(Continued from Part II)
If all that is required to count as an error theory about evaluative discourse is that atomic value sentences [1] systematically contain or are infected by some quantum of error, Nietzsche clearly is an error theorist about value.
But is that a useful construal of error theory? If so, then we might as well all be "error theorists" about scientific discourse. It is well known, for instance, that the two fundamental theories in physics--quantum mechanics and relativity--are mutually inconsistent. Moreover, it is the very discovery of error that drives the entire scientific project. No anomalies ==> no scientific "progress." So the practice of science itself presupposes and anticipates prospective anomalies. But of course any such anomaly once discovered will render a scientific theory "false" to at least some degree. Hence, our "error theory" would apply. Clearly, that's an error theory that disproves too much.
To see how this problem arises in Hussain's paper, we need to look at how he defines the term:
As we've seen, to be "systematically false" is not necessarily to be false to a relevant degree. What's happening here is that Hussain fixes on the semantic and substantive dimensions of error theory with no eye to the ontological considerations that underlie the substantive claim. This exclusion in turn leads to a lessening of the degree of error that is otherwise logically required by any error theory in specie, namely, 100 percent.
A standard illustrative example for how ontology figures in grounding this logical requirement for complete rather than partial error is the case of phlogiston. Any proposition of the form 'phlogiston is phi' will be completely false. The reason it is completely false, and necessarily completely false, is that there is nothing in the world that satisfies the essential conceptual content of 'phlogiston'. That is, 'phlogiston' fails to refer to anything real. So there is no truth, not even partial truth, to be had about the actual properties of phlogiston (save to say that there are none). Thus, we are all "error theorists" about phlogiston talk.
The very same ontological considerations motivate error theories about moral or evaluative discourse, and so entail the same sort of logical requirement for 100 percent falsity. Thus, if value talk can contain even a quantum of correctness, error theory does not apply.
So again, if I'm right, Nietzsche is clearly not an error theorist about value. Though he does not propound any theory of value (and why this is so is a topic for an upcoming post), he at least seems to believe that some "orders of rank" really are superior to others, and that through various epistemological reforms we can more reliably come by ever better, ever more defensible revaluations. In short, he thinks that talk about value--or at least his talk about value--gets at something real in the world.
__________________
NOTES
1. Appropriating Richard Joyce's terminology, I'll take "atomic value sentences" to be those that "imply or presuppose the instantiation of [an evaluative] property." See Joyce's "Moral Anti-Realism", The Stanford Encyclopedia of Philosophy (Fall 2008 Edition), Edward N. Zalta (ed.), URL = <http://plato.stanford.edu/archives/fall2008/entries/moral-anti-realism/>.
