Friends with the good: moral relativism and moral progress

The Philosophical Quarterly Vol. 0, No. 0 2023 ISSN 0031-8094 https://doi.org/10.1093/pq/pqad101 FRIENDS WITH THE GOOD: MORAL RELATIVISM AND MORAL PROGRESS By Eduardo Pérez-Navarro The aim of this paper is to defend moral relativism from the accusation that it would make it irrational to classify past changes in public opinion as instances of moral progress, for they would constitute an improvement only from our current point of view. The argument is this. For our assessment of a change in public opinion as an instance of moral progress to be rational, we need to take the moral claims made before the change to be false simpliciter while being open to the possibility that we ourselves change our minds at some point. These two things can be made compatible if we construe moral relativism as taking the truth of moral claims to be relative to the context of assessment. Thus understood, moral relativism is in fact the only view that makes room for talk of moral progress, as the rest of candidate positions make it irrational. Keywords: moral relativism, moral progress, assessment-sensitivity, moral contextualism, moral objectivism. I. INTRODUCTION Back in the 19th century, a majority of the population in the USA thought chattel slavery was not inherently wrong. Today, this opinion is fairly residual. It seems natural to say that public opinion 1 on chattel slavery has undergone moral progress in the USA. This seems quite a trivial fact—the fact that we talk about moral progress and, apparently, are not irrational in doing so. This simple fact, however, has been taken to be enough to discard some views in metaethics. In particular, talk of moral progress has been used to argue against moral relativism (see, for instance, Prinz 2007 : 289; Kitcher 2011 : 175, 2021 : 17; 1 Here, I will use the term ‘public opinion’ to refer to the opinion held by the majority of a given population. I admit this is quite a broad definition of the term. A narrower definition, which I will not pursue here, would require members of the population to know that other members hold this opinion, so that it can be presupposed in public exchanges (see Stalnaker 1973 ). ©The Author(s) 2023. Published by Oxford University Press on behalf of The Scots Philosophical Association and the University of St Andrews. This is an Open Access article distributed under the terms of the Creative Commons Attribution License ( https://creativecommons.org/licenses/by/4.0/ ), which permits unrestricted reuse, distribution, and reproduction in any medium, provided the original work is properly cited. Downloaded from https://academic.oup.com/pq/advance-article/doi/10.1093/pq/pqad101/7325257 by Irene Ruiz user on 30 October 2023 2 EDUARDO PÉREZ-NAVARRO Coliva & Moruzzi 2012 : 52; Rachels & Rachels 2019 : 20; Wreen 2019 : 363–4; Sauer et al. 2021 : 5). The argument can be schematically put like this: (P1) It follows from moral relativism that, when it comes to moral matters, there is no objective point of view. (P2) For our assessment of a change in public opinion as an instance of moral progress to be rational, we need to take things as they are after the change to be morally better from an objective point of view. (C) Thus, it is irrational for moral relativists to assess any change in public opinion as an instance of moral progress. The aim of this paper is to debunk this argument and prove that, in fact, moral relativism is the only view that makes room for talk of moral progress, as it is the rest of positions that might be considered that make it irrational. To talk of moral progress, we need to be able to say of some things that they are good without any qualification. If we heed to Montesquieu’s ( 1899/1964 : 947) advice that ‘the best is the mortal enemy of the good’, though, views that are only satisfied with the best will preclude us from doing so. This means that only moral relativism can be friends with the good and, consequently, make talk of moral progress rational. The structure of this paper is as follows. In Section II , I offer definitions of both moral relativism and moral progress. In Section III , I show that (P2) above is based on the idea that moral relativists cannot take the moral claims made before a change in public opinion took place to be false simpliciter , and I argue that moral relativism construed along MacFarlane’s ( 2014 ) lines can do such a thing. To make the argument clearer, I combine it with the parallel argument that it is moral contextualists who lack this possibility. In Section IV , I show that not any way of deeming past claims false simpliciter will do, as we need to do so while leaving the door open to eventually changing our own minds. The standard version of moral objectivism , on which Section V focuses, cannot do this even if it makes us deem past claims false simpliciter . Unfortunately, moreover, it will be of no help to shift to another version of moral objectivism, as, if we do so, the view will not even outperform moral contextualism. This leaves moral relativism as the only view that makes sense of our talk of moral progress. II. MORAL RELATIVISM AND MORAL PROGRESS The way I will use the term ‘relativism’ in this paper, it refers to a theory along the lines of the one developed by MacFarlane ( 2014 ). MacFarlane applies his theoretical scaffolding to several kinds of utterances, such as knowledge attributions and utterances featuring predicates of personal taste. Moral relativism, Downloaded from https://academic.oup.com/pq/advance-article/doi/10.1093/pq/pqad101/7325257 by Irene Ruiz user on 30 October 2023 FRIENDS WITH THE GOOD 3 although not taken into account by MacFarlane himself, 2 should be suitable for implementation along similar lines. Utterances of sentences in an area of language susceptible of a relativist treatment are taken by MacFarlane to express propositions whose truth-value depends on the context of assessment , 3 which is simply the context from which we ask whether a certain proposition is true or false. As MacFarlane ( 2014 : 60–1) says, if we are comfortable with the idea of a context of use (that in which we use a sentence), we should also be comfortable with the idea of a context of assessment (that from which we assess it). Moral relativism, in the sense I will favour throughout this paper, will then be the view that moral claims express propositions whose truth-value is relative to a moral standard determined by the context of assessment. For instance, by saying (1), Alice would be expressing a proposition whose truth-value is relative to a moral standard: (1) Chattel slavery isn’t morally wrong. If Beth wonders whether Alice has said something true, relativism has it that she should answer in the affirmative if and only if chattel slavery is morally right relative to the moral standard that is relevant in Beth’s context. 4 Moral relativism should be distinguished from moral contextualism, which would be the view that moral claims express propositions whose truth-value is relative to the context of use 5 (see, for instance, Harman 1975/2013 ; Dreier 1990 ). 6 For instance, contextualism would have it that Beth should answer in the affirmative to the question whether Alice has said something true in uttering (1) if and only if chattel slavery is morally right relative to the moral standard that is relevant in Alice’s context. As advanced, some opposition to moral relativism has its origin in the fact that we talk of moral progress. Sometimes, a moral claim that was commonly taken to be true turns to be commonly taken to be false. 7 For instance, while 2 MacFarlane proposes to use relativism to explain the behaviour of claims featuring ‘ought’ (Kolodny and MacFarlane 2010 ; MacFarlane 2014 : ch. 11), but this is a prudential, rather than a moral, ‘ought’. 3 This makes the version of relativism I am talking about a variety of nonindexical relativism, as the dependence of the truth-value associated with an utterance on the context of assessment is due to the dependence of the expressed proposition’s truth-value, and not its content, on that context. The latter option would be the one pursued by indexical relativists (see Pérez-Navarro 2022a ). 4 This need not be Beth’s moral standard; it can be some other moral standard made relevant by Beth’s context (see Pérez-Navarro 2021 ). 5 This makes the kind of contextualism I am talking about here a variety of nonindexical contextualism (see, for instance, MacFarlane 2014 : 88–90; cf. footnote 3). 6 Although Harman and Dreier call themselves relativists, see Pérez-Navarro ( 2022b : 60) for an argument that they are best seen as contextualists. 7 This sets the ground for characterising collective moral progress. Moral progress can be identified at the individual level too; however, I will not discuss this variety of the phenomenon in this paper. Downloaded from https://academic.oup.com/pq/advance-article/doi/10.1093/pq/pqad101/7325257 by Irene Ruiz user on 30 October 2023 4 EDUARDO PÉREZ-NAVARRO 200 years ago most people in the USA would have taken (1) to be true, public opinion in that country is now mainly convinced of that very claim’s falsity. Note that what is taken to be false is not only that chattel slavery, were it to occur today, would not be morally wrong, but also that it was not as it occurred 200 years ago. When we think about this change, many of us take it to be an instance of moral progress, not ‘just one damn thing after another’ (Kitcher 2011 : 138). Seeing the situation in this way involves at least ( i ) accepting that the change in question has occurred and ( ii ) assessing it as a transition to a morally better situation (Buchanan and Powell 2018 : 48). Apparently, we take both ( i ) and ( ii ) to hold when we say that there has been moral progress in public opinion on chattel slavery in the USA: We do so because we think that chattel slavery is badly seen where it used not to and that things are better that way. The fact I want to point to is simply that it does not seem irrational to say that there has been moral progress in a case like this. Two qualifications are in order. First, the characterisation of moral progress above is congenial to a pragmatist outlook in that it does not reply to the question what moral progress is, but to a question about the conditions under which we call certain changes in public opinion instances of moral progress. Secondly, I do not care whether we are right or wrong in doing so; all I care about is whether this is a rational move (cf. Kitcher 2021 : 39). In other words, I am not after a criterion that distinguishes moral progress from moral regress, of the kind on offer by both utilitarians (Singer 1979/1993 : 12–4, 1981/2011 : 96–124) and deontologists (Musschenga and Meynen 2017 ). I take one of the lessons of relativism, in fact, to be that no such criterion can be offered that allows us to tell, in an algorithmic way, one thing from the other. My characterisation of moral progress is minimal on purpose, so that it can cover a number of varieties of moral progress that have been distinguished in the literature, such as proper moralisation (Buchanan and Powell 2018 : 56– 7), proper demoralisation (Buchanan and Powell 2017 ), and moral progress understood in terms of ‘expanding the circle’ (Singer 1981/2011 ). 8 Saying that moral progress in any of these varieties has taken place amounts to saying that ( i ) a change of public opinion has taken place and ( ii ) things are morally better that way. I acknowledge, though, that some varieties of moral progress, such as the one that results in better moral concepts (see, for instance, Buchanan and Powell 2018 : 54–5), are not covered by my characterisation. III. CHANGES FOR THE BETTER In Section I , I presented an argument against moral relativism based on the rationality of talk of moral progress. In this section, I show that this is in fact 8 For the most comprehensive survey of varieties of moral progress I know of to this day, see Buchanan & Powell ( 2018 : 53–9). Downloaded from https://academic.oup.com/pq/advance-article/doi/10.1093/pq/pqad101/7325257 by Irene Ruiz user on 30 October 2023 FRIENDS WITH THE GOOD 5 an argument against moral contextualism, but it does not work against moral relativism. To see this, it is first important to note that, since we have chosen to understand moral claims as utterances, the most salient difference between moral relativism and moral contextualism as characterised in Section II lies in the possibility of deeming a moral claim true once and for all. In contextualism, we will be able to do so if the proposition it expresses is true relative to the context of use, which is intrinsically linked to the claim. In relativism, on the contrary, moral claims cannot be said to be true once and for all, for their truth depends on the context of assessment, and there is a potentially infinite number of contexts from which we might assess a single claim (MacFarlane 2014 : 61–2; see also Field 2009 : 273). While the contextualist would thus take the speaker’s context to determine once and for all which moral claims are true, then, the relativist denies that any point of view can play such a role. In other words, the relativist denies that there is an objective point of view when it comes to moral matters. Relativism understood in MacFarlane’s terms, but not contextualism, will thus make the first premise of the antirelativist argument true. But, while I accept (P1), this is not the case with (P2), to which I turn now. To dispute this premise, repeated below, let us first consider what lies behind it: (P2) For our assessment of a change in public opinion as an instance of moral progress to be rational, we need to take things as they are after the change to be morally better from an objective point of view. In what sense can we take things to be morally better after a change in public opinion has taken place? Following the focus on truth attribution that characterised the discussion in the previous section, I will say we do so whenever we take the change to go from assigning the wrong truth-value to a moral claim to assigning the right one to it. If we take the moral claim to be true, we will take a change from assigning falsity to it to assigning truth to be a change for the better. If we take the moral claim to be false, changes for the better will be those that go from assigning truth to it to assigning falsity. For instance, since we take (1) to be false, we take the change from assigning truth to it to assigning falsity to be a change for the better. However, note that we should not only take (1) to be false from our current point of view: When asked whether it is true or false, we should say it is false without any qualification. In other words, we should take it to be false simpliciter . Remember from the previous section that the focus here is not on the conditions under which there is moral progress, but on the conditions under which we say there is such a thing; correspondingly, falsity simpliciter is here understood as something we attribute to certain claims, and we do so when we say they are false without relativising the attribution to any parameter. I would say the focus on falsity (or truth) simpliciter is what lies behind Moore’s (1903/ 1993 : §11) ‘What’s at issue?’ argument: When wondering whether Downloaded from https://academic.oup.com/pq/advance-article/doi/10.1093/pq/pqad101/7325257 by Irene Ruiz user on 30 October 2023 6 EDUARDO PÉREZ-NAVARRO something is right or wrong, we do not wonder whether it is so according to this or that, but whether it is right or wrong, period (see also Gibbard 2003 : 23–9). Taking past claims to be false only from our current point of view would be compatible with a criterion for truth simpliciter that were met by these claims as far as they were true relative to the context in which they were made. If this were the case, however, we would not be able to talk about progress, but just about ‘one damn thing after another’, as Kitcher says. Taking (1) to be false simpliciter , of course, implies taking (2) to be true simpliciter : (2) Chattel slavery is morally wrong. In general, the move from assigning truth to a false claim to assigning falsity to it will coincide with the move from wrongly assigning falsity to the corresponding negated claim to assigning truth to it. I will thus call the required move a move from falsity to truth, paying homage to the intuition that, when we say that moral progress has taken place, we are saying that a true moral belief is now widely shared where a false one used to. If this is what it means to take things to be morally better after a change in public opinion has taken place, the second premise of the antirelativist argument can be rendered as saying that the moral relativist cannot take claims that were previously taken to hold to be false simpliciter . For instance, (1) will be true simpliciter for the relativist, the antirelativist says, as far as it is true with respect to the moral standard that held back when the claim was made. Of course, (1) will still be true with respect to that standard even if it is no longer accepted. So the change at issue cannot be said to go from falsity to truth, as (1), which should now be taken to be false simpliciter , is still taken to be true. Note, however, that I have characterised moral relativism as the view that moral claims express propositions whose truth-value depends on a moral standard supplied by the context of assessment (see again MacFarlane 2014 ). Thus understood, relativism will of course make us take (1) to be false simpliciter , as it is false as assessed from our context. The moral relativist can thus avail herself of move ( ii ), because she can say that moral claims licensed by moral standards that no longer hold are false simpliciter . 9 In making such move, however, she must take into account whether things are morally better now relative to her context, not the one that held before the change took place. The consequence is that moral relativists should be relativists about moral progress too; that is, they should say that claims like 9 My reply to the antirelativist argument does not thus rely on the notion of trumping (Coliva and Moruzzi 2012 ). The basic idea behind any strategy that relies on such notion is that our moral standard ‘trumps’ any other one, so, once we have adopted a certain standard, moral claims are assigned relative to any other one the truth-value ours would assign to them. For instance, if our moral standard makes chattel slavery wrong, then, relative to that standard, it is also true that the moral standard that previously held makes chattel slavery wrong. This seems highly implausible to me, as does to Coliva & Moruzzi ( 2012 : 54–5). Fortunately, I do not need to undertake any such commitment. Downloaded from https://academic.oup.com/pq/advance-article/doi/10.1093/pq/pqad101/7325257 by Irene Ruiz user on 30 October 2023 FRIENDS WITH THE GOOD 7 (3) When it comes to chattel slavery, there has been moral progress in the USA. are assessment-sensitive. This means that they cannot be said to be true or false from an objective point of view; instead, each time they are so declared the assessment is made from a particular standpoint. This is just as it should be, as classifications of changes in public opinion as instances of moral progress are as normative as ground-level moral claims are. When we discuss whether something amounts to moral progress, we are at the end of the day discussing what is right and what is wrong. By taking classifications of changes in public opinion as instances of moral progress to be as assessment-sensitive as ground-level moral claims, I distance myself from authors who, although sympathetic to moral relativism, think claims identifying moral progress can only rely on extramoral considerations (Prinz 2007 :289–302; Kitcher 2021 ). The idea is that, when we change our moral opinions in an apparently progressive way, this appearance of progress can only be justified by reference to non-moral values, ‘such as social cohesion, welfare, and well-being’ (Prinz 2007 : 301). This is so, it is said, because, if reference were made to moral values, the ones we embrace would trivially render as instances of progress those changes that accord with them and deny that title to those that do not (see next section). I agree with Prinz’s point that there is a number of standards that might be taken into account when assessing the truth of a claim identifying moral progress (Prinz 2007 : 291), and that some of these standards might not be moral ones. However, I would not go as far as to say that none of them is (Prinz 2007 : 292), as I think doing so would deprive assessments of moral progress of their normative character. I suspect it is moral contextualism that critics have in mind when they accuse moral relativism of making talk of progress irrational, as it is contextualism that makes us take into account the moral standard that held before the change in public opinion took place when assessing the truth of (1). Relative to this standard, the proposition expressed by (1) is true. Since a moral claim will be true simpliciter if and only if it expresses a proposition that is true relative to the moral standard that matters, we should then take (1) to be true simpliciter . The consequence is that the change in public opinion that goes from rendering (1) true to rendering it false can no longer be taken to be a move from falsity to truth, and thus the second requirement that must hold for us to say that such a change constitutes an instance of moral progress is not fulfilled. It is a separate question, of course, whether antirelativists are wrong in construing relativism in the terms I have associated with contextualism. The argument above could as well be seen as supporting the conclusion that moral relativists, now taken to propose the kind of view I have called ‘moral contextualism’, should abandon it in favour of the one advanced here. The question how to understand the argument ultimately comes down to which Downloaded from https://academic.oup.com/pq/advance-article/doi/10.1093/pq/pqad101/7325257 by Irene Ruiz user on 30 October 2023 8 EDUARDO PÉREZ-NAVARRO characterisation better captures the actual commitments of authors who have called themselves relativists, but I will not press this issue here. 10 IV. ONGOING PROGRESS The view I have called ‘moral contextualism’ does not reject the idea of an objective point of view, but it makes talk of moral progress irrational. The view for which I have kept the label ‘moral relativism’ captures the intuition that moral relativism makes no room for an objective point of view, but, pace the antirelativist argument, it apparently accommodates talk of moral progress. If I say ‘apparently’, it is because not any way of rendering (1) false simpliciter should count. Moral relativism could do so in an inappropriate way, thus resulting unable to account for moral progress after all. In this section, I explore what is needed to say that the change in truth-value is appropriately accounted for. We will see that moral relativism still accounts for the change in this sense, but in the next section I will show that other views that apparently account for it do so inappropriately. We have seen that talking of moral progress requires taking the moral claims that were made before the change at issue took place to be false simpliciter . But what we would declare to have taken place would be merely apparent progress if this were done in a way that made no room for future talk of progress to be rational too. If my current claims identifying progress were rational in a way in which my future ones could never be so, the claim that progress has taken place would be trivially true, and this is not what we are after (see Prinz 2007 : 289). So we also need to make it rational for us to, at some point in the future, assess a further change in public opinion as an instance of moral progress. This, in turn, requires two things. The first is for public opinion to change again, that is, for condition ( i ) for talking about progress to be eventually met. The second is that we end up taking the things now taken to be true to be false simpliciter , thus fulfilling condition ( ii ) for talk of progress. Note, however, that the picture for the future is a little more complicated than the one for the present. Current claims identifying progress do not necessarily involve a change of one’s mind: one of the cases in which we call a change in public opinion an instance of moral progress is that in which public opinion evolves to coincide with what we have always thought. Future claims of progress that took place after one made now, though, would involve such a change of mind, as if we say current public opinion is the result of moral progress it is in part because we share that opinion. So, along with the usual ( i ) and ( ii ) for public opinion, future claims identifying progress will require us 10 For some textual evidence that the positions of many actual relativists should be construed along MacFarlane’s lines, though, see Pérez-Navarro ( 2022b : 59–60). Downloaded from https://academic.oup.com/pq/advance-article/doi/10.1093/pq/pqad101/7325257 by Irene Ruiz user on 30 October 2023 FRIENDS WITH THE GOOD 9 to ( i S ) accept that we have changed our minds and ( ii S ) take such a change of mind to have been for the better inasmuch as our past moral claims are false simpliciter . Can moral relativism and moral contextualism make sense of move ( ii S )? Suppose that (1), repeated below, used to be thought true at t 1 , but now, at t 2 , it no longer is: (1) Chattel slavery isn’t morally wrong. Instead, it is (2), repeated below, that is commonly thought true at t 2 even if at t 1 it was commonly taken to be false: (2) Chattel slavery is morally wrong. Since we share public opinion on chattel slavery at t 2 , we take (1) to be false and (2) to be true simpliciter . It would be moreover natural that, since we take (2) to be true simpliciter , we ourselves said such a thing at some point. But we should be open to the possibility that, at some point in the future t 3 , we end up taking our own claim (2) to be false simpliciter . Like with ( ii ), moral relativism gets us this, but moral contextualism does not. If our moral standard indeed changes, relativism has it that we will turn to taking (2) to be false simpliciter , as it will be so relative to the standard that will hold at t 3 . According to contextualism, though, the standard that will be decisive in assigning (2) a truth-value simpliciter is the standard that holds at t 2 , so we will take it to be true simpliciter . 11 V. MORAL OBJECTIVISM Le mieux est le mortel ennemi du bien. (Montesquieu 1899/1964 : 947) By allowing us to accept that both ( i ) public opinion and ( i S ) our own have changed and take the changes in both ( ii ) public opinion and ( ii S ) our own to have been for the better, moral relativism makes rational both our current classifications of changes in public opinion as instances of moral progress and those carried out by our future selves. A view’s fate in one case, however, need not be linked to its fate in the other. To show this, I will now discuss a view that makes moves ( i ), ( ii ), and ( ii S ) possible but leaves the change of mind involved in ( i S ) utterly mysterious. This is moral objectivism, at least in one of its versions. 11 Of course, since the standard the moral contextualist thinks should be taken into account when asking about the truth of (1) is the one that held at t 1 , at t 3 she will take (1) to be true simpliciter , as should be the case. But this is just one of the two times a day moral contextualism’s broken clock is right. Downloaded from https://academic.oup.com/pq/advance-article/doi/10.1093/pq/pqad101/7325257 by Irene Ruiz user on 30 October 2023