There two basic types of ethical fact: (i) values, i.e. facts about what is good, bad, or neutral; and (ii) duties, i.e. facts about what is permissible, obligatory and forbidden. In this post I want to consider whether or not there is a defensible non-theistic account of values. In other words, is it possible for values to exist in the godless universe?
Obviously, I think it is, and I have defended this view in the past. But today I’m going to look at Erik Wielenberg’s defence of this position, as outlined in his excellent little book Robust Ethics. The view he defends can be called robust ethical non-naturalism. According to it, moral facts are non-natural and metaphysically basic. Wielenberg holds that this is true of all moral facts (i.e. duties as well as values) but I’m only going to focus on values for the time being.
Robust ethical non-naturalism is difficult to support in a positive way — i.e. in terms of arguments for its specific conclusions. It tends to be defended in a negative way — i.e. by showing how no other argument succeeds in defeating it. This makes sense given that it holds that ethical facts are metaphysically basic. Such facts tend to be those that are left standing after all attempts to reduce them to other facts or to argue against their existence seem to fail.
So it is no surprise that Wielenberg’s defence of the view is largely negative in nature. But this negative structure allows him to do something important: it allows him to show how robust non-naturalism provides an account of moral value that is — at the very least — no worse (and possibly a good deal better) than the theistic accounts that are commonly used against it. In particular, he shows how the account of moral value supported by Robert M. Adams, William Lane Craig and J.P. Moreland is vulnerable to many of the same objections they level against robust ethical non-naturalism. I am going to try to show how he does that in the remainder of this post.
1. A Brief Sketch of Robust Ethical Non-Naturalism
We need to start with a slightly more detailed understanding of robust ethical non-naturalism. The view relies heavily on the distinction between intrinsic and extrinsic value. Something is intrinsically valuable if it is good in and of itself (i.e. irrespective of its consequences and other extrinsic properties). Robust ethical non-naturalism holds that all moral value is ultimately rooted in a set of metaphysically basic, but intrinsically valuable states of affair.
In fact, it goes further than this and holds that those intrinsically valuable states of affair are necessarily good or bad. To take two examples: the experience of pain is deemed to be intrinsically and necessarily bad; while the experience of pleasure is deemed to be intrinsically and necessarily good. But these are only the most obvious examples. There are others. For instance, Wielenberg thinks that being in a loving relationship with another person is necessarily and intrinsically good.
Why think that these things are intrinsically good? Wielenberg admits that this is difficult to prove, but he follows other philosophers (GE Moore and Scott Davison) in suggesting that two tests are apposite.
The Isolation Test: Imagine that the phenomenon of interest (e.g. pain, or being in a loving relationship) exists in a simple, isolated universe (i.e. a universe in which all of the usual extrinsic accoutrements are stripped away). Does it still seem to have the value you originally attached to it?
The Annihilation Test: Imagine that the phenomenon of interest is completely annihilated (i.e. no trace of it is left in the universe). Is the universe now shorn of the value it had (i.e. does the universe seem better or worse off)?
Wielenberg argues that things like pain, pleasure or being in a loving relationship pass both of these tests. For example, if you imagine a universe in which nothing except your loving relationship exists, then it still seems like you have something that is good; and conversely, if you imagine a universe in which that loving relationship is completely annihilated, it seems like the universe is slightly worse off as a result. Consequently, being in a loving relationship seems like it is intrinsically good. This isn’t a water-tight argument, to be sure, but there is nothing obviously wrong with it.
What then of necessity of such facts? Wielenberg thinks that all ethical properties arise, necessarily, from an underlying set of non-moral properties, in such a way that the non-moral facts make (cause to be) the moral facts. (I discussed this view of moral supervenience in a previous post). But this doesn’t mean that he thinks that all ethical facts are groundless and metaphysically basic. Some ethical facts are grounded in others. For example, the wrongness of torture could be grounded in facts about the badness of pain and the moral status of sentient beings.
That said, Wielenberg does not think that ethical facts can be reduced to non-moral facts. Indeed, he thinks that there are several problems with the notion that ethical facts can be reduced in such a manner (problems discussed by the likes of David Hume, GE Moore and Mark Schroeder). So instead, he holds that there is a set of necessarily true, and metaphysically basic ethical facts from which all others proceed. These are likely to include things like the intrinsic badness of pain; the goodness of love; the badness of injustice; and so forth.
That, in a nutshell, is Wielenberg’s account of moral value. The question now is how it stacks up against theistic alternatives.
2. Robert Adams’s Theistic Account of Value
Not all theists think that God accounts for moral facts. For instance, Richard Swinburne has famously argued that certain foundational ethical truths are analytic in nature, and so do not depend on God for their existence. For those theists who deny the connection between God and moral value, Wielenberg’s account may seem pretty attractive.
But there are others who insist that God is the origin of all things, including moral facts. For them, Wieleberg’s account represents a challenge. To see whether they can fend off that challenge, we must first consider the view they themselves hold. There is, of course, no single view that garners universal approval, but the one that is typically trotted out is Robert M. Adams’s account from Finite and Infinite Goods. This is used by William Lane Craig and JP Moreland in their defence of the Christian worldview.
Adams tries to offer an account of three phenomena (i) the Good, which is the transcendent and perfect form of goodness; (ii) finite goodness, which is the type of goodness we find our world; and (iii) moral obligations. We’ll ignore the third for now and focus on the first two.
According to Adams, the Good is simply equivalent to God’s divine nature. In other words: Good = God. The divine nature simply is the transcendent and perfect paradigm of goodness. This is an identity claim, not an explanatory claim or a semantic claim. Adams is not saying that the divine nature explains goodness or that the term “Good” is semantically equivalent to the term “God”. In fact, Adams models his “Good = God” claim after another identity claim, the “Water = H2O” claim. We are all now familiar with this latter identity claim. It tells us that the substance we call water simply is the molecule captured by the chemical formula H2O. That molecule does not explain the existence of water, nor are references to H2O semantically equivalent to references to water. It is just that the latter is equivalent to the former. So it is with Good = God.
Adams’s account of finite goodness then builds upon this identity claim. In brief, Adams holds that all finite goods — like the goodness of a loving relationship — are such because of their resemblance to the divine nature. We can say that a relationship is good because it bears a resemblance to one of God’s key attributes. This is a particular account of moral supervenience — the resemblance account — that I outlined in a previous post. I offer the same diagram I offered there to illustrate how it works.
There are two important features of Adams’s account. First, like Wielenberg, Adams accepts the existence of certain metaphysically basic ethical facts. In Adams’s case those facts include things like “the Good exists, and that the Good is loving, that the Good is merciful and that the Good is just”. These facts are ethically basic because of the way in which Adams links God to the Good. Second, and related to this, Adams’s account does not provide a metaphysical foundation for the Good. Just as it would be nonsense to claim that H2O is the foundation of water; so too would it be nonsense to claim that God is the foundation of the Good. On the contrary, the Good has no foundation on Adams’s account because, like most theists, he thinks that God has no metaphysical foundation (He just is). Hence, facts about his nature are ethically basic facts.
As we shall now see, Wielenberg exploits these features in his defence of the atheistic view.
3. Does the Atheistic View make Sense?
While Robert Adams is himself open to the possibility of values in a non-theistic universe, other prominent Christian philosophers are more closed. William Lane Craig, for instance, argues that without God there can be no moral value. Furthermore, he explicitly relies on Adams’s account of goodness in defending his position. But what is it that the atheistic view lacks that Adams’s view has?
In a passge written with fellow Christian philosopher JP Moreland, Craig makes the case:
Atheistic moral realists affirm that objective moral values and duties do exist and are not dependent on evolution or human opinion, but they also insist that they are not grounded in God…They just exist. It is difficult, however, even to comprehend this view. What does it mean to say, for example, that the moral value justice just exists? It hard to know what to make of this. It is clear what is meant when it is said that a person is just; but it is bewildering when it is said that in the absence of any people, justice itself exists. Moral values seem to exist as properties of persons, not as mere abstractions — or at any rate, it is hard to know what it is for a moral value to exist as a mere abstraction. Atheistic moral realists seem to lack any adequate foundation in reality for moral values but just leave them floating in an unintelligible way.
(Craig and Moreland 2003, 492 - passage is repeated in many other writings by Craig).
We get from this that they are incredulous at the notion of robust ethical non-naturalism, but they don’t formulate their objections as an argument. For ease of analysis, I will try to rectify this. I think what they are saying can re-interpreted in the following way:
- (1) If an account of moral values entails that moral values (i) “just exist”, (ii) are not properties of persons, and (iii) float free of metaphysical foundation, then that account is false (or inadequate).
- (2) Robust ethical non-naturalism entails (i), (ii) and (iii).
- (3) Therefore, robust ethical non-naturalism does not provide an adequate account of moral value.
The obvious corollary to this is that a theistic account can provide an adequate account. But is that right?
Wielenberg argues that it isn’t. Let’s start with the claim that on robust ethical non-naturalism moral values “just exist”. Is this right? Sort of. As we saw above, Wielenberg thinks that there is a set of metaphysically basic ethical facts. These facts are necessarily true because they necessarily supervene on certain non-moral facts. There is nothing more to be said about them. But that doesn’t differentiate Wielenberg’s account from Adams’s. After all, on Adams’s account God just exists, and facts about His nature are equivalent to metaphysically basic facts. So if the “just exists” condition undermines robust ethical non-naturalism, it must also undermine the theistic view, since the divine nature just exists.
What then of the claim that ethical non-naturalism denies the fact that values are properties of persons? Wielenberg points out that this view has little to recommend to it. For starters, Adams’s view also entails that values are not properties of persons. Adams says that the Good = God, not that goodness is property of God. In other words, he is claiming that the Good is a person, not a property of a person (if it were then it would be a mere abstraction). So, again, if (ii) really is a criticism of robust ethical non-naturalism, it must also be a criticism of the theistic view. In any event, it seems silly to insist that values must be properties of persons. As environmental ethicists have long pointed it, it is arguable that values supervene on states of affairs concerning animals and the natural environment that have no persons involved in them.
Finally, what of the claim that on Wielenberg’s view values float free of a metaphysical foundation? This is true, but it is, once again, also true of Adams’s view. As I outlined above, Adams does not think that God provides a metaphysical foundation for the Good. God is the good; like water is H2O. Furthermore, there is nothing deeply mysterious or unintelligible about the account that Wieleberg is proposing. His view rests on the notion that values necessarily supervene on states of affairs and the non-moral properties of those states of affairs. Consequently, his view is no more unintelligible than any metaphyiscal view that posits the existence of states of affairs and properties (which is pretty much all of them). As he puts it himself:
With respect to justice, my view is that there are various obtaining states of affairs concerning justice, and that when individual people have the property of being just, it is (in part) in virtue of the obtaining of some of these states of affairs. For instance, I hold that it is just to give people what they deserve, thus, anyone who gives others what they deserve thereby instantiates the property of justice. The state of affairs that it is just to give people what they deserve obtains whether or not any people actually exist, just as various states of affairs about dinosaurs obtain even though there are no longer any dinosaurs….This approach is perfectly intelligible and no more posits mysterious, floating entities than does any view committed to the existence of properties and states of affairs.
Craig and Moreland’s critique is, consequently, unpersuasive. It does nothing to support the Christian worldview over the atheistic one.
To briefly recap, I have tried in this post to answer the question “Is there a defensible atheistic account of moral value?”. I have used Erik Wielenberg’s work on robust ethical non-naturalism to answer that question. According to robust ethical non-naturalism, certain moral values necessarily supervene on certain states of affairs. Some of these values are metaphysically basic (e.g. the goodness of pleasure; the badness of pain etc.). They are not founded in a deeper set of ethical or non-ethical facts.
This account of moral value is intelligible and certainly no worse than Robert Adams’s beloved theistic account of moral value. Indeed, the criticisms levelled against the atheistic view by the likes of William Lane Craig and JP Moreland can easily be turned back on the theistic view they themselves defend. Both views hold that certain ethical values “just exist”, that the values are not always properties of persons, and that the values float free from a deeper metaphysical foundation.