Comments on “Contextualismo e relativismo na ética”: Relativism in ethics

André Fuhrmann¹

Reference of the commented article: MENDONÇA, W. Contextualismo e relativismo na ética. Trans/Form/Ação: revista de filosofia da Unesp, v. 46, n. esp. “Filosofia Autoral Brasileira”, p. 627-668, 2023.

It is nearly a commonplace that the truth of moral assessments is relative. Only nearly so, because there is the respectable view that moral assessments do not belong to the kinds of speech acts that can be evaluated for truth or falsity. But there have been attempts – notably by Simon Blackburn and his followers – to make even that view compatible with the commonplace. What makes the statement a commonplace is, of course, a hidden quantification: The truth of moral assessments is relative in some sense. As soon as we start spelling out possible senses of “relative”, we obtain a wide range of claims. Some of them are obviously true. For a start, whether “Polygamy is wrong” is true or false, depends on what we mean by “polygamy”. Further, once meanings are fixed, it also depends on what the consequences of polygamy are (at least for consequentialists in the widest possible sense). There certainly are worlds in which polygamy only spells bliss and there are worlds in which polygamy is desastrous in every single case. So, whether polygamy is actually wrong depends on how things are. These are commonly understood dependencies.

As everyone else, philosophers procure arguments that have commonplaces as premisses but not as conclusions (despite occasional evidence to the contrary). If we took the thesis that truth in morals is relative as supported by the kind of dependencies mentioned above, then that thesis would be just as interesting as, say, the thesis that the reference of the first person pronoun “I” is relative to who is speaking – not the kind of thesis in need of an argument. We need an interestingly contentious and reasonably clear sense in which the truth of moral assessments is relative.

A paradigm case of substantial relativism is its most vulgar version: In morals “anything goes”. According to this view, morals is even worse than taste. Taste has at least a physiological basis that makes people “unanimously” spit when drinking methylated spirits. But in morals there are no arguments that could command unanimous consent; one is at liberty to take any stance whatsoever. The liberty is only more or less tempered by custom and compassion. So moral assessment is tenuously relative to social groups and genetically programmed sentiments. There is no sense in the notion that your group or genes may misguide you.

Repugnant as vulgar relativism may be, there is a grain of truth in it. Moral judgements essentially depend on a moral standard. But how exactly does such a standard come into play when asking about the truth of a moral claim? What is the nature of the dependency? It turns out that there are theoretical options here with clearly differing consequences. This is the topic of Wilson Mendonça’s contribution to this volume. How does a moral standard enter into an evaluation for truth? In particular, is there a sustainable answer to the question which would support the claim that some moral judgement is true relative to one standard and false relative to another? If there is, then the relativist claim can be given an interesting and intelligible reading.

When asking how the truth assessment of a statement can depend on a parameter, such as a moral standard, we are asking at bottom a formal question. For, in principle nothing depends on the particular content of the statement nor on the nature of the parameter. Once this is realized, it is natural to turn to extant theories of semantic dependency in other domains. Temporal discourse is the best investigated domain in this respect. Wilson clearly lays out the relevant varieties of dependency and transfers them to the moral domain. There is no need to repeat details here and no space to comment on them. Let me instead point out what I take to be the fundamental problem. I do not claim originality here, save perhaps in the way of presenting it. Anyway, the problem seems to me important enough to be recalled and firmly kept in mind. (Of course, Wilson is well aware of it and I see it hinted at in his text.)

The problem is a double dilemma. First, we must choose between global und local relativism. Global relativism claims that the truth of all judgements – irrespective of subject matter – is relative to some variable parameter P. Local relativism claims that relativity only affects certain classes of judgements – about matters of taste, beauty, or right-or-wrong and the like. If we opt for global relativism, then we are postulating a plethora of extensionally distinct predicates for sentences – P₀-true, P₁-true, and so on. Now, even though ‘truth’ may be a problematic notion, we do know enough about ‘truth’ so as to rule out some sentence predicates as candidates for synonymy with ‘is-true’. (If we did not, we wouldn’t even know what we are aiming at when theorizing about truth.) Without rehearsing well-known arguments let me simply record that it is today widely agreed among professional philosophers that none of the P-true predicates comes even close to reaching the qualification round for being a truth predicate. (I am thinking of the standard objections to epistemic theories of truth, such as pragmatism, verificationism, coherentism, or discourse theories.)

Thus we should go for the second horn of the first dilemma, local relativism. Choose some domain D of judgements for which relativity of truth is claimed, say, morals. Again, let P₀, P₁... be the parameters of truth so that within D all statements are judged P₀, P₁...-true or -false, as the case may be. Now we face a new dilemma. Suppose, first, that we treat each of these distinct predicates as primitive. Then, by the same considerations marshalled against global relativism, none of them conforms to our conjoint truisms about ‘truth’; they may be more or less remote surrogates for, but they are not even close to being truth predicates. We get a doctrine of some kind of relativity, but not a relativity of truth. — So we proceed to the second horn. Suppose that we take each P-true predicate as a composite of absolute truth and a parameter P. There is, of course, a challenge in the “taking-as”. Some sort of reduction is required: “A is P-true” needs to be resolved into a clause in which “P ” and “true” occur. But note, first, that every way of meeting the challenge undermines the thesis that in D truth is relative. For, since we offer a reductive analysis of the relative notion, the relativity is revealed as a mere surface phenomenon. We get a kind of surface-relativism that functions best as an error theory, explaining why we are inclined to take truth to be relative in D. Note, second, that even if we know how to relate P-truth to plain truth, we may still face the objection that P-truth differs significantly from plain truth. The same arguments that apply to global relativism may be used to question that P-truth is a kind of truth at all.

Whether the second objection applies, depends on the way in which P- truth is reduced to a composite of the parameter P and plain truth. We cannot rule out antecedently that the reduction may be such that P-truth inherits from its truth-component the associated truisms. But it is unlikely. For, relativists take it as an advantage of their view that they give some sense to the notion that Manoel may be right, in his way, to pronounce something true that Maria, in her way, judges to be false. The notion is, of course, alright as far as it goes. But true relativists offer the explanation that this sort of relativity attaches to the notion of truth, not to that of pronouncing or judging. If so, then their explanation is at odds with hard to dispute hallmarks of truth, such as the possibility that Manoel believes something to be true that Maria believes to be false – in a univocal sense of ‘true’ and ‘false’.

Now back to the interpretations of “dependency on a moral standard” as described in Wilson’s contribution. If they bypass the problem, then it must be dubious whether they can subserve the relativist claim. If they confront the problem, then the reader will have to judge how well they can respond to it.

Reference

MENDONÇA, W. Contextualismo e relativismo na ética. Trans/Form/Ação: revista de filosofia da Unesp, v. 46, n. esp. “Filosofia Autoral Brasileira”, p. 627-668, 2023.

Received: 12/04/2023

Accept: 18/04/2023