Expressivism and Contrary-forming Negation

A well known challenge faced by any version of metaethical expressivism is to provide an adequate treatment of the so-called Frege-Geach problem. The task is to make sense of logically complex sentences like 'If I insulted the host then I ought to apologize', to do so by making sense of the states of mind such sentences would be used to express, and to preserve the validity of intuitively valid arguments, e.g., " If I insulted the host then I ought to apologize; so, since I did insult the host, I ought to apologize. " The overall Frege-Geach problem includes, as a special case, the problem of making sense of negative sentences like 'Murder is not wrong', making sense of the states of mind expressed by such sentences, and explaining why the states of mind respectively expressible by 'Murder is wrong' and 'Murder is not wrong' are logically inconsistent with one another. This special case raises certain specific difficulties of its own for expressivism, over and above the familiar difficulties raised by the generic We will call this particular package of difficulties the negation problem. The plan of this paper is as follows. In section 1 we describe the negation problem. In section 2 we propose a solution that looks potentially available to various different versions of expressivism, including the version that we ourselves espouse (Horgan and Timmons 2006). In section 3 we argue that our solution is theoretically preferable to two alternative solutions currently on offer in the literature on this topic. In the appendix we briefly describe how to modify and improve our own version of expressivism (" cognitivist expressivism "), including its treatment of the generic Frege-Geach problem, in order to incorporate the treatment of the negation problem proposed in the present paper.