Ethics: Inventing Right and Wrong Paperback – May 17, 1991

A seminal classic in the ethical naturalist cannon.

This book is the father of all error theory books. He originates some of the central critiques of moral thought and discourse that guys like Richard Garner rely on (see also: Beyond Morality). They think there is something bizarre about morality's commitment to categorical reasons (things that you ought or ought not to do regardless of what your desires happen to be). They think moral properties are queer because he says that don't fit into a naturalistic worldview and because they are thought to supervene on non-moral properties (i.e. the moral properties "fix" the moral facts... so the non-moral fact of kicking a puppy is always accompanied by the property of wrongness).

As you may notice, the theory presented in this book does not agree with me. Error theory argues that morality is systemically erroneous... what do we do with moral thought and discourse then? If Mackie thinks supervenient relationships are flawed, exactly how do ordinary natural facts determine the moral facts? He argues for conservationism (rather than eliminativism) because he recognizes the instrinsic vlue of morality, although I don't see that working. I agree more with the ideas of David Boonin, David Enoch, and Michael Huemer.

Mackie's errors about value talk have spread a lot of confusion, in particular his flawed subjective/objective distinction.

Think first of an example of non moral goodness. A good chair is such as to serve ends which are widely shared and revealed by the word ‘chair’ itself. To ask if the goodness of a given chair is "subjective" or "objective" is to start out in the wrong way. On the one hand, it is not an entirely subjective issue whether this is a good chair, because once the ends in view are fixed it is a matter of observable fact whether or not this chair is adequate to satisfy them (whether it wobbles, is sturdy enough to support a person’s weight, etc.). On the other hand, the issue of goodness is not entirely objective either because the desires, needs and practices of chair users set the ends for which chairs are employed.

The subjective/objective distinction doesn't give two mutually exclusive or jointly exhaustive alternatives. It does not present a real dilemma at all. The case in morality is more complicated in part because we evaluate ourselves with respect to our own needs and objectives. But it doesn't help to ask whether a thing's sufficiency to meet our goals is a subjective or objective matter.

The first chapter of Mackie's Ethics: Inventing Right and Wrong is the locus classicus for error theories in contemporary meta-ethics. There he argues that ordinary moral discourse and thought involve an assumption that there are what he calls "objective values," and that this assumption is false. Consequently, ordinary moral thought and language are infected by an error that precludes any ordinary moral claims and thoughts from being true.
Mackie first argues for a cognitivist interpretation of moral language. In other words, he argues that ordinary moral claims purport to describe facts about the world. In particular, ordinary moral language and thought purport to describe facts about objective moral values. What are objective moral values? They have two defining characteristics: (i) mind-independent existence (think of how chairs, trees, people, and electrons exist), and (ii) "intrinsic and categorical prescriptivity": that is, they are such that the mere apprehension of them will motivate a person to act in a certain way. The former characteristic is the source of their objectivity; the latter is the source of their normativity.
But, he claims, we have good reason to think that no such things exist. Mackie's fundamental worry about these putative objective values is that these things are especially "queer," that they are unlike any other things we have good reason to think exist. As I understand Mackie, underlying his worries about the queerness of these putative entities is his perception of a tension in their nature. He appears to believe that the objectivity of these putative entities is in tension with their intrinsic and categorical action-guidingness. That is, it is unclear to Mackie how something that exists as a mind-independent part of reality could have the sort of influence on human behavior that these objective values are supposed to have. It is unclear how something could be both objective and normative. The things that scientists study and that we encounter in the everyday world simply don't have this sort of categorical action-guidingness built into them. So, given the naturalistic conception of the world that Mackie favors, we have good a posteriori reasons to doubt the existence of objective moral values.
But, if Mackie is correct about the nature of ordinary moral thought and language, this commits us to regarding ordinary moral thought and language as involving a very fundamental sort of error, an error of presupposing that objective moral values exist. Mackie then completes his error theory by providing an explanation of our tendency to make this error, to mistakenly suppose that ordinary moral thought and language involve our successfully coming to know about the sorts of things he claims don't exist.
Mackie's book doesn't end here, however. Indeed, this is only the first chapter, and Mackie goes on to cover a wide range of territory in normative ethics and meta-ethics, along with a few issues in metaphysics (the existence of God and freedom of the will) that have some bearing on moral issues. In fact, despite his worries about the objectivity of morality, Mackie goes on to defend a substitute for morality, one that looks quite a bit like a broadly consequentialist moral theory, and he even weighs in on several controversial moral issues that are still with us. In short, in a little over two hundred pages of exceptionally clear prose, Mackie covers just about everything of interest in moral philosophy.
This book is, of course, essential reading for anyone interested in meta-ethics. Understanding some of the material and its importance may require some background knowledge, but enough of the book is more generally accessible that it also constitutes a good wide-ranging introduction to issues in both meta-ethics and normative ethics for a person with some background in philosophy (and perhaps for the general reader). Furthermore, the book, while not a work of history, is sufficiently informed about the history of the issues it discusses to provide the reader with an entry into study of the history of the subject.
If you're especially interested in Mackie's meta-ethical views, you should attempt to track down a copy of Morality and Objectivity (Ted Honderich, ed.), as it includes interesting and important reactions to Mackie's views by some major names (John McDowell, Simon Blackburn, R. M. Hare, Bernard Williams, et al.) in moral philosophy.