Thursday, October 14, 2010

Total nonsense

Mary Warnock picks five books on morality without religion. One is Anscombe's Collected Papers on ethics, religion, and politics. Obviously thinking of "Modern Moral Philosophy," Warnock says:
There is one essay in this book in which she says that if you don’t believe in the commands of God then you have no business as a moralist to talk about duties or what people must or must not do. There can’t be any commands except the commands of God. I think that is total nonsense, but it is extraordinarily powerfully argued and one must respect the kind of relentless logic with which she argues that the concept of duty cannot exist without the authority of God who commands it.
This is not true.

Simon Blackburn gets Anscombe's wrong in a similar way, which I criticized in an earlier draft of my forthcoming book on Anscombe's ethics. Since I have cut this part of the book out, I think it's OK for me to publish it here:

In his review of Human Life, Action and Ethics for the Times Literary Supplement, Simon Blackburn objects strongly to the suggestion that we should abandon the notion of moral obligation. He gives an example of professional irresponsibility and his imagined disapproving reaction to it. Such disapproval, Blackburn insists, is moral even without the “supernatural prop” of belief in divine authority. He seems to think that this shows Anscombe’s most famous thesis about ethics to be “poppycock.” However, an examiner who, in an example that Blackburn develops, accepts bribes in return for favorable grades is hardly different from an example that Anscombe develops in which one person bilks another. She explicitly has no objection at all to saying that “a man should not bilk.” This is not, as she sees it, a use of the “ordinary (and quite indispensable) terms ‘should’, ‘needs’, ‘ought’,” or ‘must’” in the special sense to which she objects. There are all sorts of reasons why examiners should not take bribes and these are just as evident to Anscombe as they are to Blackburn. His disapproval would only be moral in the sense to which she objects if he meant to imply some sort of verdict on the action, a verdict that was wholly moral, not legal or concerned with violations of professional codes of conduct or expectations. If he meant, perhaps we can say, that the action was sinful. If he means something else, then Anscombe has no objection.

Why then all the fuss about the word “moral”? Why does Anscombe insist that people who do not believe in divine law cannot talk of moral wrong, moral obligation, and so on, as Blackburn implies? In truth, she does not. What she claims is that if non-believers take from the Christian tradition a use of words that implies a judge and a law, a use of words that implies a verdict, “with a characteristically solemn emphasis,” and yet deny that any such judge or law exists, then the use of those words has a psychological effect but not the desired, original meaning. Like Blackburn, Anscombe does not object to any particular words. It is a particular use of a related set of words that she complains about. Anscombe’s complaint is much milder than Blackburn implies. There is no laying down the law on what people may or may not say. What Anscombe objects to is not any words but “the notion ‘morally ought’,” as I have described it, and what she says about the use of this notion is that, “It would be most reasonable to drop it” because “It has no reasonable sense outside a law conception of ethics” and “you can do ethics without it.” Blackburn writes of the words “morally wrong” that “I may choose to avoid the words, if I wish, but that is by itself of no interest, and if I feel I must avoid them because I have been told that they are the private preserve of people who believe in divine law, then I have been hoodwinked and robbed.” Anscombe, though, claims no such thing, and so can be guilty of no such hoodwinking or robbery. And the fact that Blackburn could avoid those words if he wanted supports Anscombe’s claim that you can do ethics without them. More to the point, as Anscombe implies, unless we are relying on the mesmeric effect she describes, we will do moral philosophy better if we avoid these words. We can be much more specific, for instance by calling an act untruthful or unjust rather than simply wrong, and, when we are so, it will sometimes be clear immediately that the act in question is wrong. This is the basis of her claim that the concepts of moral obligation and moral duty ought to be jettisoned if this is psychologically possible because they are unnecessary and harmful outside the context of belief in divine law.
Now back to Warnock. Anscombe quite explicitly says that it is all right, quite possibly even necessary, for us to use the words "must" and "ought." It is obvious also that there can be duties and commands that do not come from God. What Anscombe objects to is a notion of duty and obligation that is not ordinary but rather moral in some way that is meant to be both special and yet utterly secular.

It might help to keep in mind two things. One is that Anscombe was very concerned indeed with the badness of deliberately targeting innocent civilians in war (with the wrongness, I think it's fair to say, of terrorism). The other is that she rejected as incoherent G. E. Moore's naturalistic fallacy, according to which: "if [a man] confuses "good," which is not in the same sense a natural object, with any natural object whatever, then there is a reason for calling that a naturalistic fallacy." Described this way committing the fallacy seems really insane. If someone thinks that "good" names an object then surely they are reasoning badly. So why did Anscombe disagree with Moore? Perhaps she thought this was too great a mistake for it to count as a fallacy.

But there is another possibility. The naturalistic fallacy is closely linked to Moore's open question argument, according to which we can see that "good" does not denote anything like the pleasant or the divinely commanded, because it is always an open question (i.e. it always makes sense to ask) whether such things are good. And it might be this idea that Anscombe rejected. Moore says we might think that "good" means something like "conducive to well-being," but (he says) we can see that this is wrong because it makes sense to ask whether it is good to be conducive to well-being, whereas it makes no sense to ask whether what is conducive to well-being is conducive to well-being (or, if that does make sense, it is at least a different question, one with a much more obvious answer). (There seems to be some confusion in Moore's thinking here involving a mix-up of sense and reference, but I'll leave that aside.)

Moore apparently thinks that it always makes sense to ask, "Yes, but is x good?" Presumably the same would go for "bad." So someone might say that doing such-and-such would bring about world peace, and Moore would think it OK to say, "Yes, but is bringing about world peace good?" Or you might point out that if I explode this device it will kill all those orphans, and then Moore might say, "Yes, but would it be bad?" Anscombe's doesn't say what is incoherent about the naturalistic fallacy (or the idea that there is such a fallacy), but she certainly would not like questions such as these. Of course world peace is good and killing orphans is bad.

Similarly, what she objects to primarily about the idea of "moral obligation" is that people say such things as: "Murdering the thousands of people in that city would be unjust, but would it be morally wrong (or 'contrary to duty' or 'impermissible' or etc.)?" She thinks that it makes sense to ask questions like this if you believe in God, because God can command unjust acts (see the story of Abraham and Isaac, for instance). If you don't believe in anything beyond the world that could possibly justify injustice then your question does not make sense. That is, it cannot be an essentially similar but God-free version of the theist's question.

It might mean anything you like, but any such meaning is unlikely to be wholesome. "Yes it's unjust, but will it be fun?" makes sense but is depraved, for instance. And if you are really talking about fun, then you should use the word 'fun,' not 'morally wrong' or whatever. So talk about moral obligation, moral duty, and the rest is either nonsense, unclear, or corrupt. Regular talk about obligation, duty, etc. is fine.

Or so Anscombe says.


  1. being naive about moral philosophy, i have to ask—

    was i wrong in hearing a substantial similarity between macintyre's thesis about the fragmentation (or whatever) of traditional moral vocabularies, and what anscombe says in 'modern moral philosophy' regarding the provenance of concepts of obligation, duty, command, etc?

  2. Not wrong, no. Peter Winch thinks MacIntyre got it all from Anscombe. Obviously MacIntyre says more than Anscombe, but the basic idea comes from her.

  3. hm, well then what are the chances it comes from somewhere more 'cultural' e.g. catholic?

  4. Catholics like it because they can connect it to the Reformation. At least Anscombe tries to do so. I think her history is a bit shaky, but the rest of it is OK.

  5. hm. maybe when i read it i assumed her history was more defensible because i was reading the greek part of macintyre's 'history of ethics' into it.

  6. She might be right. I'm not sure. She has been criticized on this score, and I don't think it matters very much (unless we want a stick to beat Protestants with), so I'm prepared to let it go. Here are some bits of the book that deal with this:

    "[Roger] Crisp argues that Aristotle does have available to him words that express the concepts that Anscombe turns to Aristotle in order to avoid. For instance, hamartanein means ‘to miss the mark,’ and Aristotle uses this word in contexts where to miss the mark is blameworthy unless done involuntarily. He also uses the word dei, which means something like ‘one ought’ or ‘one must.’ So sticking to the vocabulary available to, and used by, Aristotle will not get us away from ideas of moral obligation after all."

    "[Charles] Pigden’s main objection to Anscombe is historical. Aristotle might not have had a law conception of ethics, he says, but other Greeks did. Some early modern Christian writers have distinguished between what ought to be done and what God commands. Protestants have believed in a Divine Law.

    "To conclude [this summary of what Pigden says]: Anscombe’s conceptual thesis is based on an historical claim—that the moral Ought is a Christian product. Cicero’s On Duties demonstrates that this is false. Analogues of the modern moral concepts antedate Christianity. Her argument for giving up the moral Ought fails since it is founded on factual error. And the neo-Aristotelian alternative she proposes is not a viable option.

    "According to Pigden, Anscombe’s idea that we should abandon moral philosophy until we have done enough philosophy of psychology to allow us to work out how an unjust man or act is a bad one, in other words, to derive an ‘ought’ from an ‘is,’ is only plausible if we accept her view of the moral ‘ought.’ On the standard view, after all, there is no point in trying to derive an ‘ought’ from an ‘is’ because we know this to be impossible. Her view of the moral ‘ought’ is that it makes no sense because it is an essentially Catholic Christian concept transplanted into a modern, secular context. So to defeat these two theses it is enough to show that the history of the moral ‘ought’ is other than Anscombe takes it to be. But Anscombe’s thoughts on the cause of the problem are speculative. Her main claim is that contemporary philosophers use words such as ‘ought’ in bad ways. Nothing about Cicero can disprove this. "

  7. thanks for the quote.

    someone who appeals to a concept of duty tied to the stoic ontology to rebut someone who ties duty to the christian god does not seem bound to get what he wants.

  8. No. I'm starting to think Anscombe's history might be OK after all. At least she doesn't have much to fear in this area.

  9. On the Macintyre/Anscombe front: Cora Diamond's essay "Losing Your Concepts" has some nice material comparing the two of them (and several other figures who express similar ideas).

  10. Thanks, Daniel. I'll have to re-read that essay.