I consider myself a moral realist: I believe that ethical sentences express propositions that refer to objective features of the world, insofar as they represent those features accurately. That last bit is very important to this post. Furthermore, it is a pretty uncontroversial inference from this position that moral facts exist - facts that represent facets of objective morality.

I will be arguing that not only can moral facts be approximated - we have good reason to think that approximate moral facts exist.

I will now introduce the first part of my thesis - a heuristic for finding moral facts: we can analyze approximate moral facts in terms of the semantics of their constituent parts and compare the combinations of constituent parts to the outcomes arrived at by their application to yield a basis for combining said constituent parts into a more accurate approximation of moral facts through trial and error.

Essentially we would insert different combinations of semantic units into an ever-closer approximation based upon how we know they interact from trial and error, and then check the consistency/closeness of the approximation against its constituent parts.

To demonstrate an application of the heuristic: we know that combining the statement “it is wrong to steal on the sabbath” with “it is okay to steal on days that are not considered holy” yields the normative statement: “it is wrong to steal on the sabbath, but it is not wrong to steal on some days that aren’t the sabbath”. We need only check this law against the truth of its constituent parts to then decide whether or not it is a valid approximation. If it is consistent, and this one is, then we have a law governing when it is okay to steal - on some days other than the sabbath. Exactly which days it would be okay to steal on would require more calculations.

Why I think there are approximate moral facts is that, according to a moral realist, there are objective features of the world that the propositions contained in ethical sentences represent. If we ground ethical sentences in such features it follows that we can approximate a moral fact if the approximate moral fact is sufficiently close to the corresponding moral fact such that it functions no differently given a certain set of true ethical sentences; the approximate moral fact would also merely represent the features of the world that would make the corresponding moral fact true, without regard to whether or not it contradicts or represents other actual moral facts accurately.

This may be all well and good, you might think, but what does it actually tell us? Where is the benchmark? At the moment there is none, but if we can at least approach the meaning of some true ethical sentences it seems to me we can use the heuristic to approximate moral facts.