hope to gain, by acting in a prudentially good way. Prudential benefits are external to the presumed morally good trait or action while what makes them morally good (or not) must be intrinsic to them, part of their very nature. Making and acting on moral claims must be independent of personal benefit, the value-granting feature (the presumed property of moral goodness the action is thought to possess) existing somewhere and somehow within the act itself -- or in our description of it. It's the presence or absence of such an extra feature, a feature we discern by observation and inquiry and which we assert to be there in statements which are subject to true or false judgments, that makes any claim of moral goodness reliable (or not) and so fit to be acted on (or rejected).

Intuitionists suppose such extra features are recognized by us because we have a kind of sense of them when they are there, picking up whatever the goodness feature is in a fundamentally unanalyzable way. Just as we see colors and hear sounds and taste flavors, so, for intuitionists, we have a kind of parallel capacity to apprehend moral goodness. That was G. E. Moore's notion and the notion of others, like Henry Sidgwick, who followed him down the intuitionist path. But intuitionism lacks a certain respectability in the modern intellectual world. If we can't explain the intuitive mechanism, as we can explain how and why we see colors, etc., in some physically demonstrable way (as a function of our sensing faculties and as phenomena sensed and thus relayed to the brain), then the idea of intuitions of goodness just looks spurious.

The modern intuitionist, Michael Huemer, tries to explain intuitions of this sort as akin to Kant's a priori mechanism. Just as we know some things because they are built into our concepts (for Kant the concept structure itself is built into the kinds of creatures we are as well), so, Huemer suggests, we know properties of goodness in some things. He argues that just as "to know that purple is a color is to know a property of purple," so to know that pleasure is good is "to know a property of pleasure." But Huemer's analogy breaks down in a number of ways. Knowing a property of something can mean more than one thing in English usage. It can mean knowing the thing is, itself, a property or just knowing that it's a property of something. In the case of purple, for instance, we might say that even if purple is itself a property, and knowing that it is that is to a know "of a property" called "purple", as a property, itself, it also has certain properties of its own (such as absorbing more light than yellow, or producing a third color, brown, when mixed with pigments of red). In the ordinary sense of "property" even properties can have properties, so speaking of "knowing a property of purple" can sound okay. But it is not this sense of being a property that matters in the analogy Huemer wants to draw.

In Huemer's example of seeing purple, the property itself, purple, that we know (on seeing it) just consists of recognition that we use the word spelled p-u-r-p-l-e in English when we see that particular color. Knowing purple the color is to know how to use the word "purple" among other things. But that use, once one knows the English word in question adequately, is a constant. Knowing how to use the word is, indeed, evidence that we know what purple is, just as knowing how to respond to its presence in other observable ways would be, so we can say we know a property called purple in lieu of saying we know the property purple or a property "of purple."

But knowing "a property of pleasure" called goodness is not the same thing for, while all standard uses of the word "purple" (if not applied in an idiosyncratic way, say as someone's name) involve associating it with that particular color on the color spectrum, not all uses of pleasure involve claims that it's good. The ascription of "good" is contingent in a way that an ascription of purple is not. So goodness is not a property of pleasure in the way the color purple may be a property of some physical things, nor is it, in fact, obviously a property at all -- even if one were determined to speak of goodness as a property. Goodness is something we sometimes ascribe to particular instances of pleasure while the color we know as purple is designated in English by the word "purple" all the time. If "purple" denotes a property, the property of purple (which we call by that word), "good" does not similarly denote anything that we can attach to pleasure in the same conceptual way. So, despite Huemer, one cannot say that goodness is intuited in an a priori way akin to how we recognize things like colors.

Other efforts to claim a cognitive content for assertions of goodness hinge on a return to the once discredited notion of naturalizing the claim. Where Moore urged us to reject the idea that whatever we mean by "good" must be understood as equivalent to some naturally occurring phenomena of the world (i.e., that "good" just denotes some natural phenomenon or phenomena such as pleasure or happiness as the hedonistic utilitarians would have it), latter day naturalistic ethical theorists have revived the argument for naturalism, concluding, in the wake of intuitionism's failure to provide a basis for believing in moral facts, that naturalism is the best explanation of how we actually use a word like "good." Candidates for being that sort of goodness on this view include notions of human flourishing as elaborated by the ancient Greeks, i.e., as being the best one can be, given the nature of what one is.

Here what counts can be happiness, just as the utilitarians might have it, but also things like living a certain kind of life, avoiding certain kinds of problems, achieving certain mental states (equanimity in the face of adversity or, perhaps, maximal realization of one's intellectual capacities), etc. To the extent such flourishing can be understood as one or more of these conditions, behaviors capable of realizing them, or moving us closer to them, are supposed to have natural goodness, too. What makes such goodness natural is that what is pursued consists of the goodness thought to be intrinsic to (part of) the natural phenomena sought, the activities of pursuit being, themselves, natural phenomena, too. If the acts are good in this naturalistic sense it's because that goodness is a function of some naturally occurring goods in the world. What makes the sought after states or conditions good is that they are good for the agents who pursue them because the agents are so constructed as to need and, when they are thinking clearly enough and have sufficient information, to want them.

Goodness, seen in this way, enables us to make claims of truth or falsity about our moral judgments because whether any condition is present or absent, or attainable or not, or whether any act may conduce to such a condition or not, are all questions of fact. And it does seem to be the case that, to argue convincingly for our moral beliefs we have to be able to show others why they are true or their opposites are false. If stealing or killing are bad (wrong to do) and keeping one's word or looking after one's children are good (right to do) then it looks like we must be able to tell others why such claims are, in fact, good, i.e., why it is true to say that they are?

But are truth and falsity even the right terms to use here? Do we really need to argue that it's true that one should pay one's debts to others in some special moral sense (over and above claiming that a contract is a contract and can be enforceable under certain circumstances with consequences for the contract violator) in order to make the claim that one should pay one's debts? Why do we think that questions of truth or falsity should enter into this at all?

In fact, when we think about truth claims, what we're really doing is considering a species of valuation, just as we are when we examine goodness claims. We speak of truth values just as we speak of moral values. Truth and goodness claims seem to be similar phenomena, both valuations but performed along different axes. And it's not clear that claims about what's good in a moral sense require measurement along a true-false axis anymore than claims about what's true require judging them along a good-bad axis. Indeed, whether it's good to speak the truth in any given circumstance is a value question and not, itself, a truth question since telling the truth may not always be the best possible thing to do. That some statement is true is not, at least by itself, reason to make it.

Of course questions of truth and falsity have a place within the framework of moral discourse. We can and do ask whether certain features of a situation or object or objective or some other object of reference are the case, which is just to ask whether assertions that such features are (or are not) present are true. The fact that there are at least two sorts of valuation axes at work (the good-bad and true-false axes) does not mean that they operate exclusively of one another. But perhaps we should recognize that one thing they don't do, and don't need to do, is measure one another. Do we measure feet in meters? We can convert one to the other and sometimes such conversions will serve as necessary translations between users of the two different measuring standards. But we don't need to measure them in terms of each other. Of course this is an imperfect analogy because both measure the same thing, i.e., distance. But we don't measure distances in ounces either, or temperatures in dollars and cents. In the case of both goodness and truthfulness measurements, we are similarly comparing along quite different vectors which may work in tandem and may even complement one another but which are not, themselves, obviously applicable to the other in either case.* That we may invoke both in discussing the same thing ('is it true that X has Y?' and 'is X's feature Y good?') isn't evidence that one must be measured in terms of the other.

That both vectors of measurement may be applied to the same natural phenomena does not in any way imply that one of the vectors is susceptible to measurement along the scale of the other. One mistake made in the moral realist/anti-realist argument seems to be the supposition that 'X is good' if and only if 'it is true to say that X is good'. But what makes any statement true is the extent to which that statement fits some schema of truth (whether by correspondence or coherence or instrumentally or in some other agreed upon fashion), while what makes any action good will be the extent to which it provides a potential actor adequate reason to do it and moral judgments are about valuing actions. But there is some ambiguity here. For we do use "good" for many other kinds of referents than just human actions. A good statement, for instance, may be that to the extent it serves its user by being true because the user will thus have a reason to accept and act on it or to otherwise make use of it. A false statement, acted upon, may lead to very unpleasant consequences so, of course, truth matters. But so does goodness and the two, goodness and truth, are not the same -- though both represent modes of valuation.

Ascriptions of goodness work by placing their targets on a scale of actionability (i.e., they announce the fitness or unfitness of taking some actions) while ascriptions of truth serve to place their targets on a scale of believability. Believability has implications for what we do, of course, but it does not determine what we do alone for there is always the further question of what we want to do or come to think we should want to do. Knowing what's true is never enough to generate action without also having some motivating reason to act. There is, indeed, no obvious reason to think acting in certain ways rather than others, when the facts are the same (as established by what is known to be true), should hinge on knowing that to act in that way is true. Indeed the very usage sounds peculiar for, in the usual sense of "true," as in the statements we make matching (however that is determined) the world, such matching of our words to states of affairs has no relevance for our motivations -- except to establish the terms and conditions within which motives may arise or be suppressed. It's not an action's truth that we need to know in order to choose to do it but its goodness, even if the truth of statements which describe or indicate the action's objectives is relevant. Being good is a completely different dimension of assessment than being true. As Hume showed a long time ago, facts are facts and do not imply our judgments of goodness though, of course, claims of fact may affect them to the extent they accurately reflect changes in the phenomena we are attending to and, thus, judging.

If this is so then the whole argument for moral realism, whether naturalistically arrived at or via some claim of intuition, must strike us as misguided. Certainly claims of goodness and badness have a reality to them but it is not a truth-susceptible reality. What makes a moral claim real isn't the extent to which it expresses some true-false claim, insofar as we can find evidence for it in the world (or adduce a logical proof for it based on unassailable premises). What makes it real is its capacity to express a sentiment we have reason to have (and thus to nurture in ourselves and others). We must, of course, be able to argue for or against our positions re: the advisability or lack thereof of taking any action. But the reason to have some sentiments (and act on them), as opposed to others, doesn't rest on what is or isn't observably or logically true but on what we recognize in ourselves -- in what we are.

Our actions are measured, not by whether they are true (for how could they be true or false in the usual sense of those terms?) but by whether, and to what extent, they reflect the underlying nature we have and which we discover by thoughtful attention to the kinds of creatures we are. Claims of moral goodness stand on self-realization which is finally a personal matter and not some objectively knowable phenomenon to which we can point, expecting others to see it, too. But this personal nature is shared just as our knowledge can be shared (which we do via truth claims exchanged with others in our linguistic community), if only along a different measurement vector. For, to the extent that we, being the kinds of creatures we are, have a common nature (recognized by each of us as shared with others in our linguistic and behavioral communities), a common nature, that is, that we can discover in ourselves if and when we are prepared to look, we also stand together with others of our kind in a common world. We don't need truth to determine what's right or wrong in this sense of a shared world, just the capacity to understand others as being creatures like ourselves.

____________________________________________________________________________________ * Although Robert Brandom argues that truth claims can be understood as an outcome of pragmatics (value claims) because what we think about anything is a function of what we take ourselves to be authorized and/or obligated to do by some statements and which may lead to statements we make which similarly authorize or oblige others to act in certain ways in their turn.