In a previous entry, I explicated W.D. Ross’s argument that “morally right action” does not mean the same thing as “morally good action”, with special attention given to the premise that was the most essential to his argument: it is the not the case that it is obligatory to act from a good motive. His intention was to conclude that because a good action is not necessarily the right action to do, if there are multiple good actions to choose from in a situation it doesn’t follow that one is morally obligated to do all of those actions. Rather, there is one right action for the situation, as determined by the strength of the reasons for doing that action given the competing duties one has.
So we can see that Ross’s positive argument for his duty-based pluralism depends on the success of his argument that a good action is not the right action, and with it, the premise that it is not obligatory to act from a good will. In the previous post, I suggested that one of two arguments that Ross gave for why it is not obligatory to act from a good will succeeded. That argument was that if one is obligated to act from the motive of duty (Ross openly assumed that the motive of duty is the most fundamental of morally good motives, such that if one is not obligated to act from the motive of duty then one is not obligated to act from any other good motive) then one must have the motive of duty in order to be motivated to have the motive of duty, but one must have the motive of duty in order to be motivated to have the motive of duty in order to be motivated to have the motive of duty, and so on, ad infinitum. In other words, our having the motive of duty never explains why we are obligated to do that action, we will always be one motive short in explaining why that act is required because all we can ever explain is the obligation to be motivated to do the act, not the obligation to do the act.
I think it is a decent, and original, argument, so it should come as no surprise that throughout The Right and the Good Ross quickly dismisses views that contradict that premise. One of the more interesting uses of the premise occurs when Ross rejects the notion that we are obligated to do the action that we think is our prima facie given partial knowledge of the facts. Ross thinks that our actual duty is the duty we would know to be our actual duty when given full knowledge of the facts. Given that Ross is defending a robust account of moral realism this is hardly unexpected – Ross is noting that the moral facts of the situation are what they are, and one’s duty is determined by all of those facts, not a partial set of facts, which could be arbitrary and lead to erroneous conclusions. Let us take a look now at how the notion that we are not obligated to act from a good will leads Ross to conclude that partial knowledge cannot provide actual (the strongest prima facie) moral duties.
For Ross, our duty cannot be determined by anything short of all the facts of the situation, because if a duty was constituted by the partial facts of the situation, then it is possible that one think action A is their duty, but when more facts become available they see that action B was there duty. Ross thinks that we must say that it was not wrong to do act A given that the available information was what it was. What this reveals, for Ross, is that partial information tells us that action A is morally good, and extrapolates to the claim that partial knowledge tells us only whether actions are morally good, and cannot tell us whether actions are right, in the sense of being one’s duty to do them. Furthermore, Ross argues, the only way we might understand a person’s action as morally good if we do not have all the facts of the situation is by identifying the motive for the action as a good motive. If we only have partial knowledge of a situation then we may only say that the action is good, and only because without having all the facts the sole thing that can make an act good is a good motive for the act. Again we see Ross’s objection to the idea of the right action being the one done for a good motive as he has argued here that the good motive in a situation with partial knowledge only makes the action morally good, not morally right.
The argument is fascinating in Ross’s continuing barrage on monistic deontology (i.e. Kantianism), which I think is valuable in itself. However, Ross leaves unexplored the interesting questions this argument raises for the epistemological theories supporting moral realism. That is — while this argument defeats ideal observer metaethical views (that the right action is the one that an agent with a certain amount of a certain type of knowledge would choose), it seems the truly pressing question posed by this argument is whether, if we for the most part only ever have partial information of the situation, do we ever perform the morally right action? If not, can we say that we have any moral knowledge?
These questions, I think, are some of the most perplexing issues that the moral realist, and particularly non-naturalists, face. I tend to think that if moral realism is true then moral truths will be of a particular sort: abstract and general propositions that we can see to be true a priori. For example, it is morally better than the alternative to cause a person pleasure than to cause them pain. I think this is self-evidently true, and we do not need to cause a certain number people of pain and a certain number of people pleasure in order to conclude that it is better to cause pleasure than pain. Rather, just in understanding what ‘pain’ and ‘pleasure’ are we understand that it is better to cause ‘pleasure’ than ‘pain’, in the same way that understanding what ‘the square root of 64’ is and ‘8’ is leads us to understand that the square root of 64 is 8. It might take time to see these identity relation as necessarily true, but once we see that it is, we see that it has to be true, in such a way that upon realizing the proposition’s truth we recognize the propositions self-evidence.
These very general and abstract a priori truths can be useful in guiding our actions, but when they compete and conflict it does not seem that there is a quick way to resolve which evaluative proposition is more commanding in which action to take. If we had full information for the situation we might more easily know the answer to this problem, a problem that Ross acknowledges for his pluralistic view, but I think it also seems true that even if we knew all the facts of the situation we still might not be sure which evaluative proposition is most commanding in the situation. This is a version of Hume’s “is-ought gap”, and there is some cogency to it, but for now I only want to tend to the issue posed by partial knowledge, not full knowledge and the is-ought gap.
I think it is uncontroversial to claim that we only ever have partial knowledge of the moral terrain we are in. To have full knowledge of the situation seems to entail being able to see long distances into the future, and into the past (insofar as the past relates to issues of restitution and deservingness in our moral acts). But if this is so, then, according to Ross’s argument, we might only ever do morally good acts and never the right action. How then do we explain our thinking that persons frequently do the right thing, even though they did not have full information of the situation? Perhaps we are incorrect in thinking that they did the morally right thing when they only performed a morally good action, but this seems a difficult bullet to bite, as Ross himself has noted that the data of metaethics are our moral intuitions, and those intuitions are against him here. So it seems we should not reject our good moral sensibility as unreliable.
It might be that persons are actually just frequently lucky; given their partial knowledge they think act A is the right action to do, and after the fact they receive more information, close to complete information, let us conjecture, whereupon they find out that act A really was the right action for the situation. I think this actually fits the phenomenological experience of our moral psychology – we frequently see more facts of the situation once we complete the action, and we realize that, though our action was morally good, there was a better action, which was the right action. Thus, it seems that in those instances where we succeed and perform not just a good action, but also the right action for the situation, we get lucky. I think this is a logical consequence of positing a robust moral realism though not a bane to it, after all, it seems plausible that if there are moral truths that exist independently of what we have to say about them then it is possible that we lack epistemic access to some moral truths given our only ever partial knowledge of a situation.