On Two Reviews of Atheism, Morality and Meaning (2003)
Two reviews of my book, Atheism, Morality, and Meaning–one by Taner Edis and one by Jeff Lowder–have recently been posted on the Secular Web. Both reviewers, although initially praising the book saying that it is “well worth reading” and that the territory covered by the book “is impressive,” argue that it has many problems. Indeed, the great bulk of their reviews consists in listing its alleged shortcomings. Someone who has only read their reviews might well wonder if the book is impressive and worth reading. But might someone who has actually read the book wonder about the accuracy and fairness of their reviews? I would like to take this opportunity to comment on the criticisms in these reviews.
Let me start with Edis’ critique of the Ideal Observer Theory (IOT), which plays a central role in my book. Edis argues that different IOs could have different reactions to the same moral situation. He points out that I try to meet this well-known objection (see pp. 99-102) but he says that that I do not do so “convincingly.” Unfortunately, he does not say why he objects to my defense. Edis also poses what he calls a theistic objection to the IOT, namely that God could bring about the convergence of the reactions of IO. According to his last objection, my attempt to show that the reactions of IOs converge is unconvincing. According to this new objection, the reactions of IOs do converge and God can explain this. Edis does not deny, however, that there are strong independent reasons to suppose that God does not exist. So if there is convergence, God is not likely to be the explanation. Another potentially damaging charge made by Edis is that I introduce ordinary language and everyday moral intuitions to support moral objectivity. He reminds us that many common-sense intuitions have been rejected by modern science. However, I briefly appeal to ordinary language on one page in the book (p. 12). My main thesis relies not on ordinary language but on the belief that my metaethics gives the best account of the meaning of moral discourse and moral reasoning (Chapters 3-5). As far as moral intuitionism is concerned, I reject it (p.23) and argue that the position of Wide Reflective Equilibrium (which I uphold) is not committed to it (p. 105).
According to Edis, I have lost sight of religious life as a way of integrating common sense perceptions and moral intuitions, and he himself goes so far as to suggest that our brains are built in a way that it seems compelling to most people that morality should be linked to supernatural realities. But surely, the philosophical point is that this integration of perception and intuition is deeply flawed and should be rejected. Moreover, I reject the view that there is an innate tendency linking morality and religion. After all, approximately 20% of world’s population consists of nonbelievers, yet we have a morality. Furthermore, if the innateness thesis is correct, children raised in nonreligious homes would have a religious morality, but this is not borne out by the facts. In addition to these alleged sins of commission, Edis brings up several alleged sins of omission. Edis argues that my book is very much an “old school book” since I do not use, for example, evolutionary psychology, cognitive science, and the findings of anthropology. He is correct that I do not, but what is not clear from what he says is why recourse to these disciplines would have helped me do what I was trying to do: criticize theistic metaethics, formulate a plausible metaethics for atheists, criticize the Christian foundation of the meaning of life, and develop a foundation of the nonreligious meaning of life. What, for example, is the relevance of cognitive science to the metaethical task that I set for myself? One suspects that Edis would have liked me to have written an entirely different book and that he takes the relevance of these disciplines to his preferred project for granted. Edis finds it “peculiar” that I don’t explore whether moral relativism (MR) is really an intolerable position and says that many naturalists are comfortable being moral relativists. However, he never says what he means by MR. As I understand MR, it entails that child abuse, torture, and the like are morally permissible relative to certain groups, times, or persons. Is this what he understands as well? If so, it seems obvious to me that MR is an intolerable position and that naturalists who are comfortable with it are misguided. If MR does not have this implication, Edis should explain why, against all appearances, it does not. For reasons that are unclear, Edis points out that Wide Reflective Equilibrium can be used by moral relativists. This is irrelevant to my point that it coheres well with the IOT and gives it practical import.
Edis comments that I limit myself to the brittle approach of conservative theists. More-liberal-minded believers, he says, do worry about the meaning of life and believe that atheists have no response to such issues as undefeated evil and a stubbornly imperfect world in the face of every human effort to improve our lives. Edis points out that religion–despite its intellectual defects–helps people cope with such worries by its assumption of cosmic harmony. Indeed, he wonders “if religiosity remains the most pragmatic, even rational, choice for many.”
But I do not limit myself to conservative religion. To be sure in Chapters 16-18, I concentrate on the problems connected with the meaning of life in Christianity. But in Chapter 15 my critique of the religious meaning of life is rather wide-reaching, as is my defense in Chapter 13 of the meaning of life without religion. Although Edis is surely correct that we live in an imperfect world, it is a mystery why he seems to suppose that, despite advancements in medicine, improved standards of living, and so on, humans have made no progress in improving their lives. No doubt religion helps some people cope with worries about the meaning of life, but given the incoherence of the concept of God, the problem of evil, the problems of religious morality, and the difficulty of making sense of the meaning of life in a religious world view, this coping comes at great price: the sacrifice of the intellect and intellectual honesty. Edis seems to suppose that this sacrifice is worth it for “many” and that, consequently, belief in a religious meaning of life is for many “the most pragmatic, even rational, choice.” On this point Edis and I differ but only in degree. In my book, Atheism: A Philosophical Justification (p.33), I argued that pragmatic considerations can justify belief but only on rare, special occasions. Although Edis apparently thinks these occasions are not so rare, it is not clear how far he would allow pragmatic considerations to govern the ethics of belief.
The seemingly most-telling critique of Atheism, Morality, and Meaning that Lowder gives relates to my defense of IOT. However, his argument is based largely on an egregiously unsympathetic reading of Chapters 3 and 5 and on serious misunderstandings. First, he says that the IO sounds like God. But, as I point out (p. 86), the IO is in at least five important respects very different from God: he is not all-powerful, not all good, not the creator of Universe, not necessarily disembodied, and he is purely hypothetical. One might add that the IO is not an object of worship, not a being such that no greater being can be conceived, not perfect, and not simple. It is important to note that the only one of these differences that Lowder cites is the fact that the IO is hypothetical. Based on Lowder’s review, a reader would naturally but falsely suppose that the IO is simply a hypothetical God. Second, Lowder claims that it is not clear how one would know what an IO would contemplate with a feeling of approval. Consequently, he is skeptical that a moral claim such as “Honesty is a moral good (MC)” could ever be epistemologically justified. Perhaps, says Lowder, the IO would contemplate honesty with a feeling of disapproval or indifference. To be sure, without empirical evidence–a priori–there would be no reason to expect one reaction of the IO over another. But on the IO Theory, ethical claims such as MC are empirical. They are verified by approximating to the attributes of the IO. One becomes as fully informed as one can, as unbiased as one can, as empathetic as one can, etc. and then one determines if one contemplates honesty with a feeling of approval. If one does, one tentatively bases one’s ethical judgment on this. Of course, such a judgment is fallible and can be changed in the light of new information, exposed biases being brought to light, and so forth. Lowder seems to object to this procedure because humans are “considerably less than ideal” and “rarely come near impartiality.” But this is to state the obvious and does not affect the point that human beings can approximate to the properties of an IO. Of course, some human beings can approximate to an IO more than others and closer approximation is possible in some situations and not in others. It is well to remember that we all make comparative judgments that some people are less biased than others with respect to a given moral issue, that some people have more relevant information with which to decide a moral issue than others, and that some people are better able to empathize with the affected parties in moral disputes than others. This is made clear in Chapters 3 and 5 on the IO Theory and also in my discussion of Wide Reflective Equilibrium in Chapters 4 and 5, which Lowder completely neglects. There I argue that human beings who make considered judgments in a state of Wide Reflective Equilibrium might be called “the real world analogues of the Ideal Observers (p.79)” Extrapolated to their ideal limit such judgments become the theoretical foundation of the IO Theory. But the basis of the theory in actual moral practice is that it appeals to relevant information, lack of bias, empathy and so forth. Lowder’s misunderstanding of my text may be based in part on a quest for certainty. Unless ethical judgments can be established with certainty, he may assume, we must be epistemological skeptics with respect to moral claims. But Lowder should know that moral claims are fallible and tentative. The IO Theory indicates why this is so and how, by specifying the procedures for making them less tentative and fallible, the justification of moral claims can be improved. Third, Lowder cannot see how the reactions of a hypothetical IO can provide metaphysical foundation for moral properties. And he asks how the IO Theory differs from error theory. Lowder asked: “Isn’t the whole point of error theory to grant that moral statements have truth value, but deny that any moral statements correspond to objects, states, or properties of that actually exist?” However, Lowder misunderstands error theory. According to John Mackie’s classic discussion of it, the objectivity of morality is an ingrained belief but this ingrained belief is false. There are no moral facts although people firmly believe that there are. (John Mackie, Ethics, Penguin 1979, p. 49). By contrast, on the IO Theory ethical statements can be objectively true since there are moral facts. To be sure, the truth of moral claims is analyzed by the IO theory in terms of the reaction of an IO: if there were an IO, then the IO would…. But this does not show that some moral claim (C) is not true. It does not follow from:
1. If IO, then R
2. IO does not actually exist
3. “C” means the same as “If IO, then R” that
4. “C” is not true
More importantly, it does not follow that, since as moral property such a moral obligation is analyzed in terms of the reactions of an IO, it is not a real property, that it is fake, that it is an illusion and so forth.
Finally, using a variant of the Euthyphro dilemma Lowder argues that the IO does not offer an explanation of moral facts. If X is good because the IO contemplates X, then, according to Lowder, goodness is subjective and the IO would be an inventor of moral facts, not just a passive observer of them. But if the IO contemplates X because it is good, moral goodness is independent of the reactions of IO and some other theory would have to do the work of explaining moral facts. It seems to me that Lowder has created a false dilemma. The IO does not invent goodness. Nor is the feeling of approval the reaction to some state of goodness that is independent of the IO’s reaction. In its primary sense, moral goodness is a relational property (p.53). Moral goodness is not invented but is the IO’s reaction in relation to the situation. For the same reason, it is not an independent property. Consider this analogy: one might argue that aesthetic worth is neither in a work of art nor invented by the viewer, rather it is a relational property, the feeling of aesthetic satisfaction of a viewer who has certain properties such as sensitivity, knowledge of art, and so on in relation to viewing the work of art.
Another potentially damaging criticism is Lowder’s critique of my treatment of moral objectivity. He says that my definition of moral realism in terms of the existence of moral facts is acceptable, but he complains that I give no “rigorous” definition of objective morality. One usually means by a “rigorous” definition one that is precise and formal. Although I give several definitions of objective morality, none of them is precise and formal, not even my preferred definition in terms of moral facts. Usually precise and formal definitions are not helpful and I doubt whether they would have been in the present case. In any case, it would seem that Lowder’s real objection to my account of objective morality in terms of moral facts is that it does not capture “the intended meaning” of objective morality. Yet, this seems to beg the question by assuming that there is one intended meaning whereas “objective morality” has different senses. Using Sayre-McCord’s account, Lowder argues that defining objective morality in terms of moral facts does not exclude subjectivism. He says that objectivism is not the belief in moral facts, but the belief that the appropriate truth conditions of moral claims make no reference to anyone’s subjective state or to the capacities, conventions or practices of any group or people. Yet he does not explain why this captures the intended sense. Religious believers maintain that theistic morality is objective and atheists have to work hard to show that theistic morality is not. But on Lowder’s account it is immediately shown to be subjective. Prima facie, this implication suggests that it does not capture at least one commonly intended sense of objective morality. The IO Theory–whatever its ultimate problems–is thought by many to be the most objective form of naturalism. Yet, on Lowder’s account, it is shown at once to be subjective. Prima facie, this implication indicates that it does not capture at least one intended sense of objective morality. So I remain unconvinced that more rigor is what is needed or that we have a clear understanding of what Lowder is after.
Lowder also is concerned that I do not meet the objection that atheists cannot have motivation for being moral. Interestingly enough, he latches on to only one small strand of my defense that atheists can be motivated–and misunderstands it. For the record, I advocate externalism: the view that moral facts are not inherently prescriptive and that if a moral fact is motivational, this is contingent both on what the moral fact is and on the psychological state of the agent (p.37). I also refute two arguments against the possibility that atheists cannot be morally motivated (pp.27-34) and in the process indicate that atheists can be motivated by everything from high moral ideals to prestige (p.31). In addition, I show that atheists might be motivated to work for the triumph of justice even if this were unlikely so long as such triumph were considered a great value (p.33) and I also say that in some cases atheists might be motivated to be moral even if it were known for certain that evil would triumph (p.33). Furthermore, I show that even if atheists have motivational problems, so do Christians (pp.177-180).
In a curiously unsympathetic reading, Lowder neglects most of these points and focuses on my claim that atheists can work for the triumph of justice even if this is unlikely saying, “if the desired outcome is unlikely, then one may well be unmotivated to perform it.” But this neglects my point that a small chance of success combined with the great value of the outcome can be motivational. If the high probability of an outcome were necessary for motivation, no one would be motivated to buy lottery tickets, and social reformers would not be motivated to work against overwhelming odds. Lowder wrongly takes my argument as to why people might be motivated to be virtuous even if it was certain that evil would triumph to be based on confusion between the normative point that people should be virtuous and the motivational point that people are motivated to be virtuous. A more sympathetic reading should reveal no confusion. For example, I suggest that moral compassion and charity would help people cope with a reign of terror. Clearly what I intended is that one could be motivated to adopt certain virtues in order to practically cope with evil.
In addition to my alleged sins of commission, Lowder brings up alleged sins of omission. Lowder cites three moral arguments for God’s existence that he says I do not cover in Atheism, Morality, and Meaning although, he claims, they are among the most popular and sophisticated available: The law implies a lawgiver argument, Adams’s Argument from Social Nature of Obligation, and Moreland’s Hybrid Moral Argument. Pointing out that I did briefly address the law implies a lawgiver argument in my other writings, he finds it “odd” that I don’t explicitly address it here. Since, however, I did address it elsewhere, and what I say in Atheism, Morality and Meaning entails its falsehood, I did not think it necessary to discuss it here. As for the two other moral arguments, I am not sure that Lowder’s claim that they are the most popular and sophisticated available is correct. Certainly, reasonable people can differ whether discussions of them should have been included in my book.
Lowder also maintains that I did not consider the justification of morality in my book. This issue is usually formulated by the question: Why should I be moral in those cases where being moral is not in my self-interest? Perhaps I should have said something about this, but I don’t think there is much to say. If the justification of morality is in terms of self-interest and if self-interest sometimes diverges from morality, then by definition there is no justification in those cases where it diverges. For atheists it is true that there is divergence but the amount of divergence is an empirical matter. However, even if God exists, insofar as Heaven is a reward for faith and not for moral behavior, it is also true for Christians: being moral would not be correlated to heavenly reward. My discussion in Chapter 17 is relevant to this point. Moreover, if God does not exist, and believers hold that Heaven is a reward for moral virtue, they are no better off than atheists in terms of justification: they believe falsely that self-interest and morality always converge.
I am most grateful for the detailed attention Edis and Lowder have paid to Atheism, Morality and Meaning and I only wish that their criticisms had been more accurate. I hope that readers of their reviews will read my book with these comments in mind.
 Taner Edis, “Can Speculative Philosophy Give Us Objective Morality?: A Review of Atheism, Morality, and Meaning” (Prometheus Books, 2002).