David Chalmers wrote a paper discussing what makes a verbal dispute and also how we can go about unmaking one. The basic method is to taboo the words you’re having the disagreement about and see if the conflict dissolves and or instead you find a substantive disagreement. Toward the middle he describes a difficult case:
Say that a Kantian and a consequentialist disagree about when an act is right. The Kantian holds that an act is right iff it has a certain deontological property, while the consequentialist holds that an act is right iff it has a certain consequence property. We can apply the usual method to determine whether this is a verbal dispute, barring ‘right’ and introducing ‘right1’ and ‘right2’. Is there a residual disagreement? We might find a dispute over ‘People always ought to perform right1
acts’ and the like. But ‘ought’ (in the relevant sense) is more or less cognate with ‘right’, so this is not much progress. In any case, we can bar ‘ought’ and introduce ‘ought1’ and ‘ought2’, which are stipulated to apply to acts with the deontological and the consequence property respectively. Is there a residual disagreement now?
As we proceed, the disagreement gets harder and harder to state. It is plausible that once all moral terms are gone, no disagreement can be stated. We might agree on all the nonmoral properties of the relevant actions, but still disagree on whether it is right.
Now that’s worth a try! I’m going to try tabooing morality from these moral theories:
1. An act is right because of its deontological properties.
2. An act is right because of its consequence properties.
3. An act is right because of its virtue properties.
4. An act is right because of its divine properties.
Every moral term must go: right, wrong, moral, good, bad, ought, should, must, proper, virtuous, higher, lower, etc. Here are the alternative sentences I came up with:
1. A sufficiently reflective person will develop a category of concepts that is important to them because the person feels that internal consistency with knowledge of the concepts requires the person to act in certain ways and not in others. Whether an act is in the category is best explained in terms of universal rules.
2. Society developed a category of concepts that is important to it because the concepts are used to guide people’s behavior so that society can reliably achieve the diverse ends it desires. Whether an act is in the category is best explained in terms of the consequences it has.
3. Society developed a category of concepts that is important to it on account of the concepts describing properties that reliably make people satisfied with their own and each others’ conduct. Whether an act is in the category is best explained in terms of personal habits and traits desired by many members of society.
4-(Atheist). Societies developed diverse stories about characters who personified things important to the societies, and then the societies figured out as best they could how such idealized persons would act in different situations, how the idealized persons would prefer ordinary people to act, and how the idealized persons would reward or punish ordinary people who did or did not behave in the preferred way. The rewards and punishments features in the stories were effective at achieving social order in primitive societies and so favored by cultural evolution. Whether an act achieves social order is best explained in terms of the characteristics attributed to the idealized persons.
4-(Theist). The previous version of 4 describes all societies except one; in that one special society, one or more actual ideal persons revealed to ordinary people the rules by which they could receive rewards or punishments, and these revelations were important to ordinary people because they wanted to get rewards and avoid punishments. Whether an act is permitted or forbidden by the revelation is best explained in terms of the characteristics of the ideal person(s) who gave the revelation.
Given those nonmoral moralities, 1, 2, and 3 all look very reasonable. I reject both versions of 4.
The first three theories, with their moral content eliminated, reveal very different substantive structure. It’s exceptionally interesting, then, that all three manage to reproduce so much of our conventional moral intuitions. They all each fail some of our moral intuitions, of course, which either shows that our moral intuitions are in error or the theories need modification. One could imagine that, in the most convenient case, all three theories could be successfully modified to always produce the same results, and that they would then be freely interchangeable on a whim, as distinct but isomorphic models of the same moral axioms. In the least convenient case, the three theories would be irreconcilable, all three would be equally important to us, and all three would have an equal claim to being called “morality”. The worst case is not particularly troublesome, though! We could simply snap the three theories together like LEGOs and say that we are more satisfied with our own and each other’s conduct when it displays habits and traits many of us desire, is consistent regarding the principles involved, and has widely desired consequences. Or that rational people would feel that acting consistently with their principles includes development of the habits and traits, and achievement of the consequences, that they desire. Or that the diverse ends desired by society include satisfaction with each others’ conduct, acquisition of certain traits and habits, and the knowledge that we have acted consistently with our principles.