Explaining historical moral convergence: the empirical case against realist intuitionism
Over the course of human history there appears to have been a global shift in moral values towards a broadly ‘liberal’ orientation. Huemer (Philosophical Studies 173: 1983–2010, 2016) argues that this shift better accords with a realist than an antirealist metaethics: it is best explained by the discovery of mind-independent truths through intuition. In this article I argue, contra Huemer, that the historical data are better explained assuming the truth of moral antirealism. Realism does not fit the data as well as Huemer suggests, whereas antirealists have underappreciated resources to explain the relevant historical dynamics. These resources include an appeal to socialization, to technological and economical convergences, to lessons learned from history, to changes induced by consistency reasoning and to the social function of moral norms in overcoming some of the cooperation problems that globalizing societies face. I point out that the realist’s explanans has multiple shortcomings, that the antirealist’s explanans has several explanatory virtues, and conclude that the latter provides a superior account of the historical shift towards liberal values.
KeywordsMichael Huemer Moral progress Moral convergence Rationalist intuitionism Consistency reasoning
Moral realists maintain that moral truths are underwritten by mind-independent moral properties or facts and that these properties or facts are epistemically accessible to moral agents.1 It has often been argued that moral realists cannot make sense of the fact that between cultures and over the course of history there has been pervasive moral disagreement (e.g. Mackie 1977; Doris and Plakias 2008). But the empirical record suggests that there is also agreement on a substantial number of moral issues. Moreover, attitudes about many moral issues seem to have converged over time. If moral divergence provides evidence against realism, does moral convergence provide evidence in support of it?
Huemer (2016) has recently argued so. According to Huemer, the historical record exhibits a clear trend of moral convergence: over the course of centuries, many societies worldwide have been moving towards a set of broadly liberal values (I identify these in Sect. 3). This trend meshes well with the contention that moral inquiry is fuelled by a somewhat reliable process of tracking mind-independent moral truths through intuition, Huemer claims. If we have at least some epistemic access to mind-independent moral facts, then over the longue durée of human history our moral values should start to converge.
Huemer presents his argument as an inference to the best explanation, concluding that given the good fit that a realist account has with the historical data ‘it is reasonable to endorse moral realism, unless and until a better account appears’ (idem, p. 2008). The aim of the present article is to provide a better account. I argue that the relevant historical data, wedded with insights from cognitive science, moral psychology and evolutionary anthropology, are best explained by assuming some version of an antirealist metaethics—that is, by assuming that there are no mind-independent moral truths.2 Not only can antirealists overcome the challenges that Huemer raises for them; in fact, their account has several virtues which makes it superior, in terms of empirical explanation, to Huemer’s realist account.
The article proceeds as follows. In Sect. 2, I outline the realist position that Huemer favours, and the view of moral progress that it entails. In Sect. 3, I summarize the empirical data that Huemer cites in support of this view. In Sect. 4, I point out why Huemer thinks that his realist explanation for these data is superior to that of antirealists, by highlighting three challenges Huemer poses to antirealists. In Sect. 5, I address these challenges, and argue that antirealists can draw upon powerful and underappreciated resources to explain the historical trend of moral convergence. In Sect. 6, I move from defence to offense: not only can antirealists overcome Huemer’s challenges, but in fact their account has several explanatory virtues, whereas Huemer’s account has several explanatory shortcomings. I conclude that the relevant historical data are best explained by assuming an antirealist metaethics.
2 Realism and diffusing explanations
Realists have fashioned various responses to the challenge of moral disagreement (e.g. Shafer-Landau 2003; Enoch 2009). Similar responses play a role in realist arguments in favour of moral convergence. Typically, such responses involve two components: an empirical challenge to the claim that widespread moral disagreement exists, and an explanation of extant disagreement in terms compatible with realism. Following Doris and Plakias (2008), let’s call the latter a diffusing explanation.
As these diffusing explanations reveal, realists need not commit themselves to the claim that moral inquiry necessarily leads to the successful discovery of moral facts. Indeed, such a commitment would be extremely difficult to maintain in the light of recent work in moral psychology, which shows that moral reasoning is susceptible to various biases (e.g. Schwitzgebel and Cusman 2012; Fitzpatrick 2014). But in order to resist global scepticism, realists should maintain that moral inquiry is at least somewhat reliable: we are not hopeless at tracking the mind-independent moral facts.
Application of the same moral principles yields different verdicts in different cultural contexts (e.g. in situations of scarcity versus situations of plenitude);
Apparent moral disagreement is actually due to disagreement over non-moral issues (e.g. the truth of theism);
Individuals are making moral judgments on an epistemically deficient basis (e.g. by echoing cultural biases).
Diffusing explanation (3) plays a key role in the version of moral realism—rationalist intuitionism—that Huemer (2008, 2016) defends. The central tenet of rationalist intuitionism is that human beings have a capacity to acquire a priori knowledge of mind-independent moral facts. The means by which we acquire such knowledge are our moral intuitions, which Huemer characterizes as ‘initial intellectual appearances, that is, states of its seeming to one that something is the case upon intellectual consideration (as opposed to sensory observation or introspection), but prior to reasoning’ (Huemer 2008, pp. 370–371). Not all moral intuitions are reliable; acquiring moral knowledge is a fallible process, just like acquiring scientific or mathematical knowledge.3 There are various types of psychological bias, reasoning errors and cultural dogmatism that might lead us to have defective intuitions. These biases are pervasive and not easily overcome; attempting to overcome them has taken up the bulk of moral history and is an ongoing process.
On Huemer’s view, then, moral progress is made by overcoming biases. In the distant past, most people had a very misguided set of moral beliefs—just like most people had a very misguided set of scientific beliefs. But a capacity to recognize—however vaguely—mind-independent moral facts has created a slow yet systematic convergence of people’s moral beliefs towards the moral truths. This push was triggered by moral reformers, who diverged from social dogma: their ‘intellectual seemings’ were less biased than those of others, and their views spread. Thanks to these reformers—whose role is not unlike that of protagonists in the history of science and mathematics—societies have gradually changed their moral stances and have adopted roughly the same set of liberal values. This is not to say that the process of moral development has reached its end and that, at present, we have become acquainted with the precise or complete set of moral truths. But we do know, Huemer claims, that our current liberal values are better than the values of the past. Whatever the precise moral truths turn out to be, it is likely that the historical shift towards liberal values has taken us a long way towards recognizing them.
3 The historical explanandum
By liberal values, Huemer does not mean the values of political liberalism. Instead, what he has in mind is a broad ethical orientation which consists of three related commitments: (1) recognition of the moral equality of persons; (2) respect for the dignity of the individual; (3) opposition to gratuitous coercion and violence. Huemer cites several empirical trends in support of the view that societies have been converging towards this liberal orientation, many of which are familiar from Steven Pinker’s discussion in The Better Angels of our Nature (2011). Key data are the lowering rates of violence, war and murder from prehistoric societies up until the present; the widespread abolition of torture and execution over the past four centuries; the universal outlawing of slavery over the past two centuries; the ongoing opposition against racism and sexism, particularly over the last century; the spread of democracy, particularly over the last few decades; and the non-violent means by which former colonies of liberal states have gained sovereignty in modern times.
Of course, there are counterexamples to the trends Huemer cites. Unfortunately, in spite of legal prohibitions, practices of slavery, racism and sexism are still widespread. Over the past decade, some countries have clearly lessened their commitment to (1)–(3). Over the past century, we have witnessed two world wars and many instances of violence, as well as genocide. Over the past millennia, commitments to human equality have waxed and waned. Rather than following a straight line leading up to the adoption of liberal values, our moral history has been full of twists and turns. Undoubtedly, recognizing a historical trend towards liberalism is partly a consequence of selection bias and partly the assumption of a Western perspective.
One strategy that critics of Huemer’s (2016) argument could employ, therefore, is to dismiss the supposed trend towards liberal values on empirical grounds. But we should keep in mind that specific historical counterexamples need not suffice to undermine Huemer’s general empirical claim: diffusing explanations might help realists to explain local episodes of moral regress. Perhaps, at certain points in history, some cultures have not become less, but more clouded by bias. As long as the overall historical trend suggests that there has been a trend of moral convergence, realists might argue, the empirical data still give support to liberal realism.
In what follows I shall, for the most part, pursue a different strategy. While I think that the cited trend is less robust than Huemer suggests, it would be too quick to dismiss the data that Huemer (2016, Section 3) puts forward as a mere fluke. Huemer makes a good case that there has been a general moral development, accelerating over the last two centuries, in a direction that we typically regard as progressive—an explanandum that is metaethically relevant. Moreover, what makes this trend particularly striking is the fact that it can be recognized in different societies and with regard to different issues (war, murder, slavery, democracy, women’s suffrage, racial segregation, torture, execution, colonization), each moving in a similar direction.
How should we explain this coincidence? The best explanation, Huemer contends, is that over time different societies have come to accept the same mind-independent liberal moral truths. I shall argue, by contrast, that the relevant historical data are better explained in antirealist terms.
4 Why Huemer takes the realist explanans to be superior
It is important to keep in mind that in speaking of the best explanation in this context, what interests us is the best historical explanation of the relevant data. This historical explanation, in turn, may proceed against a background of specific metaethical assumptions—in the present case either by positing the existence of mind-independent moral truths (moral realism), or by denying that such truths exist, or that they are mind-independent (moral antirealism). Hence, we are interested in the following question: from a historian’s point of view, are the relevant data better explained against the background of a realist or an antirealist metaethics?4
Huemer attempts to establish the greater plausibility of a realist explanation inter alia by advancing three challenges for antirealist explanations.5 I outline these challenges in the present section, and reply to them in Sect. 5.
4.1 Evolutionary explanations and the ‘scope challenge’
In recent years it has become fashionable among antirealists to explain the basic contents of our moral judgments from a broadly evolutionary perspective (e.g. Joyce 2006; Street 2006; Kitcher 2011). But are evolutionary explanations also suitable to account for more recent developments, such as the historical trend towards liberalism? Huemer thinks that they are not: this recent trend lies beyond the scope of evolutionary explanations (see also FitzPatrick 2012; Buchanan and Powell 2015). We may call this Huemer’s scope challenge: the challenge for antirealists to show that recent value changes are not beyond the explanatory scope of evolutionary theory.
Consider the anti-racism movement in the United States, which only led to the abolishment of the Jim Crow laws in the 1960s. It is highly implausible that this moral accomplishment had much to do with the spread of specific genes. Huemer argues that the realist’s alternative is much more plausible: what explains the success of the anti-racist campaign is that the campaigners discovered a moral truth, and that this truth—racism is morally wrong—was generally recognized during the second half of the twentieth century.
It should be kept in mind that realists are not committed to the view that this moral shift was entirely driven by the recognition of this mind-independent moral truth, or that evolution plays no role whatsoever in explaining our moral beliefs (cf. FitzPatrick 2012, p. 174). The point of the scope challenge is that appeals to evolution do not suffice to explain the contents of our moral endorsements. Especially where moral values are concerned that have only recently been adopted, such explanations seem problematic. Therefore, Huemer claims, the antirealist’s evolutionary account falls short.
4.2 Cultural explanations and the ‘coincidence challenge’
Antirealists do not have an obvious explanation for this coincidence, Huemer claims. Realists, on the other hand, do: different cultures have converged by recognizing the same mind-independent moral truths.
Are we to believe it is coincidence that, at the same time that slavery was becoming economically inefficient, some other trend was leading women’s suffrage to become more popular (…), another trend was causing democracy to spread across the world, another was causing war to seem less glorious, another made torture seem less beneficial, and so on? This is not just a series of unrelated changes; they are all changes in line with a certain coherent ethical perspective: all the changes fit together, in one way or another, with the value of equal respect for the dignity of persons (Huemer 2016, p. 1999).
4.3 Rational explanations and the ‘inference challenge’
A third possible explanation for the shift towards liberalism refers to our rational capacities. In progressive moral transitions, moral reasoning presumably plays an important role. Consider the view advanced by Peter Singer, according to which moral progress is driven by an expansion of people’s circle of moral concern. That we should expand our circle of concern, Singer holds, is a truth that we can track through moral reasoning. But Huemer points out that it is not a truth of logic. In fact, it seems inescapable to appeal to intuitive moral truths in explaining progressive moral shifts. We may call this Huemer’s inference challenge: the challenge for antirealists to explain how moral reasoning, unaided by intuitions about mind-independent moral truths, can lead to progressive moral changes.
On Singer’s view, this kind of inference is illustrative of progressive moral change. But where does the new moral insight (2) come from? There appears to be no good explanation for this insight, other than an appeal to a deep-seated moral intuition—precisely as Huemer’s rationalist intuitionism suggests. Hence, it appears that realist intuitionists are better able than antirealists to elucidate the role that rational intuitions play in progressive moral change.
Prior judgment We should respect the interests of individuals within our tribe.
New moral insight There is no morally relevant difference between individuals within and outside our tribe.
Posterior judgment We should respect the interests of individuals outside our tribe.
5 How antirealists can counter Huemer’s challenges
In the previous section I outlined Huemer’s main considerations for thinking that, assuming an antirealist metaethics, we cannot adequately explain the historical shift towards liberalism. In this section I challenge these considerations. I will argue that antirealists can meet Huemer’s coincidence challenge and inference challenge. The scope challenge cannot easily be met, but for many antirealists this won’t be much of a concern, since the challenge does not target their view.
If my arguments in this section will be along the right lines, they undercut Huemer’s historical argument for the presumed advantage of realism over antirealism. Moreover, in Sect. 6, I will proceed to argue that the antirealists’ historical explanation is in fact superior.
5.1 Answering the ‘scope challenge’
Let’s start with the scope challenge. We should agree with Huemer that the contents of some—perhaps many—moral judgments are beyond the explanatory scope of evolutionary theory. But it should also be noted that antirealists are not typically committed to thinking otherwise. Kitcher (2011), for example, holds that a capacity for normative guidance originally served the evolutionary functions of overcoming a lack of responsiveness to others and fostering human cooperation. But Kitcher also holds that the emergence of moral thought eventually prompted an ongoing ethical discussion, which can go in any number of directions—and is not strictly tied to its evolutionary function. Or take Street (2006), who argues that our ‘basic evaluative tendencies’ have been directly shaped by natural selection. But Street is not committed to the adaptationist view that the same holds for the contents of our ‘full-fledged evaluative judgments’. By contrast, she grants that various social, cultural and historical influences, as well as the sui generis influence of rational reflection, have influenced these contents (Street 2006, p. 114). For these antirealists, then, it won’t be much of a concession to admit that values that are of relatively recent origins are beyond the scope of evolutionary explanations. What matters is whether they can adequately be explained in other antirealist-friendly terms—that is, without positing the existence of mind-independent moral truths (see Hopster 2018 for discussion).
5.2 Answering the ‘coincidence challenge’
A promising strategy for antirealists to dismantle Huemer’s coincidence challenge is to point out that the historical processes Huemer cites are entangled. Here I pursue a strategy along these lines, in three steps. First, I argue that the coincidence poses less of an explanatory burden for antirealists than Huemer suggests, because the coinciding phenomena directly influence each other. Second, I point out that the coinciding phenomena are indirectly influenced by shared common causes, which further reduces the antirealist’s explanatory burden. Additionally, I point to the directionality of historical change: for what remains of a coincidence, antirealists are well placed to explain it in functional terms.
5.2.1 First step: direct cultural influence
Coincidences are in need of explanation insofar as the coinciding phenomena are independent (if not, this explains their coincidence). But for many of the historical trends that Huemer highlights—the abolition of torture, execution and slavery; the opposition to racism and sexism; decolonization and the spread of democracy—such independence is questionable. As a result, the coincidence constitutes less of an explanatory burden for antirealists than might initially seem to be the case.
First, these trends are not fully independent because they have influenced each other. For instance, throughout history slavery has tended to be a racist practice; the opposition to slavery and racism are historically entangled. Similarly, the rise of democracy was affected by the process of decolonization and the emerging ideal of popular sovereignty. The arguments, tactics and language of empowerment used by oppressed groups have affected the emancipatory struggles of others. Consider the nineteenth and twentieth century struggles for racial and sexual equality in the United States: even if these movements did not always support each other’s cause, they did influence each other with their mutual appeal to rights.6 While such influences do not explain the direction of the historical shift itself, they do explain, to a large extent, the commonality of this direction with regard to many distinct issues. Rather than being seen as a concurrent series of directional changes, then, the historical trend that Huemer cites can largely be regarded as an interconnected historical process.
Second, societies have influenced each other. Rather than being developments that were driven by internal social dynamics, there is ample evidence that many of the convergences Huemer cites are actually due to the fact that societies have pushed each other in certain directions—by setting examples, by using soft power, by means of force and through negotiation. The ideals of the American Revolution set a precedent for its French counterpart; the Indian independence movement inspired many others. In a world with a global traffic of goods and information, in which societies depend on each other for resources and share mutual goals, it should not come as a surprise that they also come to adopt a roughly shared moral outlook. Indeed, given that social conformity is a very common phenomenon in moral reasoning (e.g. Haidt and Bjorklund 2008), and that people often internalize the norms of their cultural environment, it would have been a surprise if, after centuries of globalization, we had not seen values shift in a roughly shared direction.
Perhaps the most dramatic example of a global shift towards liberal values has been the adoption of the Universal Declaration of Human Rights (UDHR 1948), which has the defence of human dignity and human rights at its very core. Obviously, not all countries independently came to support this declaration. Instead, its near-global endorsement was the outcome of deliberation and negotiation through a global platform: the United Nations. The same goes for many other international treaties and declarations, which have contributed to the joint adoption of a broad moral outlook: they are not the product of individual trajectories, but of mutual cultural influences.
5.2.2 Second step: non-moral convergence
The Universal Declaration of Human Rights brings us to another major explanatory component of the shared direction of moral change: the common historical experience of different societies. The UDHR was adopted after the Second World War, which provided the immediate impetus for its creation. Historical experience can provide a rich source for gaining moral insight—insight about which norms lead to human flourishing and to human suffering, and which institutions are effective in fostering human cooperation. Moreover, historical experiences are often shared by different societies and may constitute a common cause for convergent moral trajectories.
In fact, there are several common causes that may explain, at least in part, the global shift towards liberalism. For instance, over time many societies have developed similarly in terms of material wealth, technology, healthcare and scientific knowledge. These non-moral convergences, in turn, may explain much of the apparent convergence in moral terms. Of course, societies have not universally moved in the same direction—but neither is the trend towards liberalism a universal trend. That said, regarding Western countries, where the shift towards liberal values is most apparent, there have also been notable parallels in terms of non-moral changes.
Consider one of the trends that Huemer cites as part of his explanandum: the lowering rates of violence, war and murder over the past millennia. This trend is clearly correlated with an increase in wealth, health, education and commercial activity, as Pinker (2011, p. xxiii) recounts. Moreover, while the exact causal dynamics will be difficult to pin down, we may hypothesize that the latter developments can also explain—at least in part—why violence, war and murder have declined: plausibly, conditions of poverty, poor health and poor education more easily give rise to violent struggle than conditions of affluence do. Or consider the abolition of slavery, which is correlated with the transition, at least in many Western countries, from agricultural to industrial societies. Huemer (2016) is aware of this correlation, but thinks it would be speculative to regard the relation as causal. But in fact it is not: Pleasants (2010) provides clear historical evidence that the rise of the market economy proved to be a crucial trigger for an effective abolitionist campaign (see also Sect. 6.2).
5.2.3 Third step: functional explanation
Similar material and informational conditions often invite similar social problems, and lend themselves to similar solutions—including solutions along moral lines. Many metaethicists—realists and antirealists alike—maintain that moral norms and values are typically functional. For instance, Gibbard (1990) argues that moral norms serve the function of coordinating our actions so that they comply with adaptive rules of conduct. Copp (2008) claims that morality has the function of enabling a society to meet its needs. Kitcher (2011) holds that moral norms serve the function of solving ‘altruism problems’ and fostering social cooperation. Even if their accounts differ regarding whether they provide a predominantly evolutionary or sociocultural explanation for the function of moral norms, these authors—along with many other metaethicists—stand united in arguing that the contents of moral norms can typically be explained in functional terms.
Let’s assume that these accounts are along the right lines and that moral norms are typically functional—in evolutionary, social or other terms. This observation is relevant for predictions about moral convergence, since functions give rise to constraints. If moral codes tend to be fitness-enhancing, then we should expect that in different societies, under similar conditions, these codes will typically evolve along convergent trajectories. Likewise, if moral norms serve to overcome social cooperation problems, then we should expect that in different societies facing similar cooperation problems, such norms will be similar in outline. Convergence need not be perfect; some problems can be solved in multiple ways. Often, though, similar problems call for similar solutions. Therefore, some degree of historical moral convergence should be our default expectation—for realists and antirealist alike.
With regard to the broad patterns of convergence that Huemer highlights, this broad appeal to moral functionalism provides antirealists with another instrument to explain the empirical data. Recall that the broadly liberal orientation in which Huemer is interested—of respecting human dignity, recognizing equality and opposing gratuitous coercion—gained global adherence over the last few centuries, with accelerating speed. Two major historical developments that took place during the same centuries were the transition to an industrial economy and the emergence of a global network of reliable and increasingly rapid information exchange. These developments, antirealists might argue, are very much entangled with our shifting moral perspectives. First, they initiated processes and made salient problems that were quite specific to this globalizing and industrializing world, such as the sustained interaction between people of different ethnicities and the emergence of global inequality. Second, the economic and industrial changes afforded new forms of human interaction and cooperation, such as the transition from a system of slave labour to a system of wage labour.
Against this background, antirealists are well positioned to explain why over the course of history a coherent shift in our moral outlook has taken place, in a broadly liberal direction. In part, this shift can be explained as the result of cultural pressures and non-moral convergences; in part, the adoption of liberal values may be regarded as solving some of the cooperation challenges specific to a globalizing world. Additionally, to a large extent this shift has been a response to lessons learned from history. The outcome of the two world wars played a pivotal role in the accelerated adoption of liberal values during the second half of the twentieth century: Europe’s traumatic experience of rule under the decidedly violent and racist Nazi regime fuelled the liberal wave in the decades that followed. Western rivalry with—and the later collapse of—the Soviet Union subsequently aided the spread of liberal values.
5.3 Answering the ‘inference challenge’
To provide a comprehensive explanation for the shift towards liberalism, antirealists also have to account for the role that reasoning played in it, without appealing to intuitions about mind-independent truths. Hence, they have to overcome what I have called Huemer’s inference challenge. I will argue that, contrary to what Huemer suggests, antirealists have a powerful—and underappreciated—resource for doing so: the phenomenon of consistency reasoning (Campbell and Kumar 2012).
As Kumar points out, the upshot of exposing moral inconsistencies can be twofold. Either the moral agent revises (or is pressed to revise) one of her extant beliefs, such that consistency is restored, or she imposes consistency upon her extant beliefs by arguing that there is a morally relevant difference between them. In theory, of course, it is also possible to reject the metanorm that we should hold moral beliefs that are consistent, but in practice this option is rarely favoured: striving for consistency is an aim that moral reasoners are typically unwilling to give up.
Suppose that although I am an avid dog owner, I also eat meat. A vegetarian friend, however, presses me: what’s the difference between factory farming and practices that I already consider abhorrent, like dog fighting? Faced with this challenge, and provided that I am disposed to trust my interlocutor, I should either decline to condemn dog fighting or, more likely, change my opinion about factory farmed meat. Unless I can find some morally relevant difference between these two practices, I should treat like cases alike Kumar (forthcoming).
Campbell and Kumar (2012) hypothesize that over the course of history dialogical appeals to consistency have served as an important engine of progressive moral change. This hypothesis seems plausible, too, with regard to the shift towards a liberal moral orientation. For instance, the demand to apply the same norms and privileges consistently to different people, irrespectively of gender, race or sexual orientation, has been a key argument in modern emancipatory struggles. Similarly, many of Singer’s influential arguments in support of animal welfare—which constitute a recent extension of Huemer’s trend towards liberalism—are consciously driven by a demand for consistency.
Against this background, let’s revisit the inference challenge: the challenge for antirealists to explain how moral reasoning, unaided by intuition, can lead to progressive moral change. Taken by itself, the phenomenon of consistency reasoning does not suffice to explain such change in non-moral terms. That we should expand our circle of moral concern is not a truth of logic, as Huemer correctly observes, but a moral premise. Instead, how we specify the circle of partners to whom we should extend our concerns, and which issues or agents we single out as demanding consistent moral treatment, is itself driven by moral considerations.
Huemer criticizes antirealists for their inability to explain how moral reasoners arrive at this new moral insight (2). But antirealists can actually come up with a clear account of reasoning-induced moral change without appealing to intuitions about mind-independent moral truths: such change comes about through consistency reasoning.
Prior judgment We should respect the interests of individuals within our tribe.
New moral insight There is no morally relevant difference between individuals within and outside our tribe.
Posterior judgment We should respect the interests of individuals outside our tribe.
As said, a demand for consistency does not necessitate that moral reasoning leads to greater inclusivism. In fact, it might just as well lead to less inclusivism—and historically, there have been plenty of examples of societies moving in this alternative direction too. Additionally, it is not necessary that moral reasoners adhere to the metanorm of being consistent. But given our psychological inclination to favour a consistent moral stance, it is perfectly explicable, in antirealist terms, that moral reasoning has often been a trigger for progressive moral change.7 The search for consistency has imposed upon the history of moral debate its own internal dynamic, which provides antirealists with yet a further ingredient to explain the broad trend towards liberalism.
6 Why the antirealist’s explanans is superior
In the previous section I have argued that Huemer’s three main challenges for antirealists to explain the historical convergence of moral values can all be met. If this argument has been successful, then Huemer can no longer claim that realists are better placed than antirealists to explain this convergence.
Moreover, in this section I will argue that a historical explanans which assumes an antirealist metaethics is in fact superior to Huemer’s realist explanans. I do so by highlighting specific virtues of the former explanation, as well as shortcomings of the latter. Some of these virtues and shortcomings are closely related, and perhaps reducible to each other. I distinguish between them for purposes of exposition, but their dialectical weight is most evident when considering them in tandem: they show that in accounting for the relevant historical phenomena, antirealists have the better explanation.
6.1 Antirealist explanations are more robust
One reason to favour an antirealist explanation is that it can illuminate a broad range of historical explananda, including historical episodes of moral divergence and regress, which are more difficult to account for assuming a realist metaethics. As mentioned in Sect. 3, there are various counterexamples which do not easily fit the historical trend of convergence towards liberal values, such as the extreme racism of Nazi-Germany. Additionally, depending on the historical period under consideration and the examples chosen, we can identify several instances of increasing moral divergence, both within and between populations. Historical explanations which assume antirealism—i.e. which assume that either there are no moral truths, or moral truths are mind-dependent—can accommodate the many twists and turns of our moral history fairly straightforwardly, whereas realist explanations for divergence and regress (e.g. the Nazi’s were extremely clouded by bias) typically provide little historical insight (see also Sect. 6.4), and are sometimes ad hoc. If antirealist accounts can explain a broader range of historical cases, and are less dependent on cherry-picking the right examples, then we should favour them on historical grounds.
6.2 Antirealist explanations have a better fit with the data
Not only are antirealists better able to explain episodes of moral regress and divergence—historical explananda that go beyond the trend which Huemer has picked out. Also with regard to some of the episodes of progress and convergence which Huemer highlights, antirealist explanations have a better fit with the data. Consider the abolition of slavery. What made abolitionist campaigns succeed? Assuming rationalist intuitionism, we would expect that the intellectual recognition of the moral fact that slavery is bad was the trigger for change. But in fact the disputed moral status of slavery was recognized well before the abolitionist movement gained momentum. Pleasants (2010) argues, on the basis of detailed engagement with historical sources, that the success of abolitionism required that a competitive alternative to slavery and other forms of coerced labour could be offered. With the emergence of the market economy and the system of wage labour such an alternative arose, and the conditions were put in place to advance strong arguments against slavery. Hence, it was not merely a novel intellectual recognition that triggered moral change; practical barriers to envisioning slavery’s dispensability had to be overcome.
These data fit well with an antirealist account, in which moral reasoning—driven by a demand for consistency—is socially embedded and takes place against various economic and technological constraints. What typically drives moral progress, antirealists maintain, is not just the acquisition of a new moral insight but also, at least in part, changing social and material conditions. Of course, realists can also appeal to changing social and material conditions to explain moral change. But on realist accounts such factors can only be of secondary importance; the most important explanatory factor is people’s recognition of mind-independent moral truths.
6.3 Antirealist explanations are more predictive
If a historical explanation is predictive of actual historical changes, this counts as an important virtue of the explanation. There is at least one respect in which antirealist accounts clearly have this explanatory virtue, and there is at least one respect in which realist accounts do not.
Assuming antirealism, an important predictor of shifts in moral value is the variety of attitudes taken into account in moral discussion. Antirealists typically think that novel moral insights come about through moral dialogue; moral progress can be made by examining the merits and weaknesses of the arguments put forward. In turn, the variety of people participating in this dialogue will influence which voices are heard and which arguments are put forward. When the variety of attitudes taken into account in moral discussion increases or diminishes greatly, we should expect that this will typically result in the endorsement of a very different set of moral values. We should keep in mind that greater inclusion does not invariably lead to greater consensus, but antirealists are not committed to thinking otherwise: they can give a unified explanation of both convergence and divergence. What matters is that, assuming antirealism, changing levels of inclusivity in moral discussion will typically coincide with changes in society’s prevailing moral stance. This is indeed what we find when looking at historical data. Especially with regard to the recent value shifts Huemer alludes to—e.g. changing stances towards slavery, racism and sexism—the inclusion of previous out-groups in moral discussion provides a clear predictor of moral change (Anderson 2016).
Assuming realism, and assuming—as Huemer seems to do—that discovering moral truths is akin to discovering non-obvious scientific or mathematical truths, we would expect to find, throughout the course of history, small groups of moral experts who defend and debate non-obvious moral claims. But this is not what we actually find. The historical protagonists of moral change were not typically individuals solving moral conundrums, but interest groups pushing to be given their due. Admittedly, we occasionally do find moral protagonists who, on the basis of a reasoning process, arrive at moral views that a majority of people finds too radical to endorse. But antirealists are well placed to account for this in terms of consistency reasoning. As Campbell and Kumar (2012, p. 288) observe, ‘the motive to be consistent can keep one from conforming to the group, as when one bravely calls attention to and seeks to resolve an inconsistency that others in one’s community refuse to face’. Rather than having a superior capacity to discern mind-independent moral truths, antirealists may argue that the distinctive quality of moral reformers is their capacity to apply consistency norms in novel contexts, thereby taking moral discussions in new directions.
6.4 Realist explanations are insufficiently detailed
As the case of slavery suggests, antirealist explanations are often richer in detail than realist explanations. With concern to historical explanations, such richness of detail typically counts as an explanatory asset. Historical explanations that build on a realist metaethics, by contrast, have a tendency to remain very general, and contribute less to our historical understanding. Consider the question of why slavery was abolished. Following Huemer’s account, what explains its abolition is first and foremost the fact that slavery is unjust. But in and by itself this is not very illuminating, as becomes apparent when we further inquire about the details of slavery’s abolition. Why was the unjustness of slavery only generally recognized during the nineteenth century? Why then, and not before? This is all the more surprising given that arguments against slavery date from much earlier periods. If these earlier arguments were fuelled by the recognition of a mind-independent moral truth, why did they not persuade people? The antirealist offers a more articulate hypothesis here, by underscoring the importance of changing socio-economic conditions as a precondition for moral change.
6.5 Realist explanations are incomplete
Another weakness of realist explanations stems from the vagueness of a realist epistemology. How do we acquire moral knowledge, assuming realism? As noted, rationalist intuitionists argue that acquisition of such knowledge is typically culturally induced but that the ultimate sources of moral knowledge are the intuitions of moral protagonists, who have recognized mind-independent moral truths. This explanation leaves unclear, however, how these mind-independent truths are intuited in the first place and how other people know that the intuitions of moral reformers are reliable. To elucidate this seems necessary to shed sufficient light on the explanandum; in the absence of such elucidation, the realist’s explanation is incomplete.
Perhaps such elucidation is forthcoming, but the prospect of realists to specify how the intuitions of moral reformers allow them to gain reliable access to mind-independent moral truths is fraught with difficulties. Huemer, for instance, leaves open the possibility that apart from being mind-independent, moral truths are causally inert: they do not causally influence the intuitions of moral reformers. But if moral facts exist independently of the intuitions of moral reformers and do not causally influence these intuitions, then it is unclear how these reformers could have reliably tracked moral truths—a problem reminiscent of the Benacerraf-Field challenge for mathematical Platonism (Hopster 2019).
Note that this is a familiar problem for realism, and not specific to the argument I have presented. But it is related to the present argument, since absent a clear account of how we are able to acquire moral knowledge, it remains equally unclear how we have been able to acquire such knowledge over the course of history.
6.6 Realist explanations are unclear
To explain the historical shift towards liberalism, antirealists can appeal to cultural changes (e.g. the changing variety of people involved in moral discussion), material changes (e.g. the consequences of the Industrial Revolution) and changes in moral attitudes (e.g. recognizing that one group has been morally privileged, and arguing, by way of consistency reasoning, that other groups are entitled to the same privileges), as well as various other factors. Realists may appeal to the same factors, but must additionally appeal to a recognition of mind-independent moral truths. But if the foregoing criticisms have been along the right lines, then such an appeal does not help the explanation: it makes it vaguer than explanations in which mind-independent truths do not figure. For instance, if we lack a comprehensive account of how mind-independent moral truths are grasped (Sect. 6.5), then the resultant explanation is bound to be more obscure than an explanation that does not refer to mind-independent truths. Hence, unless the realist can overcome these previous criticisms, the antirealist’s explanation should also be preferred for reasons of clarity.
6.7 Realist explanations contain redundancies
Realist accounts of historical moral change invoke an explanatory factor which antirealist accounts do not invoke: over the course of history, we have tracked mind-independent moral truths. Moreover, realists hold that this is a necessary component of a comprehensive historical explanation. By implication, realists maintain that historical explanations that proceed against the background of moral antirealism are incomplete.
In Sect. 5, however, I have argued that this contention is mistaken. I have shown that antirealists have ample resources to account for the apparent coincidence of our global value shift, and for the role that reasoning plays in progressive moral change. If my argument has been along the right lines, then the antirealist’s explanation is by no means incomplete; it can shed light on the explanandum in a comprehensive way. Moreover, if the antirealist has indeed provided a comprehensive explanation, then arguably the realist’s explanation contains a redundancy. Parsimony considerations come into play: if all other things are equal, apart from the fact that the realist’s explanation contains a redundancy whereas the antirealist’s explanation does not, then the latter should be preferred.8
Realists may want to object to the relevance of parsimony and argue that in deciding which philosophical view to favour, parsimony considerations are irrelevant and question-begging (cf. FitzPatrick 2015). However, we should keep in mind that the present debate does not turn on the question of which metaethical view we should favour all things considered, but on the question of what is the better historical explanation: one that proceeds against the background of realism, or antirealism. With regard to competing historical explanations, parsimony considerations are certainly relevant: if mind-independent truths need not be appealed to in order to provide an adequate historical explanation, then historians should not appeal to them.
Alternatively, realists may want to reason as follows: one can only submit that the best historical explanation does not appeal to the existence of mind-independent moral truths if one has already rejected moral realism. If one has not, then mind-independent moral truths may still be regarded as explanatorily indispensable. Hence, we arrive at a stalemate in the debate: whether one thinks that an appeal to mind-independent moral truths is required for our best explanation depends on one’s underlying metaethical commitments (cf. FitzPatrick 2014, p. 248).
But in fact the present dialectic is quite different. I have discussed the realist’s best arguments, as presented by Huemer (2016), for thinking that mind-independent truths are indispensable in explaining historical moral change, and found them wanting. At this point in the dialectic, the burden of argument is with realists; they need to show that my criticisms are mistaken and that mind-independent moral truths are explanatorily indispensable after all. Unless and until such an argument is given, we may presume that the arguments I have provided are along the right lines and that mind-independent moral truths are indeed redundant in explaining historical moral change.
6.8 Realist explanations have a poor fit with other disciplines
An additional drawback of Huemer’s realist explanation is that it does not easily fit with our background knowledge from other disciplines, such as psychology and cognitive science. For instance, there is no indication from these fields of research that moral reformers engage in a process of intuiting mind-independent facts. By contrast, cognitive scientists do find that inconsistency is typically a trigger for belief change, which lends support to the antirealist’s view of reasoning-induced moral change. Realists may appeal to the same phenomenon of moral consistency reasoning, but they additionally need to appeal to the existence of mind-independent moral facts—an appeal that finds no support in findings from other disciplines. If antirealist explanations can more easily be unified with our scientific background knowledge than realist explanations, this provides a further reason to favour the former over the latter.
Huemer (2016) recognizes the relevance of historical data for assessing the adequacy of metaethical theories, and appropriately so. But contrary to what Huemer suggests, when taking into account all of the evidence, these data provide better support for an antirealist than a realist metaethics. Huemer’s criticisms of antirealist explanations are unconvincing; moreover, antirealist explanations have several virtues compared to realist accounts.
That moral antirealism has a better fit with our historical background knowledge than moral realism should not be taken to conclusively vindicate the former, or to debunk the latter. After all, more considerations are relevant for evaluating the adequacy of metaethical theories than their compatibility with our best historical explanations. But the arguments given, if successful, do raise the overall plausibility of antirealism compared with realism, as they suggest that antirealists are better positioned to shed light on an important historical explanandum. Other things being equal, we should prefer some version of moral antirealism over Huemer’s realist intuitionism on historical grounds.
To what extent moral truths are epistemically accessible, assuming realism, depends on the realist view in question, as I illustrate in Sect. 4.2. Realists are not committed to the view that moral truths are accessible to anybody at any moment in time. But all realists maintain that moral truths are at least somewhat accessible, such that the skeptical implications pressed by Street (2006, 2016) can be resisted.
Here and in what follows I use the label ‘antirealism’ in the same broad sense as Huemer (2016, p. 1983), as denoting all views which either deny that there are mind-independent moral truths, or that we can have knowledge of such truths. Note that this is an encompassing set of views, which includes versions of non-cognitivism, error theory, as well as the constructivist view that moral truths are underwritten by mind-dependent moral properties or facts (e.g. Hopster 2017).
A challenge for Huemer is to elucidate how people are able to differentiate between truth-apt intuitions and intuitions clouded by bias. Huemer’s (2008) main strategy for doing so is to differentiate between intuitions at different levels of generality and to argue that intuitions about some levels (e.g. intuitions about concrete cases) are more susceptible to bias than others (e.g. formal intuitions).
Naturally, to accord with the historian’s best explanation is not the only desideratum for metaethical theories. But it is certainly a weighty desideratum, which may tilt the overall balance of metaethical plausibility.
In fact, Huemer (2016, Section 4) discusses—and dismisses—not three but four possible antirealist explanations for the shift towards liberalism. In the present section I highlight only three of these explanations—those that strike me as most plausible.
Rather than being understood as direct cultural influences, historical connections such as these may also be understood as stemming from a common cause—e.g. from the Enlightenment ideal of egalitarianism, or from a common belief in natural rights. In explaining complex historical dynamics, the distinction between direct causes and common causes is often difficult to make. For the purposes of the present argument, however, this distinction is only of secondary importance; antirealists can call upon either kind of influence to call into question the independence of convergent historical trajectories.
Of course, in order to provide a full account for this, antirealists also need to specify what makes moral change progressive. For an antirealist defence of moral progress, see, for example, Kitcher (2011).
As a reviewer pointed out, some realists may want to deny that other things are equal, and argue instead that they substitute the antirealist’s historical explanation with a completely different explanation. If the explanations of realists and antirealists turn out to be different in many respects, then parsimony considerations are indeed beside the point. They only come into play if, apart from its appeal to mind-independent truths, the realist’s explanation is (almost) identical to that of the antirealist.
Funding was obtained from Nederlandse Organisatie voor Wetenschappelijk Onderzoek (Grant Number 360-20-340).
- Doris, J., & Plakias, A. (2008). How to argue about disagreement: Evaluative diversity and moral realism. In W. Sinnott-Armstrong (Ed.), Moral psychology (Vol. 2, pp. 303–331). Cambridge: MIT Press.Google Scholar
- Gibbard, A. (1990). Wise choices, apt feelings. Oxford: Oxford University Press.Google Scholar
- Haidt, J., & Bjorklund, F. (2008). Social intuitionists answer six questions about moral psychology. In W. Sinnott-Armstrong (Ed.), Moral psychology (Vol. 2). Cambridge: MIT Press.Google Scholar
- Hopster, J. (2018). Evolutionary arguments against moral realism: Why the empirical details matter (and which ones do). Biology & Philosophy, 33(5–6), 1–19.Google Scholar
- Hopster, J. (2019). Striking coincidences: How realists should reason about them. Ratio (online first), 1–15.Google Scholar
- Joyce, R. (2006). The evolution of morality. Cambridge: MIT Press.Google Scholar
- Kumar, V. (Forthcoming). The ethical significance of cognitive science. In S. -J. Leslie & S. Cullen (Eds.), Current controversies in philosophy of cognitive science. New York: Routledge.Google Scholar
- Mackie, J. L. (1977). Ethics: Inventing right and wrong. New York: Penguin.Google Scholar
- Pinker, S. (2011). The better angels of our nature: Why violence has declined. New York: Viking.Google Scholar
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.