Some Thoughts on Naturalism and Morality

It is supposed, by some, to be difficult for naturalism to account for moral properties (both axiological properties like goodness and badness and deontic properties like rightness and wrongness). William Lane Craig and Paul Copan, have each argued incessantly that naturalism cannot account for moral properties. Craig has offered the following argument:

  1. If God does not exist, then objective moral value does not exist.
  2. Objective moral value does  exist.
  3. God exists.

This argument has always struck me as a paradigmatic example of an extremely unconvincing argument. I see no reason at all to believe that premise (1) is true. This is for two reasons: First, I can’t see how the existence of God is supposed to account for the existence of moral value; and second, I can’t see any reason to think that moral value cannot exist in a universe without God. The intuition that Craig (and many others) has is difficult for me to grasp. I just don’t get it.

Consider the following argument:

  1. Killing babies is horrendous.
  2. Even if God does not exist, killing babies is horrendous.
  3. So, even if God does not exist, objective moral values do exist.

This, to my mind, is so compelling as to be decisive. I do not see what the existence of God adds to the universe such that, if he did not exist, the killing of an innocent child would not be horrendous.

Suppose that God does not exist. How does that change anything about a situation in which a person is killed in an undeserved and excruciating manner? What possible reason could we have to think that God’s non-existence entails that this situation is not horrible?

In this post I am going to do my best to try to articulate the intuition behind the thought that, in a universe without God, there can be no moral value. I will then show that this intuition is completely misguided.

Here, I think, is the line of reasoning that lies behind the thought that without God, objective moral value cannot exist: If God does not exist, there is no basis for claiming that one state of affairs (or action) is better than any other. Any claim to the effect that something is better or good or bad or right or wrong is just a subjective and arbitrary preference. For objective moral value to exist, there must be some ground or foundation of this value and this foundation cannot be the merely subjective or arbitrary whim of human beings. If God does not exist, there is nothing objectively special or significant about any object, state of affairs, or action.

Often the above reasoning is supplemented with a kind of argument from incredulity concerning naturalism. On naturalism, the reasoning goes, all there is is matter and the void. Naturalism provides no basis for accounting for objective moral value. Everything, on naturalism, is the product of mindless and valueless natural laws operating on mindless and valueless physical objects. How can you get value out of that?

That, such as it is, is the best I can offer to capture this line of reasoning. I think that if you listen to or read Craig and Copan (and others) on this issue, you will not come across anything more compelling than the above (though, if you can find something more compelling, please bring it to my attention). There are presentations of the reasoning, to be sure, that are more sophisticated in virtue of making fine distinctions and so forth; but the essential reasoning remains the same.

Now, the first thought I have when I encounter such reasoning is this: How is the existence of God supposed to change any of this? How does the existence of God explain the existence of objective moral value? After all, if value depends on God, wouldn’t value just be a matter of God’s subjective and arbitrary preferences? I think that this thought, fully articulated, can be used as the basis of an argument that is devastating for the no-moral-value-in-a-godless-universe intuition. But let’s put that issue aside for now because I want to show that there is just no basis for thinking that moral value is impossible on naturalism. Indeed, the belief that there is no such basis is itself founded on a false conception of what moral value is.

I think that sometimes when people talk about the basis or ground of objective moral value, they are thinking along the lines of an analogy with the way in which macro-level physical properties are reduced to micro-level properties in physics and chemistry. For example, according to thermodynamics, the heat of a room is identical to the mean kinetic energy of the air particles in the room. In this instance, a macro-level property, the heat of a room, is reduced to a micro-level feature (average kinetic energy of a population of microscopic particles). One of the important consequences of this reduction is that it tells us that heat is nothing but mean kinetic energy. That is, on the basis of this thermodynamic reduction, we discover that heat, at a fundamental level, just is air-particle movement.

I suspect that many people have this kind of model in their mind when they demand an account of the grounding of moral value. They want to know what objective moral value really is at a fundamental level. Indeed, when you study the religious moral theories that ground moral value in God, you often find defenders of these theories utilize precisely this kind of analogy with scientific reduction. On Robert Adams’ divine command theory of moral obligations, for example, we are told that moral obligations just are divine commands, in just the same way that water just is H2O. Adams’ himself offers this analogy and I think it is very significant that he does.

According to the scientific reduction analogy, moral properties are analogous with physical properties. Goodness, badness, rightness, and wrongness, are properties in much the same way that size, shape, mass, kinetic energy, etc. are properties. The task for moral theory, on this conception, is to discover the fundamental features of reality that underlie and thus ground moral properties; just as the task of science is to discover the fundamental features of reality that underlie and account for the macro-level properties with which we are familiar through our five senses.

I think that this analogy is fundamentally misguided. Moral properties are not features of reality in the way that coolness, heat, wetness, solidity, etc. are features of reality. Morality is not concerned with a special class of properties, the moral ones. Rather, morality is concerned with a special kind of significance: moral significance.

So, what is moral significance? Well, let’s put aside ‘moral’ for a minute and just think about significance. Significance concerns mattering; when we ask what is significant we are asking what matters. And, it is clear that significance comes up in life in many different ways that are unproblematic. For example, it arises in the context of evidence. Suppose that someone you know is being investigated for a murder. Suppose that the medical examiner has determined that the victim was slashed with a large kitchen knife. You and I both know the suspect and have visited his house on many occasions. When we learn of the medical examiner’s determination as to the cause of death, you say that you are worried that our mutual friend, the suspect, might actually be guilty because you’ve noticed that he has a knife block on his kitchen counter that contains a large knife of just the type indicated by the examiner. Is this fact significant? Pretty clearly it is not. That the suspect owns a large kitchen knife is not at all significant due to the ubiquity of such knives; nearly every house on the block has such a knife and so the fact that the suspect owns such a knife can hardly be thought to be significant with respect to the question of his guilt or innocence.

Notice that this is not a matter of opinion. Nor is it a matter of subjective or arbitrary preference. It is not just that I don’t want the existence of the knife to be significant. It is insignificant in a completely objective way, regardless of what anyone thinks, wants, needs, or values.

Another example: A friend tells you that he now believes in Bigfoot because he went camping in Northern California last weekend and, during the night, he heard strange grunting sounds that were unlike anything he had ever heard before and that, after much reflection, he could not account for them. “What else could it be?” he asks. Well, once again, the evidence on offer is not significant with respect to the claim ‘Bigfoot exists.’ That your friend could not identify the noises in no way indicates that they were the noises of a North America great ape. Now, the discovery of the body of a dead, unidentified apelike creature in the North American woods would be significant, at least much more significant than the strange noises.

Once again, the claims about significance that I have made are not subjective, nor are they a matter of preference or arbitrary whim. They are completely objective. And this is an important (IMPORTANT!) point: the significance of lack or significance of a piece of evidence is not an extra, special property of the piece of evidence. That is, it is not as if the dead body has, but the sounds do not have, some special property called ‘significance.’ There is no such special property, at least not if we are thinking of properties in the sense of physical properties. Once all of the properties of the bead body are accounted for (its height, weight, color, age, bodily proportions, anatomy, etc.) there is no additional property corresponding to its significance. The significance of the body is just a matter of its being properly related to the claim ‘Bigfoot exists.’ The body is properly related to this claim in a way that the unidentified noises are not. That is what the significance of the body, as a piece of evidence for the claim that Bigfoot exists, consists of.

So, significance is not a special kind of property that is analogous to physical properties. Thus there is no reason to suppose that the significance of something has to be reduced to some basic, fundamental feature of reality. The significance of the body of a dead Sasquatch would in no way be reducible to some fundamental feature of reality, and to suggest that it must be so-reducible is to miss the point entirely. Imagine if someone gave the following argument:

  1. If God does not exist, then objectively significant evidence does not exist.
  2. But objectively significant evidence does exist.
  3. So, God exists.

Obviously this is a ridiculous argument. It is ridiculous because it has an inappropriate (one might say overly-metaphysical) understanding of evidential significance. (And, by the way, if significance in the realm of evidence did depend on God, then it would hardly count as objective since it would be a matter of God’s subjective and arbitrary preferences.) I think that, in exactly the same way, Craig’s argument (and all suggestions that naturalism cannot account for objective moral value) are based on an inappropriate (and overly-metaphysical) understanding of moral value.

Now, let’s get back to the ‘moral’ in ‘moral significance.’ What does it mean to say that something has moral significance? Well, based on what I’ve said so far, we know that it is not a matter of a thing’s having a special kind of property that is analogous to physical properties. Rather, it is to say that the thing matters. But to say that it matters morally is to say that it matters is a particular kind of way with respect to particular kinds of concerns. Morality concerns practical reason; that is, it is a matter of what we should do (deontic moral value) and what is worth pursuing (axiological moral value). This is one way in which moral significance differs from evidential significance. Evidential significance concerns theoretical reasoning, while moral significance concerns practical reasoning. The other essential feature of moral value is that moral value is (or at least tends to be) overriding. That means that moral values will override other values and other concerns in practical reasoning. Here is one example that illustrates this point: Suppose that I have a desire for some particular consumer good. Acquiring this good has value in virtue of the fact that it satisfies my desire. But suppose that acquiring that consumer good would be or would lead to a morally bad state of affairs. In that case, the moral value trumps the other concerns and thus, all else being equal, I ought not pursue that particular consume good. So as to focus on the relevant aspects here, I will choose an example that is relatively uncontroversial. Suppose I want to buy a yacht and suppose further that, just to be absolutely silly, I have the money to purchase one. But suppose that acquiring this yacht would involve morally bad states of affairs. Perhaps the production of the yacht involves the exploitation of labor or perhaps, once I buy the yacht, I will not be able to pay for the surgery that my wife needs to save her eyesight. Here is the point: to say that there are moral values is just to say that there are objects, states of affairs, and actions that, at least potentially, override my desire to purchase the yacht. In this example, pretty uncontroversially, my wife’s eyesight is more significant than my desire for a yacht. Her loss of eyesight has moral significance; it is so significant that it overrides other concerns and other interests.

So, moral significance is a matter of there being things (objects, states of affairs, conscious states, actions, for example) that have significance that tends to be overriding. Since moral value is a matter of significance, we cannot think of moral value as a special kind of property. Whether something has moral value is thus not a matter of whether it has some special property, but whether it matters in the right kind of way.

Now, with that out of the way, I want to say that there is no reason to think that moral value cannot exist on naturalism. Naturalism is the claim that the only things that exist are natural objects, natural laws, and natural properties. But ‘moral value’ is not the name of a special kind of property; rather it is the name for a special kind of mattering. Notice, first, that naturalism makes no claim about what matters; so at least prima facie, there is no contradiction involved in believing in naturalism and also believing that some things matter morally. The claim that moral value can exist on naturalism is just the claim that some natural properties or objects matter in this special kind of way. That is, some natural properties, objects, or states of affairs are significantly related to practical reason such that their significance overrides (or tends to override) other values (that is, other things that are significant or that are desired). Now, what is supposed to be the problem with this? Why are we supposed to think that nothing that is purely natural can matter in this way?

I think that pain has moral significance. And I think that pain is natural phenomenon. Why can’t pain be such that it is significant and its significance tends to be overriding? Indeed, isn’t this precisely the case? Well, suppose someone asks, “But in virtue of what is pain morally significant? In virtue of what does pain have this special status?” That is a really good question and one that we should tackle. I suppose that a more general version of it, asked in a kind of rhetorical way, might serve as the basis of a criticism of my claim that moral value can be accommodated by naturalism: “Jason, you say that moral value is just a matter of things be significant in the right kind of way. But, on naturalism, what accounts for the fact that anything has that kind of significance?”

As long as the question is not meant rhetorically, as if it is obvious that there is no answer, then it is a fine one to ask. And, I think it can be answered. But lets’ first remember that answering will not involve reducing the moral value of pain to some more fundamental aspect of reality; it will instead involve explaining why pain is significant in an overriding way (or, in general, explaining why anything is significant in this way). Now, once you formulate the question in the right way, it seems to me that it answers itself (at least for any person who has ever experienced pain). Pain just is significant; its significance is not reducible to or identical with any other more fundamental feature of reality. Rather, its significance is a matter the role that pain plays in the structure of human motivation. Pain, qua pain (or pain, all things being equal), is something that sentient creatures have a strong tendency to avoid. Sentient beings will not readily undertake activities that they know will cause themselves pain and will only do so when there is some other overriding good that the creature believes can only be acquired via the painful activity. So explaining why pain is morally significant is just a matter of explaining the role that pain plays in practical reason. Once we see that, it seems to me that it is not at all mysterious that naturalism can account for the moral value of pain specifically, and moral value more generally.

What about deontic value? Some philosophers have felt that accounting for moral obligation is the really hard problem for naturalism. In virtue of what can an action be morally required, on naturalism? Well, once again, we need to make sure that we understand the question correctly. First let’s note that moral obligations concern moral reasons and moral reasons are reasons for action. Once again, it is not a matter of actions having a special kind of property. To say that I have a moral obligation to do A is to say that I have a reason, which will tend to override other concerns, to do A. Now, can there be such reasons on naturalism? I don’t see why not. Unless there is some special reason to think that there cannot be practical reasons on naturalism, I don’t see the problem. What is it about naturalism that is supposed to rule out such a possibility?

Since this post is getting too long, I will end here and wait to see what devastating line of criticism are on offer.