First published Thu Feb 19, 2004;
https://plato.stanford.edu/entries/moral-relativism/
Moral relativism is an important topic in metaethics. It is also widely discussed outside philosophy (for example, by political and religious leaders), and it is controversial among philosophers and nonphilosophers alike. This is perhaps not surprising in view of recent evidence that people's intuitions about moral relativism vary widely. Though many philosophers are quite critical of moral relativism, there are several contemporary philosophers who defend forms of it. These include such prominent figures as Gilbert Harman, Jesse J. Prinz, J. David Velleman and David B. Wong. The term ‘moral relativism’ is understood in a variety of ways. Most often it is associated with an empirical thesis that there are deep and widespread moral disagreements and a metaethical thesis that the truth or justification of moral judgments is not absolute, but relative to the moral standard of some person or group of persons. Sometimes ‘moral relativism’ is connected with a normative position about how we ought to think about or act towards those with whom we morally disagree, most commonly that we should tolerate them.
1. Historical Background
2. Forms and Arguments
3. Experimental Philosophy
4. Descriptive Moral Relativism
5. Are Moral Disagreements Rationally Resolvable?
6. Metaethical Moral Relativism
7. Mixed Positions: A Rapprochement between Relativists and Objectivists?
8. Relativism and Tolerance
1. Historical Background
Though moral relativism did not become a prominent topic in philosophy or elsewhere until the twentieth century, it has ancient origins. In the classical Greek world, both the historian Herodotus and the sophist Protagoras appeared to endorse some form of relativism (the latter attracted the attention of Plato in the Theaetetus). It should also be noted that the ancient Chinese Daoist philosopher Zhuangzi (sometimes spelled Chuang-Tzu) put forward a nonobjectivist view that is sometimes interpreted as a kind of relativism.
Among the ancient Greek philosophers, moral diversity was widely acknowledged, but the more common nonobjectivist reaction was moral skepticism, the view that there is no moral knowledge (the position of the Pyrrhonian skeptic Sextus Empiricus), rather than moral relativism, the view that moral truth or justification is relative to a culture or society. This pattern continued through most of the history of Western philosophy. There were certainly occasional discussions of moral disagreement—for example in Michel de Montaigne's Essays or in the dialogue David Hume attached to An Enquiry Concerning the Principles of Morals. These discussions pertained to moral objectivity, but moral relativism as a thesis explicitly distinguished from moral skepticism ordinarily was not in focus. Prior to the twentieth century, moral philosophers did not generally feel obliged to defend a position on moral relativism.
Nonetheless, the increased awareness of moral diversity (especially between Western and non-Western cultures) on the part of Europeans in the modern era is an important antecedent to the contemporary concern with moral relativism. During this time, the predominant view among Europeans and their colonial progeny was that their moral values were superior to the moral values of other cultures. Few thought all moral values had equal or relative validity, or anything of that sort. The main impetus for such a position came from cultural anthropology. Anthropologists were fascinated with the diversity of cultures, and they produced detailed empirical studies of them—especially “primitive,” non-Western ones. At the beginning anthropologists accepted the assumption of European or Western superiority. But this assumption began to be challenged in the twentieth century, especially by some social scientists in the United States. An early dissent came from the sociologist William Graham Sumner, who proposed a version of moral relativism in his 1906 Folkways. But the most influential challenge originated with the anthropologist Franz Boas. He and his students—in particular, Ruth Benedict, Melville J. Herskovits, and Margaret Mead—explicitly articulated influential forms of moral relativism in the first half of the twentieth century. In 1947, on the occasion of the United Nations debate about universal human rights, the American Anthropological Association issued a statement declaring that moral values are relative to cultures and that there is no way of showing that the values of one culture are better than those of another. Anthropologists have never been unanimous in asserting this, and more recently human rights advocacy on the part of some anthropologists has mitigated the relativist orientation of the discipline. Nonetheless, prominent anthropologists such as Richard A. Shweder and the late Clifford Geertz have defended relativist positions in recent years.
An important early bridge from anthropology to philosophy was established by Edward Westermarck (1906–8 and 1932), a social scientist who wrote anthropological and philosophical works defending forms of empirical as well as metaethical moral relativism. In the latter half of the 20th century, moral philosophers began devoting considerable attention to moral relativism and some—most notably Richard B. Brandt (1954) and John Ladd (1957)—took quite seriously the empirical effort of anthropology to understand the moralities of different cultures, to the point of making such empirical inquiries themselves (an anticipation of the recent emphasis on experimental philosophy, to be discussed in section 3). In the past several decades there has been increasing consideration of moral relativism, and there is now an enormous literature on the subject (the Bibliography below is very limited). Most of these discussions are situated in the domain of “pure metaethics,” but not all. For example, there is considerable work on moral relativism in connection with human rights (with which some forms of relativism are in obvious conflict), political liberalism (which is sometimes allied and sometimes contrasted with relativism) and feminist philosophy (where relativism has been both criticized and defended vis-ŕ-vis feminist concerns). There are also discussions of moral relativism in applied fields such as medical ethics.
2. Forms and Arguments
In general, the term ‘relativism’ refers to many different ideas. For example, in anthropology it sometimes connotes, among other things, the rather uncontroversial notion that anthropologists should strive to be impartial and unprejudiced in their empirical inquires. However, in moral philosophy ‘relativism’ is usually taken to suggest an empirical, a metaethical, or a normative position. The empirical position is usually:
Descriptive Moral Relativism (DMR). As a matter of empirical fact, there are deep and widespread moral disagreements across different societies, and these disagreements are much more significant than whatever agreements there may be.
Sometimes what is emphasized is moral diversity rather than strict disagreement. DMR is often thought to have been established by anthropology and other empirical disciplines. However, it is not uncontroversial: Empirical as well as philosophical objections have been raised against it. Hence, it is one focal point of debate.
The metaethical position usually concerns the truth or justification of moral judgments, and it has been given somewhat different definitions. Metaethical relativists generally suppose that many fundamental moral disagreements cannot be rationally resolved, and on this basis they argue that moral judgments lack the moral authority or normative force that moral objectivists usually contend these judgments may have. Hence, metaethical relativism is in part a negative thesis that challenges the claims of moral objectivists. However, it often involves a positive thesis as well, namely that moral judgments nonetheless have moral authority or normative force, not absolutely or universally (as objectivists contend), but relative to some group of persons such as a society or culture. This point is typically made with respect to truth or justification (or both), and the following definition will be a useful reference point:
Metaethical Moral Relativism (MMR). The truth or falsity of moral judgments, or their justification, is not absolute or universal, but is relative to the traditions, convictions, or practices of a group of persons.
With respect to truth-value, this means that a moral judgment such as ‘Polygamy is morally wrong’ may be true relative to one society, but false relative to another. It is not true, or false, simply speaking. Likewise, with respect to justification, this judgment may be justified in one society, but not another. Taken in one way, this last point is uncontroversial: The people in one society may have different evidence available to them than the people in the other society. But proponents of MMR usually have something stronger and more provocative in mind: That the standards of justification in the two societies may differ from one another and that there is no rational basis for resolving these differences. This is why the justification of moral judgments is relative rather than absolute.
In recent years, there has been a proliferation of different formulations of relativism (for discussion of some of these, see Krausz 2011 and López de Sa 2011). It is important to note several distinctions that may be made in formulating different metaethical relativist positions. First, a distinction is sometimes drawn between content relativism, the view that sentences may have different contents (meanings) in different frameworks, and truth relativism, the view that sentences have the same content in different frameworks, but their truth-value may vary across these frameworks (for a discussion of this distinction in terms of moral relativism, see Prinz 2007: 180-3). In the discussions that follow, truth relativism is ordinarily assumed. Second, it is sometimes said that the truth or justification of moral judgments may be relative to an individual person as well as a group of persons. In this article, the latter will be assumed, as in the definition of MMR, unless otherwise noted. Third, that to which truth or justification is relative may be the persons making the moral judgments or the persons about whom the judgments are made. These are sometimes called appraiser and agent relativism respectively. Appraiser relativism suggests that we do or should make moral judgments on the basis of our own standards, while agent relativism implies that the relevant standards are those of the persons we are judging (of course, in some cases these may coincide). Appraiser relativism is the more common position, and it will usually be assumed in the discussion that follows. Finally, MMR may be offered as the best explanation of what people already believe, or it may be put forward as a position people ought to accept regardless of what they now believe. There will be occasion to discuss both claims below, though the latter is probably the more common one.
Metaethical moral relativist positions are typically contrasted with moral objectivism. Let us say that moral objectivism maintains that moral judgments are ordinarily true or false in an absolute or universal sense, that some of them are true, and that people sometimes are justified in accepting true moral judgments (and rejecting false ones) on the basis of evidence available to any reasonable and well-informed person. There are different ways of challenging moral objectivism. Moral skepticism says that we are never justified in accepting or rejecting moral judgments. Other views—variously called moral non-cognitivism, expressivism, anti-realism, nihilism, etc.—contend that moral judgments lack truth-value, at least beyond the truth-value implied by the minimalist claim that to assert that S is true is simply to assert S (a related view, the error theory, claims that moral judgments are always false). MMR is often distinguished from all of these views: Instead of denying truth-value or justification, it affirms relative forms of these. However, metaethical moral relativist views are sometimes regarded as connected with positions that say moral judgments lack truth-value, since the relativist views contend that moral judgments lack truth-value in an absolute or universal sense. This is sometimes simply a question of terminology, but not always. If it is said that moral judgments lack truth-value (beyond the claim of minimalism), then there cannot be relative truth-value in the sense that moral relativists usually intend (though it might be contended that there is a sense in which there could still be justification). As will be seen below, there is a debate about the relationship between MMR and non-cognitivist or expressivist positions.
Most arguments for MMR are based on DMR and the contention that it is implausible to suppose fundamental moral disagreements can always be resolved rationally. Sometimes it is said that some moral disagreements are faultless, meaning that neither party has made a mistake (see Kölbel 2004). For instance, Harman (1996), Prinz (2007) and Wong (1984 and 2006) have all stressed the importance of moral disagreements in arguing for MMR, and such arguments will be considered in some detail in subsequent sections. However, some arguments for MMR have a rather different approach, and two of these should be noted here.
First, MMR might be defended as a consequence of the general relativist thesis that the truth or justification of all judgments is not absolute or universal, but relative to some group of persons. For example, this general position might be maintained on the ground that each society has its own conceptual framework and that conceptual frameworks are incommensurable with one another. Hence, we can only speak of truth or justification in relative terms (see the discussion of incommensurability in the Summer 2015 archived version of the entry on relativism (section 4.2)). This position might be thought to have the disadvantage that it can only be put forward as true or justified relative to some conceptual framework (the suggestion is usually that this framework is our own), and many find it implausible with regard to common sense judgments and judgments in the natural sciences. However, this is one avenue to MMR. But most proponents of MMR focus on distinctive features of morality and reject general relativism. In fact, they often contrast morality and science with respect to issues of truth and justification. For example, Harman (2000b), Prinz (2007) and Wong (1996 and 2006) all associate moral relativism with naturalism, a position that usually presupposes the objectivity of the natural sciences.
Second, a metaethical moral relativist position might be defended by emphasizing aspects of morality other than disagreement. For example, Rovane (2011 and 2013) has maintained that relativism is best understood, not as a response to disagreement, but as a response to alternative conceptual schemes that portray different worlds that are normatively insulated from one another. On this account, the truth-bearers in one world are not logically related to the truth-bearers in another world (so there cannot be strict disagreement), and yet it is not possible to embrace both worlds (so they are alternatives). Rovane argues that in the moral domain, but not in the domain of the natural sciences, there may be different worlds in this sense. Hence, a moral judgment may be true for the occupant of one world, but not for the occupant of another. An implication of this view, she says, is that learning and teaching across different moral worlds might not be possible.
In a partially similar view, Velleman (2013) has claimed, on the basis of ethnographic and historical data, that different communities construct available action types differently. Moreover, reasons for action are always dependent on the perspective of the particular community since they arise out of the drive for mutual interpretability needed for social life within the community. Hence, there are no perspective-independent reasons. There cannot be straight-forward disagreement across these communities because they do not have common sets of action types. The communities may nonetheless address the basic themes of morality, but in incompatible ways given their different perspectives. So moralities can only have local validity.
Both Rovane and Velleman stress moral diversity rather than moral disagreement. They maintain, not that disagreements cannot be rationally resolved, but that there is no basis for showing that, among various incompatible alternatives, one is rationally superior to another.
In addition, it is worth noting that MMR is sometimes justified by appealing in a significant way to a distinctive analysis of moral judgments in combination with a claim about moral disagreement. For example, Prinz (2007) argues that what he calls “moral sentimentalism” implies a form of MMR once we acknowledge moral disagreements. According to moral sentimentalism, an action is morally right (wrong) if and only if some observer of the action has a sentiment of approbation (disapprobation) concerning it. Prinz defends this position on the basis of a metaethical argument that it is the most plausible account in light of empirical studies linking moral judgments and emotions. Since people often have conflicting sentiments about the same action, a judgment of the form 'Action X is right' may be true (when expressed by a person who approves of X), and 'X is wrong' may also be true (when expressed by a person who disapproves of X). On this view, the truth of such moral judgments is relative to the sentiments of the persons who make them. Moral sentimentalism is a crucial feature of this argument and many philosophers would deny that moral rightness and wrongness depend on our sentiments in this way. But most arguments for MMR are not based on moral sentimentalism.
In another example, Harman (2000a) argues that a moral judgment that a person ought to do X (an “inner judgment”) implies that the person has motivating reasons to do X, and that a person is likely to have such reasons only if he or she has implicitly entered into an agreement with others about what to do. Hence, moral judgments of this kind are valid only for groups of persons who have made such agreements. An action may be right relative to one agreement and wrong relative to another (this combines agent and appraisal relativism insofar as Harman assumes that the person making the judgment and the person to whom the judgment is addressed are both parties to the agreement).
Harman's relativism is presented as a thesis about logical form, but the relativist implication arises only because it is supposed that the relevant motivating reasons are not universal and so probably arose from an agreement that some but not all persons have made. In this sense, moral disagreement is an important feature of the argument. But the main focus is on the internalist idea that inner judgments imply motivating reasons, reasons that are not provided simply by being rational, but require particular desires or intentions that a person may or may not have. Internalism in this sense is a controversial view, and many would say that a moral judgment can apply to a person whether or not that person is motivated to follow it (see the section on 'Psychological: Moral Motivation' in the entry on moral epistemology). However, internalism is not a standard feature of most arguments for moral relativism, and in fact some relativists are critical of internalism (for example, see Wong 2006: ch. 7)
It is worth noting that internalism is one expression of a more general viewpoint that emphasizes the action-guiding character of moral judgments. Though Harman and others (for example, Dreier 1990 and 2006) have argued that a form of moral relativism provides the best explanation of internalism, a more common argument has been that the action-guiding character of moral judgments is best explained by a non-cognitivist or expressivist account according to which moral judgments lack truth-value (at least beyond the claim of minimalism). In fact, some have claimed that the expressivist position avoids, and is superior to, moral relativism because it accounts for the action-guiding character of moral judgments without taking on the problems that moral relativism is thought to involve (for instance, see Blackburn 1998: ch. 9 and 1999, and Horgan and Timmons 2006). By contrast, others have maintained that positions such as non-cognitivism and expressivism are committed to a form of moral relativism (for example, see Bloomfield 2003, Foot 2002b, and Shafer-Landau 2003: ch 1). For an assessment of this debate, see Miller 2011, and for a discussion of non-cognitivism and related positions, see the entry on moral cognitivism vs. non-cognitivism.
Finally, the term ‘moral relativism’ is sometimes associated with a normative position concerning how we ought to think about, or behave towards, persons with whom we morally disagree. Usually the position is formulated in terms of tolerance. In particular, it is said that we should not interfere with the actions of persons that are based on moral judgments we reject, when the disagreement is not or cannot be rationally resolved. This is thought to apply especially to relationships between our society and those societies with which we have significant moral disagreements. Since tolerance so-understood is a normative thesis about what we morally ought to do, it is best regarded, not as a form of moral relativism per se, but as a thesis that has often been thought to be implied by relativist positions such as DMR and MMR. Despite the popularity of this thought, most philosophers believe it is mistaken. The main question is what philosophical relationship, if any, obtains between moral relativism and tolerance.
The remainder of this entry will discuss DMR, the contention that it is unlikely that fundamental moral disagreements can be rationally resolved, arguments for and challenges to MMR, mixed positions that combine moral relativism and moral objectivism, and the relationship between moral relativism and tolerance. But first there needs to be some consideration of the recent contributions of experimental philosophy to these discussions.
3. Experimental Philosophy
Experimental philosophy is an approach to philosophy that explicitly draws on experimental knowledge established by the sciences to address philosophical questions (see the entry on experimental moral philosophy). There are three significant ways in which experimental philosophy has played an important role in discussions of moral relativism. These concern the extent to which there is moral disagreement or moral diversity among people (that is, DMR), the extent to which folk morality is committed to an objectivist or relativist understanding of moral judgments (that is, the views of ordinary people concerning MMR), and the extent to which acceptance of moral relativism affects moral attitudes such as tolerance (that is, ways in which views concerning MMR causally influence whether or not people have tolerant attitudes).
The first of these has a long history in discussions of moral relativism and in fact may be considered one of the earliest instances of experimental moral philosophy. As was seen in section 1, for more than a century the work of anthropologists and other social scientists has contributed to the development of thought about moral relativism, both by purporting to provide empirical evidence for extensive cross-cultural disagreement and diversity about morality, and by proposing the notion that moral codes are true only relative to a culture as the best explanation of this. That is, these scientists have provided empirical grounds for accepting DMR, and they have suggested that some form of MMR is a reasonable inference from this data (though these positions were not always clearly distinguished). More importantly, the work cited in section 1 by Brandt (1954) and Ladd (1957), involving both empirical investigations into the moral values of Native Americans and philosophical reflection on the significance of these investigations vis-ŕ-vis moral relativism, are significant examples of moral philosophers engaging in empirical inquiry in support of philosophical aims. Their empirical work did not immediately inspire other other philosophers to engage in similar research. Experimental philosophy in this sense--experiments or other empirical investigations conducted by philosophers--did not become prominent until nearly a half-century later. Nowadays philosophers do sometimes conduct experiments to investigate the extent of moral disagreement (for example, see the study of Western and East Asian values cited in Doris and Plakias 2008). What has been much more common in recent decades has been the citation by philosophers of empirical studies by anthropologists to establish facts about moral disagreement or diversity (for example, see Prinz 2007, Velleman 2013, and Wong 1984 and 2006). There is more about these issues in section 4.
The second concern, the extent to which ordinary people accept some form of moral objectivism or some form of MMR (or some other non-objectivist position), has been the subject of considerable experimental research in recent years. This research has sometimes been conducted by psychologists (or other scientists), sometimes by philosophers, and increasingly sometimes by both working together (for overviews of this literature, see Quintelier et. al. 2012 and Sarkissian Forthcoming). In the past, philosophers with a variety of meta-ethical commitments have sometimes claimed that in everyday moral practices people implicitly suppose that moral objectivism in some sense is correct (for example, see Blackburn 1984: 180 and Jackson 1998: 137). By contrast, on occasion some philosophers have maintained that ordinary people sometimes have attitudes that conflict with objectivism. For instance, Wong has argued that in some moral disagreements people grant that the person with the conflicting moral judgment is reasonable in accepting the judgment to the extent that these people are unsure if their own position is uniquely right--what he calls “moral ambivalence” (see Wong 2006: ch. 1). So who are correct, philosophers who claim that ordinary people accept a form of objectivism (folk moral objectivism) or philosophers who think that ordinary people at least sometimes accept something closer to MMR (folk moral relativism)?
Recent empirical research suggests that both positions have some merit: the meta-ethical views of ordinary people are rather complex. A common method for measuring whether people are objectivists or relativists about a moral belief is to present them with a disagreement between two parties concerning the belief and to ask them if at most only one party could be correct. A response that only one could be correct indicates commitment to objectivism, while a response that more than one could be correct suggests commitment to relativism (or some non-objectivist position). Several studies employing this methodology have provided evidence that, while many people are objectivists about morality, a significant number are not objectivists (for example, see Nichols 2004). Moreover, some studies have shown interesting correlations with these differences, correlations that may partly explain them. For instance, objectivists are more likely to attribute a religious grounding to morality than non-objectivists, and non-objectivists are more likely to be open to considering alternative reasons than objectivists (see Goodwin and Darley 2010). In addition, those who are more open to experience are more likely to be non-objectivists than those who are not (see Feltz and Cokely 2008).
Other studies have shown different kinds of complexity. People are more likely to be objectivists about some moral issues (such as robbery) than they are about other moral issues (such as abortion). These differences also have correlations that might be partly explanatory: regarding an issue as objective correlates with strength of belief and perception of consensus on the issue (see Goodwin and Darley 2008 and 2010). Moreover, people are more likely to be objectivists about some issues than others even when they are allowed to determine for themselves which issues count as moral issues (see Wright et. al. 2013).
Finally, it is more more probable that people give objectivist responses when they think that the parties to a moral disagreement share the same culture than when they think that the disagreeing parties belong to a very different culture. This might suggest that many of those who give objectivist responses are tacitly assuming a kind of objectivity on the assumption that the disagreeing parties have a common moral framework, but not in circumstances in which there are different moral frameworks (see Sarkissian et. al. 2011).
In short, empirical work about folk meta-ethical outlooks suggests that there is considerable diversity in the extent to which, and the circumstances under which, people express moral objectivist views or moral non-objectivist views such as MMR. This might be taken to indicate that some people are objectivists and some are not. But it might also be taken to show that some people are “meta-ethical pluralists”: they are objectivists about some moral issues, but they are relativists about other moral issues (see Wright et. al. 2013). That is, perhaps some people implicitly deny the common assumption among philosophers that all moral beliefs should be given the same meta-ethical analysis.
Various questions may be raised about the significance of this experimental work. For example, some critics might wonder to what extent the results of these experiments are indicative of the meta-ethical commitments of all human beings. In one study, subjects aged 17 to 29 were less objectivist than both older and younger subjects (see Beebe and Sackris, in Other Internet Resources). Since almost all the subjects in the studies reported above were college students, this might cast doubt on how representative the conclusions of these studies are of the population as a whole. Moreover, almost all the subjects were college students in the United States, and this might lead to concerns about how representative their views are of people throughout the world (to address part of this concern, the subjects in one of the six studies reported in Sarkissian et. al. 2011 were undergraduates in Singapore).
In any case, some philosophers may wonder about the philosophical relevance of this experimental research. On response is that it could affect criteria of success in meta-ethics. For example, it is sometimes suggested that most people are moral objectivists rather than moral relativists, and that a meta-ethical position such as moral realism gains credibility because it is in accord with folk morality so understood (see Smith 1991). The studies just cited appear to challenge the factual premise of this meta-ethical criterion. Another response is that some of the complexity revealed in these studies might lead philosophers to consider more seriously the philosophical viability of a pluralist or mixed meta-ethical position according to which, for instance, moral objectivism is correct in some respects, but MMR is correct in other respects (in this connection, see Gill 2008 and Sinnott-Armstrong 2009). There is more on this issue in section 7.
The final area in which experimental philosophy has contributed to discussions of moral relativism pertains to the relationship between relativism and moral attitudes such as tolerance. It is sometimes claimed that some forms of moral relativism provide a reason for tolerance (see section 8). But are moral relativists more likely to be tolerant than moral objectivists? Some recent psychological studies suggest that the answer may be “yes.” In the face of divergent attitudes about some issue, people were more likely to be intolerant when they regarded the issue as moral rather than non-moral (see Wright et. al. 2008), and they were more likely to be intolerant when they believed the moral issue was objectively rather than non-objectively grounded (see Wright et. al. 2014). In these studies, intolerance was measured in terms of reported willingness to interact with or help those with divergent attitudes (among other things).
These studies suggest that there is some correlation between acceptance of moral objectivism and tolerance. For those who have argued that moral relativism provides a rational basis for tolerance, this might be regarded as an unsurprising result. Of course, a psychological relationship does not show that there is a logical relationship. But some support might be derived from the fact that people are behaving in what, for this position, is a rational way. In addition, it has been claimed that an advantage of moral relativism is that, even though it does not provide a reason for tolerance, acceptance of it makes people more tolerant (see Prinz 2007: 208). These studies would provide support for this empirical claim.
4. Descriptive Moral Relativism
Most discussions of moral relativism begin with, and are rooted in, DMR. Though this is not sufficient to establish MMR, the most common rationales for MMR would be undermined if DMR (or some descriptive thesis about significant moral disagreement or diversity) were incorrect. Moreover, if DMR were generally rejected, it is likely that MMR would have few proponents. Hence, it is important to consider whether or not DMR is correct. Defenders of DMR usually take it to be well-established by cultural anthropology and other empirically-based disciplines, and many believe it is obvious to anyone with an elementary understanding of the history and cultures of the world. Examples of moral practices that appear sharply at odds with moral outlooks common in the United States are not hard to come by: polygamy, arranged marriages, suicide as a requirement of honor or widowhood, severe punishments for blasphemy or adultery, female circumcision or genital mutilation (as it is variously called), and so on (for a review of some of the literature, see Prinz 2007: 187-95). At a more general level, Wong (1984) has argued that at least two different approaches to morality may be found in the world: a virtue-centered morality that emphasizes the good of the community, and a rights-centered morality that stresses the value of individual freedom.
Though it is obvious that there are some moral disagreements, it is another matter to say that these disagreements are deep and widespread, and that they are much more significant than whatever agreements there may be. Philosophers have raised two kinds of objection to this contention: a priori arguments that DMR could not be true, and a posteriori arguments that DMR is probably not true or at least has not been established to be true.
A priori objections maintain that we can know DMR is false on the basis of philosophical considerations, without recourse to empirical evidence. One argument, expressed in general form by Donald Davidson (1984), states that disagreement presupposes considerable agreement (see the entry on Donald Davidson). According to Davidson, a methodological constraint on the translation of the language of another society is that we must think they agree with us on most matters. For example, suppose we believed there were numerous disagreements between us and another society about trees. As the disagreements piled up, we reasonably would begin to think we had mistranslated a word in the language of the other society as ‘tree’: It is more likely that (what we take to be) their false beliefs about trees are really beliefs about something else. By generalization, it follows that there could not be extensive disagreements about trees between our society and the other one. Of course, there could be some disagreements. But these disagreements would presuppose substantial agreements in other respects. Davidson (2004a and 2004b) and others (for example, Cooper 1978 and Myers 2004) have claimed that this argument applies to moral concepts. If they are right, then there cannot be extensive disagreements about morality, and the agreements are more significant than the disagreements. DMR cannot be true.
Davidson's argument is controversial. One response is that, even if it were compelling in some cases, it would not have force with respect to moral concepts. ‘Tree’ is an ordinary, descriptive concept based on direct observation. In view of this, mistranslation seems more likely than substantial disagreement. But what about concepts concerning what is amusing, interesting, or exciting? These have to do with human reactions to the world, and it may be said that our knowledge of human nature suggests that some reactions vary widely. A claim that there is much disagreement about what people find amusing—about what makes them laugh—does not immediately generate the suspicion of mistranslation. If moral concepts were more similar to ‘amusing’ than to ‘tree’, as some believe, then the Davidsonian argument might not undermine DMR even if it were convincing in other cases. Davidson, however, believed the argument applies across the board, to evaluations as well as empirical beliefs. Another response to his argument is to claim that, even if it does apply to evaluations, it would only apply to very basic ones and would leave room for substantial disagreements beyond these (if this were the case, then Davidson would have established only what I call a mixed position in section 7). For some critical responses to the Davidsonian critique of relativism, see Gowans 2004: 144-6, Prinz 2007: 195-9 and Rovane 2013: 247-62.
Another a priori objection to DMR was suggested by Philippa Foot (1978a and 1978b) in a response to emotivism. Just as there are shared criteria of ‘rude’ such that not just anything could be considered rude, she argued, there are shared criteria of moral concepts such that not just anything could be a moral virtue or obligation. For example, there are substantial constraints on what could be considered courage. Hence, there are significant limits to the extent of moral disagreements.
One response to this argument, interpreted as an objection to DMR, is that it faces a dilemma. On the one hand, if ‘courage’ is understood broadly, in terms of confronting a difficulty to achieve some perceived good, then it is likely that most everyone values courage. However, this leaves room for very different conceptions of courage. Both warriors and pacifists may value it, but they may regard very different kinds of actions as courageous. This puts less pressure on DMR, a point Foot later conceded to some extent (see section 7)). On the other hand, if courage is defined narrowly, for example, as the virtue of a warrior who faces the threat of death in battle (as suggested by Aristotle), then there may be little disagreement about the scope of the concept, but considerable disagreement about whether courage so-defined should be valued (pacifists would say no). A proponent of DMR might say that this is also a significant moral disagreement. Against this, it may be said that our understanding of human nature and culture shows that everyone values courage understood within some fairly significant limits. This is a more empirical point, in line with the objections in the last paragraph of this section.
Some versions of the a priori approach emphasize the constraints imposed by “thinner” moral concepts such as goodness, rightness, or morality itself (for example, see Garcia 1988). Once again, a defender of DMR might say that, if these concepts have enough content to preclude significant disagreement in their application, then it is likely that many societies do not apply them at all—a form of moral disagreement in itself. Another response would be to argue, following R.M. Hare (1981), that a formal analysis, for example in terms of a kind of prescriptivity, is plausible with respect to some thinner moral concepts, and that this is consistent with significant moral disagreements. However, the a priori critics question the adequacy of any such analysis. Much of this debate concerns the acceptability of formal versus material definitions of morality (see the entry on the definition of morality).
The second approach to rejecting DMR focuses on the interpretation of the empirical evidence that purportedly supports this thesis. Some objections point to obstacles that face any attempt to understand human cultures empirically. For example, it may be said that the supposed evidence is incomplete or inaccurate because the observers are biased. In support of this, it may be claimed that anthropologists often have had preconceptions rooted in disciplinary paradigms or political ideologies that have led them to misrepresent or misinterpret the empirical data. Or it may be said that even the most objective observers would have difficulty accurately understanding a society's actual moral values on account of phenomena such as self-deception and weakness of will. These concerns point to substantial issues in the methodology of the social sciences. However, even if they were valid, they would only cast doubt on whether DMR had been established: They would not necessarily give us reason to think it is false. Of course, this would be an important objection to someone who claims DMR is established or relies on DMR to argue for MMR.
Another objection, more directly pertinent to DMR, is that anthropologists have tacitly and mistakenly assumed that cultures are rather discrete, homogenous, and static entities—rather like the shapes in a Piet Mondrian painting or a checkerboard. In fact, according to this contention, cultures typically are rather heterogeneous and complex internally, with many dissenting voices. Moreover, they often interact and sometimes influence one another, and they may change over time. From this perspective, the world of cultures is closer to an animated Jackson Pollock painting than to the unambiguous configuration suggested by the first image. If these contentions were correct, then it would be more difficult to know the moral values of different cultures and hence to know whether or not DMR is true. As before, this would not show that it is false (in fact, the point about heterogeneity might point the other way). However, we will see later that these contentions also pose challenges to MMR.
Other critics try to establish that the empirical evidence cited in support of DMR does not really show that there are significant moral disagreements, and is consistent with considerable moral agreement. A prominent contention is that purported moral disagreements may result from applying a general moral value (about which there is no disagreement) in different circumstances or in the same circumstances where there is a factual disagreement about what these circumstances are. Either way, there is no real moral disagreement in these cases. For example, everyone might agree on the importance of promoting human welfare (and even on the nature of human welfare). But this may be promoted differently in different, or differently understood, circumstances. Another contention is that moral disagreements may be explained by religious disagreements: It is only because specific religious assumptions are made (for instance, about the soul) that there are moral disagreements. Once again, the apparent moral disagreement is really a disagreement of a different kind—here, about the nature of the soul. There is no genuine moral disagreement. Of course, these possibilities would have to be established as the best explanation of the disagreements in question to constitute an objection to DMR.
Finally, some objections maintain that proponents of DMR fail to recognize that there is significant empirical evidence for considerable moral agreement across different societies. Several kinds of agreement have been proposed. For example, the role-reversal test implied by the Golden Rule (“Do unto others as you would have them do unto you”) has been prominent beyond Western traditions: A version of it is also endorsed in The Analects of Confucius, some traditional Buddhist texts, and elsewhere (see Wattles 1996). Another form of this claim maintains that basic moral prohibitions against lying, stealing, adultery, killing human beings, etc. are found across many different and otherwise diverse societies. Yet another contention is that the international human rights movement indicates substantial moral agreement (see Donnelly 2013: ch. 4). On the basis of evidence of this kind, some such as Sissela Bok (1995) and Michael Walzer (1994) have proposed that there is a universal minimal morality, whatever other moral differences there may be. In a similar vein, Hans Küng (1996) and others have maintained that there is a common “global ethic” across the world's major religious traditions regarding respect for human life, distributive justice, truthfulness, and the moral equality of men and women. These contentions, which have received increased support in recent years, must be subjected to the same critical scrutiny as those put forward in support of DMR. However, if they were correct, they would cast doubt on DMR. In the final analysis, there may be significant agreements as well as disagreements in people's moral values. If this were the case, it would complicate the empirical background of the metaethical debate, and it might suggest the need for more nuanced alternatives than the standard positions.
5. Are Moral Disagreements Rationally Resolvable?
Philosophers generally agree that, even if DMR were true without qualification, it would not directly follow that MMR is true. In particular, if moral disagreements could be resolved rationally for the most part, then disagreement-based arguments for MMR would be undermined, and there would be little incentive to endorse the position. Such resolvability, at least in principle, is what moral objectivism would lead us to expect. One of the main points of contention between proponents of MMR and their objectivist critics concerns the possibility of rationally resolving moral disagreements. It might be thought that the defender of MMR needs to show conclusively that the moral disagreements identified in DMR cannot be rationally resolved, or again that the moral objectivist must show conclusively that they can be. Neither is a reasonable expectation. Indeed, it is unclear what would count as conclusively arguing for either conclusion. The center of the debate concerns what plausibly may be expected. Adherents of MMR attempt to show why rational resolution is an unlikely prospect, while their objectivist critics try to show why to a large extent this is likely, or at least not unlikely.
Moral objectivists can allow that there are special cases in which moral disagreements cannot be rationally resolved, for example on account of vagueness or indeterminacy in the concepts involved. Their main claim is that ordinarily there is a rational basis for overcoming disagreements (not that people would actually come to agree). Objectivists maintain that, typically, at least one party in a moral disagreement accepts the moral judgment on account of some factual or logical mistake, and that revealing such mistakes would be sufficient to rationally resolve the disagreement. They suggest that whatever genuine moral disagreements there are usually can be resolved in this fashion. In addition, objectivists sometimes offer an analysis of why people make such mistakes. For example, people may be influenced by passion, prejudice, ideology, self-interest, and the like. In general, objectivists think, insofar as people set these influences aside, and are reasonable and well-informed, there is generally a basis for resolving their moral differences. (They might also say that at least some agreements about moral truths reflect the fact that, with respect to matters pertaining to these truths, people generally have been reasonable and well-informed.)
Proponents of MMR may allow that moral disagreements sometimes are rationally resolved. In particular, they may grant that this often happens when the parties to a moral dispute share a moral framework. The characteristic relativist contention is that a common moral framework is often lacking, especially in moral disagreements between one society and another, and that differences in moral frameworks usually cannot be explained simply by supposing that one society or the other is making factual or logical mistakes. These moral disagreements are ultimately rooted in fundamentally different moral orientations, and there is usually no reason to think these differences result from the fact that, in relevant respects, one side is less reasonable or well-informed than the other. They are faultless disagreements. This conclusion might rest on the observation that it is not evident that mistakes are at the root of these disagreement. But it might also depend on a theory, developed to explain such observations, that the frameworks are incommensurable: They do not have enough in common, in terms of either shared concepts or shared standards, to resolve their differences, and there is no impartial third standpoint, accessible to any reasonable and well-informed person, that could be invoked to resolve the conflict.
Various objectivist responses may be made to this argument. One is the Davidsonian approach, already considered, that precludes the possibility of incommensurable moral frameworks. Another response is that incommensurability does not preclude the possibility of rationally resolving differences between moral frameworks. For example, Alasdair MacIntyre (1988: ch. 18 and 1994) has argued that, in some circumstances, it is possible to realize, through an exercise in imagination, that a conflicting and incommensurable moral tradition is rationally superior to one's own tradition. However, the most common objectivist response is to claim that some specific moral framework is rationally superior to all others. For example, it might be argued, following Kant, that pure practical reason implies a fundamental moral principle such as the Categorical Imperative (see Kant's moral philosophy), or it might be claimed, following Aristotle, that human nature is such that virtues such as courage, temperance, and justice are necessary for any plausible conception of a good life (see the sections on the human good and the function argument in the entry on Aristotle's ethics, and the entry on virtue ethics). If such an argument were sound, it might provide a compelling response to the relativist contention that conflicts between moral frameworks cannot be rationally resolved.
Proponents of MMR are unimpressed by these responses. They may say that the Davidsonian account cannot assure sufficient common ground to resolve conflicts between moral frameworks (or to ensure that there is really only one framework), and that MacIntyre's approach is likely to work at best only in some cases. And they usually consider debates about the Kantian and Aristotelian arguments to be as difficult to resolve rationally as the conflicts between moral frameworks the relativists originally invoked. They may add that the fact that moral objectivists disagree among themselves about which objectivist theory is correct is further indication of the difficulty of resolving fundamental moral conflicts.
A rather different objectivist challenge is that the position of the proponent of MMR is inconsistent. The relativist argument is that we should reject moral objectivism because there is little prospect of rationally resolving fundamental moral disagreements. However, it may be pointed out, the relativist should acknowledge that there is no more prospect of rationally resolving disagreements about MMR. By parity of reasoning, he or she should grant that there is no objective truth concerning MMR.
To this familiar kind of objection, there are two equally familiar responses. One is to concede the objection and maintain that MMR is true and justified in some metaethical frameworks, but not others: It is not an objective truth that any reasonable and well-informed person has reason to accept. This may seem to concede a great deal, but for someone who is a relativist through and through, or at least is a relativist about metaethical claims, this would be the only option. The other response is to contest the claim that there is parity of reasoning in the two cases. This would require showing that the dispute about the irresolvability of moral disagreements (a metaethical debate) can be rationally resolved in a way that fundamental moral disagreements (substantive normative debates) themselves cannot. For example, the metaethical debate might be rationally resolved in favor of the relativist, while the substantive normative debates cannot be resolved.