Monism, pluralism, and the structure of value relations

This work is licensed under the Creative Commons | © Joel Robbins. Attribution-NonCommercial-NoDerivs 3.0 Unported. ISSN 2049-1115 DOI: http://dx.doi.org/10.14318/hau3.1.006

Monism, pluralism, and the structure of value relations

A Dumontian contribution to the contemporary study of value

Joel ROBBINS, University of California, San Diego

Although the topic of values has not been a central focus of discussion in anthropology in recent decades, during this period questions concerning values have continued to be important in philosophy. One key debate surrounding values in that field takes up the question of whether monist or pluralist accounts best describe the way values relate to one another in the world. Reviewing some of the philosophical literature on this topic, I argue that it is primarily concerned not with how many values exist in any given society, but with the nature of the relations between them. Drawing on Dumont’s theoretical work, I suggest that ethnographic research demonstrates that both monist and pluralist tendencies exist in the value relations of all societies and that the key analytic task thus becomes not determining whether a society is monist or pluralist, but rather documenting which kinds of configurations of monist and pluralist relations we tend to find in actually existing societies. I present four ethnographic sketches of different configurations, demonstrating the promise of this kind of research for contributing to both anthropological research and philosophical debates about value.

Keywords: values, value pluralism, monism, Dumont, anthropological theory

Anthropology is only now returning to the serious study of values after a long period of neglect.1 Until the recent rebirth of interest in the subject, Louis Dumont was the last prominent anthropologist to direct significant attention to values, and by the end of his life he did so in something close to isolation; beyond the circle of his own students and others who followed him closely, the nature and cultural role of values were not topics of great interest in the discipline. During the period in which anthropologists neglected the study of values, however, it remained a focal concern in other areas of scholarship. In particular, questions concerning values continued to be important in philosophy. One of the key debates surrounding value in that field takes up the question of whether monist or pluralist accounts of value best describe the way values relate to one another in the world. In this article, I want to look at some aspects of the philosophical discussion around value pluralism and value monism and consider what kinds of issues it might raise for anthropologists. In doing so, I will put these philosophical issues in dialogue with Dumont’s work, and another of my goals is to show how Dumont’s approach to the study of values can be used to address contemporary questions with which he himself did not appear much concerned.

Value monism and value pluralism

I begin by introducing some aspects of the philosophical debate between those who promote monist positions and those who promote pluralist ones. On a broad level, this is not difficult to do. The debate, which is important primarily in moral and political philosophy, turns on whether the values of a society can work harmoniously together or whether by nature they are destined to conflict with one another. Monists hold that the values of a society work smoothly with one another. In this view, values are, as Plaw (2004: 109) puts it in passing, “harmoniously integrated.” This harmony can come about because, as Galston (2002: 6) phrases matters, there is “a comprehensive hierarchy or ordering among goods,” or, as in Dworkin’s (2011: 10) major recent formulation of the monist position, each important value supports the realization of all others, or, finally, because all the values we might recognize as existing in fact reduce to one “supervalue,” such as pleasure, because each gains its own value by contributing to the realization of that supervalue (Chang 2001). Pluralists, by contrast, assert that more than one worthy value exists, and some of these worthy values conflict with one another, such that pursuing one means failing to pursue the others. Isaiah Berlin and Max Weber are well-known pluralists (Lassman 2004). Berlin (1998: 239) has famously given voice to this position by stating,

If, as I believe, the ends of men are many, and not all of them are in principle compatible with each other, then the possibility of conflict—and of tragedy—can never be wholly eliminated from human life, either personal or social. The necessity of choosing between absolute claims is then an inescapable characteristic of the human condition.

In Weber’s (1946: 147) well-known and more concise phrasing, the pluralist position rests on the claim that “the various value spheres [or values] of the world stand in irreconcilable conflict with each other.” For pluralists, there is no way, or at least no legitimate, non-totalitarian way, in which value monism can be achieved such that the plurality of existing values does not lead to the possibility of conflict between them.

One of the classic cases used to illustrate the pluralist position is the sometimes conflicting relationship between the values of liberty and equality in the liberal tradition. In this discussion, it is assumed that people agree that both liberty and equality are very valuable. But it is often the case that realizing more of one of these values results in realizing less of the other. So, for example, redistribution from the more well off to the less well off realizes more equality (at least as this can be defined in economic terms), but arguably reduces the liberty of more well off people to use their wealth to pursue any ends they might choose beyond sharing their wealth with the less fortunate. Turning the argument around, increasing the realization of the value of liberty of the wealthy by letting them keep more of their wealth and deploy it as they wish reduces the realization of the value of equality in the society in which they live. On the pluralist account, the dilemmas presented by the conflict between liberty and equality are irresolvable. Choices have to be made that will result in realizing less of one or the other value, despite the fact that it would be desirable to realize both as fully as possible. For pluralists, there are a number of worthy values that conflict in this way (justice and mercy are another pair that philosophers often put forward by way of illustration), and the need to make choices that result in the loss of value (choices Berlin refers to as “tragic” in the quotation above) is an unavoidable aspect of human life.

Monists have a sunnier outlook on the relations between values. For them, value conflicts are not irresolvable, and the choices we face between them are not tragic. If one holds that all values are ultimately reducible to a supervalue (such as pleasure or happiness), then one can make one’s choice on the basis of a determination of which lower-level value contributes most to realizing the supervalue. So, for example, if happiness is the ultimate or supervalue, the choice between liberty and equality in the cases presented above would be resolvable by determining whether increasing liberty or equality would produce the most happiness. A slightly different version of monism is based not on the claim that all values are reducible to a supervalue, but that all important values work together such that the realization of one supports the realization of the others. Dworkin (2011) presents a complex argument that equality and liberty, properly understood, support each other in this way. I do not have space to present his argument in detail here, but by way of illustration, consider a simpler version of this kind of monist account. Suppose that we say that of all the kinds of equality we can imagine, the kind that matters most is equality of opportunity. Equality of opportunity, we can go on to argue, clearly supports liberty by giving everyone an equal chance to live their lives as they see fit. If we restrict our understanding of equality as a value to equality of opportunity, then, there should not arise occasions in which the realization of equality conflicts with realization of liberty. Both values support one another.2

The stakes of arguments between pluralists and monists can be quite high within moral and political philosophy themselves. For example, pluralism can be seen to raise significant problems for certain popular forms of utilitarianism, which are built around the assumption that it is possible to posit supervalues, and of rational choice theory, which assumes that it should always be possible to make reasoned rather than tragic choices between values (Chang 2001: 16140). For its part, value monism is often seen as lending support to totalitarianism (Lassman 2011; Muller 2012), and as according only poorly with people’s widespread intuition that life often does require more or less tragic choices between values that do conflict (Chang 2001).

Some of these stakes will not appear to be worth much to contemporary cultural anthropologists, who are unlikely to worry much over a loss of plausibility on the part of utilitarian and rational choice arguments. It may therefore seem eminently reasonable to us that, as is widely recognized, monism is currently largely out of fashion (Lassman 2011: 5; Chang 2001: 16139). Yet, it remains the case that the debate between these two positions cannot at all be taken to be definitively settled. And value pluralism itself, as Lassman (2011: 5) points out, is a very broad designation that “does not refer to the proposition of a clearly defined theory about values but instead to a set of overlapping concerns about the nature of the world of modern politics and the problem of how best to make sense of it.” So even if in general terms something we might call “value pluralism” is dominant at the moment, it remains an open question precisely what version of this position it promotes. I take this generally unsettled situation in philosophical debates about values as an opening to bring anthropology into this conversation in ways that I hope might not only enrich anthropological analysis in some useful ways, but also add some new perspectives to philosophical debates that are already underway.

Dumont on the nature of value relations

One thing that is quite clear in the philosophical debate between pluralists and monists is that the core issues involved do not have to do with determining how many different kinds of values there are in the world (although at one extreme, monists who posit a single supervalue can come close to arguing that there is really only one). Rather, these issues pertain to the nature of the relations between values. As Lassman (2011: 10) puts it for the pluralist side, “The basic problem of value pluralism for political theorists is not so much their plurality, but the implication of their supposed incommensurability and incomparability. These two properties are… held to be responsible for the persistence of disagreement.” Pluralists also hold that it is such relationships of incommensurability and incomparability that lead to the need to make tragic choices. On the monist side, it is relations of reducibility or mutual supportiveness that link different values to one another. Determining which kinds of relations between values really do occur in the world is thus at the heart of these debates.

My suggestion is that anthropologists can approach the nature of these relations as an empirical question, and that Dumont has presented us with important tools for doing so. Indeed, Dumont’s most famous claim about values—that higher-ranked values encompass their lower-ranked contraries—is a claim about value relations. As such, it is highly relevant to the monist-pluralist discussion.

In what Dumont (1980: 240) calls hierarchical oppositions, the value relation in question is one between, as he puts it, a set or whole and an element of that set or whole. In his most well-known illustration of this kind of opposition, he turns to the example of the notion of “man” as it is deployed in the English and French languages. In this case, “man” can stand for humankind in its entirety, which is the whole in question. But it can also stand for “man” in opposition to “woman.” In Dumont’s (1980: 240) analysis, “woman” is both a member of the whole referred to by “man,” and hence has the same value as the whole—or, as Dumont often puts it, is “identical” with the whole—and is also the opposite of “man,” and as such is less valued than the latter. This double value relation, whereby “man” and “woman” are both of equal value and “woman” is inferior to “man,” is possible because hierarchical oppositions distinguish different ideological levels, such that at the superior level, which is also that of the whole, the whole encompasses its parts, including them in its value, while at an inferior level the whole and the parts are opposed to one another, with the whole generally (although see the discussion of reversals below) being more valuable than the part (see Otto and Willerslev’s introduction). The distinctions made between levels in this way is not a contingent outcome of hierarchical oppositions, but is rather one of their necessary features: for Dumont (1980: 244), “hierarchy assumes the distinction of levels.”

Having laid out Dumont’s key notion of hierarchical opposition, one can go on to ask if it represents a monist or a pluralist understanding of value relations. I believe that anthropologists have largely interpreted it as a monist understanding, and this has been a key reason that so many of them have dismissed his work during the current age of pluralist predominance. And one can see how people who do not read Dumont closely can come to this understanding. Encompassment of the contrary, after all, looks like nothing so much as the reduction of a lower-ranked value to a higher-ranked one. Combining this with Dumont’s occasional reference to “paramount” values that represent the most overarching whole in play in a given society, one can easily be led to assume that ultimately all other values must be encompassed by and hence reduced to this one supervalue.3 Yet, even as this account of Dumont as a monist has a superficial kind of plausibility, I think it is fundamentally in error. Dumont is best understood not as promoting monism, but as developing the notion of hierarchical opposition in such a way as to capture and help explain what can be seen as the routine existence of both monist and pluralist tendencies in all societies. This is true formally, in terms of his theoretical contributions, and also in terms of his analyses of empirical materials. In the remainder of this section, I will demonstrate these points in turn.

In theoretical terms, we can begin with Dumont’s (1980: 239) statement that hierarchy is not, among other things, “a chain of beings of decreasing dignity, nor yet a taxonomic tree.” Both of the potential forms of hierarchy Dumont rejects here are straightforwardly monist ones. Chains in which successive elements are distinguished by decreasing amounts of a single valued feature are clearly organized by a single value, and taxonomic trees suggest that one can always work reductively back up through the distinctions they make to reach a singular starting point. Already, then, at the outset of his most important discussion of the nature of hierarchical opposition, Dumont has signaled his intention to avoid a wholly monist interpretation of this phenomenon. The challenge to such monism follows from the existence of differently valued levels in every hierarchical opposition.

On such differently valued levels, different values can be preeminent. This is why one routinely finds value reversals in social life, such that what is an inferior value in some contexts is superior in another. “The mother of the family (an Indian family, for example),” Dumont (1980: 241) writes, “inferior though she may be made by her sex in some respects nonetheless dominates the relationships within the family” (see also Dumont 1986: chaps. 8 and 9). Because of the existence of value reversals between levels in every opposition (or, put otherwise, in every whole), one cannot always move reductively between the values of the parts and the value of the whole. The value of a woman at the level of the family cannot, in the Indian context, be simply reduced to the value of a woman on a superior level. To quote Dumont (1980: 244, emphasis added):

As soon as we posit a relation of superior to inferior, we must become accustomed to specifying at what level this hierarchical relation is situated. It cannot be true from one end of experience to the other (only artificial hierarchies make this claim), for this would be to deny the hierarchical dimension itself, which requires situations to be distinguished by value.

Put otherwise, only “artificial” hierarchies without levels support value monism. When one recognizes the effect of levels on value relations in hierarchies that are not artificial, it is clear that Dumont operates theoretically with a complex notion of value pluralism to complement those cases of monism he might recognize.4

And yet, it remains true that Dumont’s pluralism is rather unique in that it allows room for a kind of monism as well. It is Dumont’s definition of the most valued level of a hierarchal opposition as the “whole” that leaves open the possibility of a monist complement to the pluralist picture I have just painted. At this highest level, where the whole encompasses its contrary or contraries (the existence of which make values irreducibly plural when lower levels are distinguished), monism of a sort might be seen to hold sway. Moreover, Dumont sees people as driving toward such monist encompassment as they seek always higher levels that can provide meaningful contexts for lower ones. As he puts it at the conclusion of the postface to Homo hierarchicus, the piece of his I have been following most closely here, “transcendence” toward the whole is always present “at the heart of social life” (Dumont 1980: 245).5 But the very movement toward transcendence can only be accomplished by way of constructing a hierarchical opposition that will again generate a plurality of values and of levels on which they appear, and thus this movement toward monism itself produces pluralism. In the end, then, Dumont leaves us in theoretical terms with a model in which pluralism and monism are both fundamental features of social life. This should, I want to suggest, lead to a charge to study the various kinds of interplay and arrangement of monist and pluralist features we find in the societies we study. Before going on to consider through some examples how we might analyze ethnographic materials in these terms, let me briefly argue that in his analytical practice, Dumont also carries out this kind of analysis.

If those inclined to read Dumont as a monist tend to examine his theory by focusing on the notions of paramount values and encompassment, their primary case in relation to his analytic practice would have to be Homo hierarchicus (1980). In this book, Dumont appears to give as exhaustive as possible an account of Indian ideology from the point of view of the value of purity alone. Although I think it is possible to do so, I will not dispute this reading of Homo hierarchicus here, instead leaving debate over that work to regional specialists. I will point out, though, that in all of the major works Dumont produced focused on Western ideology, value pluralism and sometimes value conflict is to the fore. Inasmuch as these works take up the way in which holist and individualist values have remained in relationship with one another throughout the history of the West, this could not be otherwise. In his last book, published in English as German ideology: From France to Germany and back (1994), the pluralism at the heart of Western societies is an explicit concern. Dumont argues in the first chapter of that work that the value of individualism can never on its own organize society, as society of necessity rests on some valuation of holism (Dumont 1994: 8). This means that all individualist societies will be marked by struggles between at least these two values, a pluralist assertion that is borne out in Dumont’s analyses of the development of both German and French ideologies. Here, Dumont also asserts more generally that by the end of the twentieth century, such tensions between holism and individualism have become present in almost all of the world’s societies. His accounts of German and French versions of the historical development of these value conflicts thus together stand as a model for how to analyze cultural pluralism more generally in the contemporary world (Robbins and Siikala 2013).

My own concern in what follows is less with following Dumont in highlighting the clash between individualist and holist values than it is with taking up the more general task his work inspires of exploring the various ways monism and pluralism can relate to each other in any given society. What is the range of variation we can expect to find between more pluralist and more monist social worlds? The case studies that follow sketch some answers to this question. They are not meant to provide an exhaustive typology of monist/pluralist configurations, but because my exposition moves from a largely monist case through two mixed cases and then concludes with a strongly pluralist case, it might be taken to at least map some key points on a continuum of configurations that I hope to suggest is worthy of further exploration.

Monism and pluralism: Four cases

Case one: Monism6

The first case I consider is that of the Bobover and unaffiliated Hasidim of Brooklyn, as studied by Ayala Fader (2009) in her book Mitzvah girls: Bringing up the next generation of Hasidic Jews in Brooklyn. The ideology of these communities comes as close as any case I have seen to realizing what would count in the theoretical terms I have laid out here as a strong monism in a Dumontian sense (that is, a monism that does not fail to recognize values other than its paramount one, nor to assign them levels of their own, but that appears wholly to subordinate all these other values and their levels under a single paramount one). The value that organizes Hasidic monism is that of the redemption of the Jewish people (Fader 2009: 87). The social activity that most directly realizes this value and that is thus itself most highly valued is male worship and study of religious texts. Hasidic life is organized so as to give men as much freedom as possible to pursue these activities. Women also participate in some forms of worship, but they are also charged with acting within the secular world of Brooklyn so as to meet the needs of their families (Fader 2009: 22). By managing this world on behalf of their families, women allow men both to have more time for worship and study and to forgo corrupting contact with the world beyond the Hasidic community (Fader 2009: 120). There are clearly two hierarchically arranged levels here: a sacred one in which men are dominant, and a subordinated secular one in which women are seen to operate most effectively. There is no doubt, even from the point of view of women’s actions in the secular world, about which value is dominant and which level is most crucial to its realization. This is what makes the monism of the Hasidic ideology so straightforward.

Fader illustrates the working of this monism most fully through her analysis of Hasidic language practices. The Hasidic people she studied use three languages. The most highly valued is Loshn-koydesh (“holy language”), a mixture of Ancient Hebrew and Aramaic that is the language of the Torah, prayers, and the early rabbinic literature (Fader 2009: 87). This language is used primarily by men in study and worship, although women also use it in prayer. The everyday vernacular of men is Yiddish, a language also identified with Judaism and of some sacred value, although it is not as highly ranked as Loshn-koydesh. Men’s use of Yiddish and their relative lack of fluency in English serve as markers of their commitment to the realization of the community’s highest value of Jewish salvation. Women, because they must engage with the secular world, speak Yiddish much less often than men do. Instead, their most common everyday language is English, the least valued language. Even as this language is appropriate to use in contexts that exist on the secular level, English has no direct link to the value of Jewish salvation, and thus people condemn its use in all other contexts.

After laying out the Hasidic hierarchy of languages and their gendered patterns of use in the way I have just summarized, Fader goes on to present a fascinating account of linguistic mixing between these codes. I cannot reproduce her analysis in detail here, but I want to sketch its broadest contours because it shows how the force of Hasidic monism reverberates throughout the levels of Hasidic ideology. As noted, in everyday contexts within the Hasidic community—contexts that are ranked below religious ones, but that are still highly valued—men speak Yiddish. To a limited extent, women also speak Yiddish in these contexts. When men or women speaking Yiddish in these contexts need to borrow English words, they are encouraged to pronounce them with a Yiddish accent. This allows speakers to “religiously uplift the English words they need,” thereby demonstrating the pull of the paramount value of achieving salvation on an everyday level that in other respects acknowledges the existence of more mundane values (Fader 2009: 94). Women, as already noted, most often speak English in everyday contexts. But their English is similarly distinctive for the way in which they borrow Yiddish intonation and phonology, and for their tendency to pepper their English speech with Yiddish words and phrases (Fader 2009: 101–3). In this way, “they adapt and redeem” English by making it “sound… Hasidic” (Fader 2009: 116). In this case as well, we see the paramount value of Jewish salvation reaching down into lower-ranked levels to exert its influence. Given the consistency with which Hasidic monism allows for this expression of its highest value on all levels of its ideology, it is perhaps not surprising that a recent push in the community for even greater religious stringency has seen men begin to ask that women make Yiddish their sole vernacular in everyday contexts within the home, thus reserving English only for worldly contexts in which its use is absolutely necessary. To this point, the campaign to increase Yiddish use among women has not been successful. But the very fact that men have pushed women in this direction again indicates the force of the pull of the preeminent value throughout all levels of the Hasidic ideology—a force that renders this ideology as strongly monist as it is possible to find within the Dumontian framework I have adopted here.

Case two: Monism with stable levels

The second case I consider is drawn from Douglas Rogers’s (2009) historical ethnographic study of Priestless Old Believers in Sepych, a town of fourteen hundred people in the Russian Urals. The Priestless Old Believers are a schismatic group that long ago broke away from the Russian Orthodox Church. Rogers (2009: 45–46, 299) traces a longstanding “ethical regime” among the Old Believers (as I will refer to them from this point forward) in Sepych that held from the mid-nineteenth century until the recent postsocialist period. Central to this ethical regime was a distinction between what Rogers (2009: 67, 91, 299), following Bourdieu, refers to as two distinct ethical “fields.” In the framework of my discussion here, these fields can readily be understood as levels in a value hierarchy. One level involves worldly withdrawal into prayer aimed at fulfilling “collective and personal aspirations for salvation from a fallen world” (Rogers 2009: 42). The other is oriented toward getting a living through labor and exchange— activities that necessitate engagement in the very world the first level defines as wholly lacking in value. Laid out side by side in this way, this is a clear case of value pluralism, with what Rogers (2009: 90–91) refers to as the values of “work” and those of “prayer” set in sharp opposition to one another.

But the Old Believers did not, of course, simply lay these levels side by side. Instead, I would argue, they ranked them hierarchically. At this point, my argument makes claims that accept and draw on but also go slightly beyond those that Rogers makes himself. Let me therefore start by laying out Rogers’s discussion. As he shows in impressive detail, the way in which the Old Believers of Sepych lived with their two distinct values of worldly engagement and worldly withdrawal, both of which they deemed very important, was to define one value as appropriate to the elderly and the very young, and the other as appropriate to those in middle age. In this distribution of values, the elderly were charged with pursuing religious salvation through prayer, and with training the young in religious beliefs and practices. Those in their middle years were the ones expected to pursue the values of the world, producing and exchanging material goods. Both values were highly respected on their own levels, but Old Believers took care not to mix them. Older people, in particular, removed themselves from worldly spheres not only of work, but also of exchange and consumption, and practiced an asceticism that left their commitment to working toward salvation unsullied by other concerns (Rogers 2009: 67). Already we can see the way in which the Old Believers have solved the problem of how to hold two conflicting values without allowing their incommensurability to foment social conflict: parceled out to distinct social categories, in crucial respects the two values never needed to clash in the lives of a single person or social group.

Where I would go beyond Rogers to some extent is in making the argument that the two values are not only sharply distinguished in their social distribution, but are also ranked vis-à-vis one another, with the religious level ultimately ranked above that of work. My primary basis for this claim is the ideal Old Believer life cycle, which at its beginning and crucially at its end is devoted to religious values. Everyone must be trained in these, and the pursuit of these religious values will end up as the project of everyone who lives out a full life. Thus, the world of work is ultimately one a person moves through, not an ultimate one at which anyone should expect to rest. On this reading, the level of work is a subordinate one in the hierarchy. But unlike in the Hasidic case, the values of industry and worldly engagement that this subordinate level promotes appear to be comfortably sovereign within its confines. There is no pressure from the religious level to remake worldly values along sacred lines to whatever extent might be possible. This is why I have referred to this as a case of monism with stable levels. It comes the closest of the cases I will consider to realizing Dumont’s ideal formal understanding of hierarchical opposition, in which a clearly preeminent highest value dominates the ideology but nonetheless accords room for other values to express themselves extensively on lower-ranked levels. The long-running stability of this organization of values throughout a significant period of Sepych’s otherwise tumultuous history perhaps provides some indication of the extent to which this kind of value relation leads to a way of structuring social life that people find in some ways manageable, or even perhaps gratifying.

Case three: (Relatively) stable pluralism

My third case is that of Avatip, a Sepik River community located in Papua New Guinea’s East Sepik Province (Harrison 1985a: 413). In a pair of now-classic articles, Harrison (1985a, 1985b) explores the opposed values that organize what he calls the “ritual” and the “secular” domains in Avatip. As with Rogers’s ethical fields, so too Harrison’s ritual and secular domains can be understood as different levels of an ideology, with the ritual domain promoting the value of hierarchical relations and the secular one devoted to the value of equality. What makes the Avatip case distinctive among those examined here is that there does not appear to be a clear hierarchical relationship between the levels involved.

As was at least formerly the case in many Sepik societies, the religious level in Avatip is centered around a male initiation cult devoted to maintaining the cosmic order and asserting men’s strength as warriors (Harrison 1985a: 417). The key value that organizes the men’s efforts is one Harrison calls “ritual hierarchy.” All important practices of the men’s cult represent the cosmos and the social world as hierarchical, and they place Avatip adult men at the top of the human portion of this hierarchy, above Avatip women and younger men, and above their enemy neighbors. This emphasis on hierarchy is reflected, for example, in the organization of the cult into a number of different initiatory levels. As they move through these levels, boys are taught sacred knowledge that they must keep secret from women and younger boys, thus asserting their ritual priority over them. As befits Avatip ritual’s general commitment to the realization of the value of hierarchy, actions in the ritual sphere are often aggressive and self-assertive.

In contrast to the ritual domain, the secular sphere in Avatip is structured by the value of equality. Everyday life, governed by this value, is marked by sensitivity to others and an interest in achieving equivalence through reciprocity in “marriage negotiations, exchanges of agricultural labor, the productive cooperation of spouses, and so forth” (Harrison 1985a: 419). Relations between the sexes in marriage are markedly supportive, with both men and women understood to contribute significantly to the maintenance of the household (Harrison 1985a: 416). Assertions of hierarchical dominance thus have little place in secular everyday life in the Avatip community.

One of Harrison’s most original claims in these articles is that we should not see ritual hierarchy as a male value and secular equality as a female value, distributed across social categories in the fashion of the values of engagement and withdrawal among the Old Believers. Instead, they are distinctive levels of Avatip ideology, the legitimacy of the values of which both men and women recognize (see also Strathern 1988). When acting on the secular level, both men and women usually realize the value of equality. Similarly, women support men’s hierarchical projects on the ritual level, and in some contexts also express hierarchical power of their own (Harrison 1985b: 121). All people in Avatip thus live their lives in relation to these two distinctive values and the levels that they organize.

Although the ritual and secular levels are distinct from each other in Avatip, it is not clear from Harrison’s ethnography that a stable hierarchical relationship exists between them. The validity of this interpretation of the Avatip situation is supported by Harrison’s (1985a: 126) observation that the “conflicting demands of these different spheres… lie at the root of a moral or existential tension which, I think, many men keenly if inarticulately feel, between the ideals” of the ritual and the secular levels. Harrison writes only about men here, but the struggle he describes them facing is precisely the kind we would expect to arise in a situation in which the hierarchical relationship between conflicting values is not managed by means of ranked levels, and thus people must sometimes reckon with their incommensurability.

Without more information than this passing remark, we cannot go further in exploring the effects of pluralist conflict in Avatip society. But it is worth noting that among the Ilahita Arapesh, another Sepik society, post-contact changes led one group in which men felt a similar conflict between the values of ritual hierarchy and secular equality to abandon the male cult system entirely in favor of a Christian revival movement that supported more egalitarian values on all levels, thus easing this particular value conflict through a clear hierarchical resolution of the tensions it produced (Tuzin 1982, 1997). I should note, however, that nothing in Harrison’s discussion predicts such a dramatic response to value conflict in the Avatip case, and this is why the pluralism he describes appears to be a relatively stable one.

Case four: Unsettled pluralism

My final case is that of the Urapmin of Papua New Guinea (PNG). The Urapmin are a group of approximately 390 people living in the West Sepik Province. Since 1970, when they experienced a mass wave of conversions during a charismatic Christian revival movement that came to them as it moved throughout many parts of PNG (the Arapesh mentioned in the last section also converted during this revival), all Urapmin have understood themselves to live in a completely Christian community. I have previously analyzed the Urapmin conversion to Christianity as in important respects involving the adoption of a new paramount value without fully jettisoning an older one, and I have described the conflict-ridden situation of value pluralism that resulted (Robbins 2004, 2007). In this section, I will just summarize the Urapmin case briefly, highlighting that ways in which it represents a pattern of value organization different from the three others I have examined—a pattern in which intense value conflict is a routine feature of people’s lives.

The Urapmin understand Christianity to be a religion in which the highest value is achieving salvation when Jesus returns. As they define Christian salvation, it is something that each person must accomplish on his or her own by avoiding sinful thoughts, feelings, and behavior. People spend a good deal of time disciplining themselves in the hope of avoiding sin, and they practice a vigorous ritual life designed to help them in their efforts and to address the effects on individuals of the sins they do commit. Because their Christianity is focused on the individual in this way, I have defined its primary value as individualism, understood as the production of the saved individual.

The production of any kind of individual was not an important goal in traditional, pre-Christian Urapmin life. Rather, at the center of that traditional life was the effort to produce and maintain relationships. This was realized through numerous daily activities and ritual practices in which people worked to bring others into relationships or to keep them involved in relationships into which they had already entered. Central to these practices was the balanced practice of what the Urapmin saw as two parts of the person: the will and the desire to act lawfully. These two parts of the person were sometimes in conflict. The will pushes them toward initiating ever more relationships, often neglecting those that already exist, while people’s lawful tendencies drive them toward meeting the legitimate demands of pre-existing relations by honoring claims for reciprocal treatment and continued involvement. The traditional Urapmin value of relationalism demanded that people balance these two aspects of themselves so as to continue to generate new relations while keeping older ones from falling by the wayside. To succeed in this task was to fully realize relationalism as a value.

The Christian promotion of individualism urges a different understanding of the relationship between people’s willful and lawful tendencies. It defines all willful behavior as sinful because it pushes others to do things that they may not want to do and often thrives on and generates sinful passions, such as anger and covetousness. Rather than court sin through the exercise of will, then, Urapmin Christians aim to meet their lawful obligations of reciprocity within existing relations and to otherwise lead lives that are as “quiet” as possible.

The two values of relationalism and individualism clearly define the ideal relationship between lawfulness and willfulness in distinct, conflicting ways. It is in regards to these divergent definitions of the relationship of lawfulness and willfulness that the Urapmin most routinely experience the conflict between the two values that contend for dominance in their lives. Many attempts to behave appropriately in the pursuit of relationalist goals involve some use of the will, which thus qualifies them as sinful in Christian terms. Conversely, the complete suppression of the will that Christian individualism demands courts failure in relationalist terms. As I have tried to demonstrate in my more thorough account of contemporary Urapmin life, the impossibility of succeeding simultaneously in both of these value systems leads the Urapmin to experience much of their lives as profoundly marked by conflict (Robbins 2004). They make significant efforts to install Christian individualism as their dominant value, thus leaving the claims of relationalism to find expression only at a subordinate level of their ideology. But in reality, relationalism organizes all the key productive and reproductive contexts of their social lives, and thus their efforts to put it in its place have not fully succeeded. At present, at least, they live more fully than any of the people discussed in the other three cases that I have treated, with a full-fledged pluralism that regularly presents them with the need to make what they experience as tragic choices in which their pursuit of one of their two values means a loss in relation to the other.


In this article, I have endeavored to find some inspiration for anthropology in philosophical debates concerning value pluralism and value monism. Having noted that it is the nature of relations between values that is central to these debates, I suggested that anthropologists could explore such relations empirically, and that Dumont’s theoretical framework could be useful in this effort. Turning to Dumont’s important discussion of hierarchical opposition, in his notions of encompassment, ideological levels, and reversals, I found the basis for an understanding of ideologies as of necessity containing both monist and pluralist tendencies. In light of this understanding, one important anthropological task becomes that of exploring diverse societies and the ideologies they harbor to begin to document the various configurations of monism and pluralism that actually exist in the world, and to explore the kinds of value dynamics that these different configurations tend to generate. I followed up this suggestion by looking at four distinct configurations of pluralism and monism drawn from published accounts and from my own work among the Urapmin of PNG. These cases in no way cover all possible configurations, but they do begin to map out a range of variations, from highly monist to profoundly pluralist ones.7

It is my hope that the argument I have just summarized and its illustration through the four cases I have presented might at this point stand on its own. Rather than asserting its wider importance in the remainder of this conclusion, I want to pick out for discussion three issues relating to the contribution of Dumont’s work to this project.

The first point is perhaps the most generally relevant. When philosophers write about values, they tend to discuss the relationships between them without any reference to the nature of social organization. One key contribution of Dumont’s work that I have drawn on but not yet mentioned explicitly is that his notion of levels tends always to move us from a discussion solely about the relative ranks of values within an ideology to one also about the ways values serve to rank ideologically defined social contexts. “Levels” in Dumont’s usage turn out to be not only ideological constructs, but also domains of social action, such as the family, the market, political or religious systems, and so forth. This lends his approach to values a dimensional complexity that more philosophical ones tend to lack, and it is precisely its openness to this complexity that allows the issue of reversals, and of the necessary existence of pluralism even within strongly monist ideologies, to come into view. I would argue that it is his attentiveness to the way value relations find expression in social life itself that has allowed Dumont to make a distinctive contribution to their study. Further anthropological advances are likely to depend on a continued commitment to keeping values and social life in focus together in this way in our future work.

A second point I would like to make has to do with the importance of religion in all of the cases I discussed. Dumont’s argument that religious values traditionally encompassed political ones in India has been profoundly controversial. But for understanding his theoretical program, it is important to recognize that the position religion occupies in his analysis of India is not a contingent one. Dumont tends to define the highest level of value as religious in any given case, for if the highest value gives rise to the whole, and if religion is the name we give to that part of an ideology that deals with the limits of a world and what lies beyond them, then we are bound to define the highest value as religious in nature. I have to admit that in the past I have been a bit allergic to the stipulative nature of this argument—the claim that whatever articulates the whole must be religious by definition. But in working through the cases I have presented here, I find it more plausible than I once did. Certainly in all these cases, religious values can be seen as at least contending for paramountcy, except perhaps in the Avatip case, where they have settled for something like co-paramountcy. And in any case, one wonders if one ever finds a situation approaching a stable monism in which the paramount value does not appear to have what in our usual understandings we tend to see as a religious character. I do not want to go further in exploring this issue here, but I have tried to lay it out with some clarity as one that bears further consideration in the future.

Finally, I want to raise briefly the question of whether Dumont, even as he has given us the tools to explore the monist and pluralist tendencies at work in all societies, himself longed for a stable holism. I believe he is widely seen this way. But as I have discussed at length elsewhere, I do not think this is a fair reading of him (Robbins 1994). As I noted earlier, Dumont at least acknowledges that almost all contemporary societies will contain both holist values (which all societies must necessarily promote at least on some levels of their ideology) and individualist ones (which have by now been carried everywhere by the globalization of modernity). Given this, Dumont recognizes that any hopes one might have for a simple monism are misplaced empirically, if not ruled out, as I have suggested in this article, on theoretical grounds. But Dumont (1994: 16) is concerned that the contemporary age might tend to produce what he calls a certain kind of “obscure blend” of holism and individualism—a particular blend that he has consistently worried can result in the rise of monist forms of totalitarianism (Dumont 1986; see Robbins 1994). Against this possible outcome, Dumont (1994: 16) hopes that in “the long run” the contemporary world will produce ideologies marked by “a clear hierarchy” between individualist and holist values. But as we have seen, hierarchy, as Dumont understands it, is inherently pluralist, even if it can also support monist aspirations. So again, even when Dumont hinted at the kind of future for which he might have hoped, he leaves us with the knowledge that value relations are complex—and he leaves us with the resources to study them as such.


Berlin, Isaiah. 1998. The proper study of mankind: An anthology of essays. Edited by Henry Hardy, Roger Hausheer, and Noel Annan. New York: Farrar, Straus and Giroux.

Chang, Ruth. 2001. “Value pluralism.” In International encyclopedia of the social and behavioral sciences, edited by Neil J. Smelser and Paul B. Baltes, 16139–45. New York: Elsevier.

Dumont, Louis. 1980. Homo hierarchicus: The caste system and its implications. Translated by Mark Sainsbury, Louis Dumont, and Basia Gulati. Chicago: University of Chicago Press.

———. 1986. Essays on individualism: Modern ideology in anthropological perspective. Chicago: University of Chicago Press.

———. 1994. German ideology: From France to Germany and back. Chicago: University of Chicago Press.

Dworkin, Ronald. 2011. Justice for hedgehogs. Cambridge, MA: Harvard University Press.

Fader, Ayala. 2009. Mitzvah girls: Bringing up the next generation of Hasidic Jews in Brooklyn. Princeton, NJ: Princeton University Press.

Galston, William A. 2002. Liberal pluralism: The implications of value pluralism for political theory and practice. Cambridge: Cambridge University Press.

Harrison, Simon. 1985a. “Ritual hierarchy and secular equality in a Sepik River village.” American Ethnologist 12 (3): 413–26.

———. 1985b. “Concepts of the person in Avatip religious thought.” Man (NS) 20 (1): 115–30.

Lassman, Peter. 2004. “Political theory in an age of disenchantment: The problem of value pluralism: Weber, Berlin, Rawls.” Max Weber Studies 4 (2): 253–71.

———. 2011. Pluralism. Cambridge: Polity.

Müller, Jan-Werner. 2012. “Value pluralism in twentieth-century Anglo-American thought.” In Modern pluralism: Anglo-American debates since 1880, edited by Mark Bevir, 81–104. Cambridge: Cambridge University Press.

Plaw, Avery. 2004. “Why monist critiques feed value pluralism: Ronald Dworkin’s critique of Isaiah Berlin.” Social Theory and Practice 30 (1): 105–26.

Robbins, Joel. 1994. “Equality as a value: Ideology in Dumont, Melanesia, and the West.” Social Analysis 36: 21–70.

———. 2004. Becoming sinners: Christianity and moral torment in a Papua New Guinea society. Berkeley: University of California Press.

———. 2007. “Between reproduction and freedom: Morality, value, and radical cultural change.” Ethnos 72 (3): 293–314.

———. 2012. “Cultural values.” In A companion to moral anthropology, edited by Didier Fasin, 117–32. Malden, MA: Wiley-Blackwell.

Robbins, Joel, and Jukka Siikala. 2013. “Hierarchy and hybridity: Toward a Dumontian approach to contemporary cultural change.” Unpublished manuscript, last modified January 13. Microsoft Word file.

Rogers, Douglas. 2009. The old faith and the Russian land: A historical ethnography of ethics in the Urals. Ithaca, NY: Cornell University Press.

Strathern, Marilyn. 1988. The gender of the gift: Problems with women and problems with society in Melanesia. Berkeley: University of California Press.

Tuzin, Donald. 1982. “Ritual violence among the Ilahita Arapesh: The dynamics of moral and religious uncertainty.” In Rituals of manhood: Male initiation in Papua New Guinea, edited by Gilbert H. Herdt, 321–55. Berkeley: University of California Press.

———. 1997. The cassowary’s revenge: The life and death of masculinity in a New Guinea society. Chicago: University of Chicago Press.

Weber, Max. 1946. From Max Weber: Essays in sociology. Translated by Hans H. Gerth and C. Wright Mills. New York: Oxford University Press.


Monisme, pluralisme et la structure des rapports de valeur. : Une contribution dumontienne à l’étude contemporaine de la valeur

Résumé : Si le problème des valeurs n'a pas été un point central des discussions anthropologiques des dernières décennies, il a continué à jouer un rôle important en philosophie au cours de cette même période. Dans ce domaine, un débat essentiel autour des valeurs concerne la question de savoir laquelle des approches moniste ou pluraliste décrit le mieux la façon dont les valeurs se rapportent les unes aux autres dans le monde. Passant en revue la littérature philosophique sur ce sujet, je montre que la préoccupation n'est pas tant celle du nombre de valeurs présentes dans une société donnée, mais la nature des relations entre elles. M'appuyant sur les travaux théoriques de Dumont, je suggère que la recherche ethnographique permet de démontrer que les deux tendances moniste et pluralistes existent dans les rapports de valeur dans toutes les sociétés et que la tâche analytique principale devient donc non pas de déterminer si une société est moniste ou pluraliste, mais plutôt de documenter quels types de configurations de relations, moniste et pluraliste, nous pouvons trouver dans les sociétés réelles. Je présente quatre esquisses ethnographiques de différentes configurations, et démontre l'avantage de ce type d'approche pour contribuer à la fois à la recherche anthropologique et aux débats philosophiques sur la valeur.

Joel ROBBINS is Professor of Anthropology at the University of California, San Diego, and is currently Jane and Aatos Erkko Visiting Professor in Studies on Contemporary Society at the Helsinki Collegium for Advanced Studies. His research focuses primarily on religion, cultural change, ethics, and values. He is the author of Becoming sinners: Christianity and moral torment in a Papua New Guinea society (University of California Press, 2004).

Department of Anthropology – 0532
University of California, San Diego
9500 Gilman Dr.
La Jolla, CA 92093–0532


1. In making this statement, I am referring to the study of “values” in the plural—roughly, those things defined as good within a society or social group. The study of “value” in the singular, a notion often in one way or another tied to Marxist or at least production-oriented points of view, received some attention even during the period during which the topic of values in the plural was almost wholly neglected (see Otto’s introduction to this issue).

2. I have argued elsewhere that the definition of equality only in terms of equality of opportunity is impossible in practice, because equality of opportunity presupposes an earlier equality of outcome (Robbins 1994). But my argument here is meant only to illustrate in formal terms the nature of monist arguments about the ways values fit together in supportive relationships, and the problems of defining equality in terms of opportunity alone are not important for this purpose.

3. It may come as a surprise to some readers that I say that Dumont’s reference to“paramount” values is occasional. But at least in recent rereadings of Homo hierarchicus and Essays on individualism, I have found this to be true. I mention this because the fact that Dumont does not regularly rely on this term is germane to the argument I am making here about the ways monism and pluralism figure in his work. The relevance of this point should become clear below.

4. In making my argument about Dumont’s recognition of value pluralism here, I have focused on his most common definition of a hierarchial opposition as one in which a whole is opposed to one of its parts. There are places in which he at least in passing defines it as also including instances in which a hierarchical relation holds between “two parts with reference to the whole” (Dumont 1986: 279). I think this is an important variant, and one that might at first appear to lend itself to a monist interpretation. This is the case because if X and Y can be ranked by virtue of their differential contribution to the whole, then the contribution to the whole emerges as the single relevant value. But even as this is true of this comparison of two parts, we must note that the very distinction between part and whole on which this comparison depends presupposes that the more canonical form of hierarchical opposition is already in play. And as we have seen, this form of hierarchical opposition is of necessity pluralist.

5. Weber, through his notion of the rationalization of different values (see Robbins 2007), and Dworkin (2011: 119) make similar points about the appeal to people of the goal of accomplishing monism, even as Weber and Dworkin disagree about the possibility of achieving that goal.

6. I have discussed this case and case three in greater detail but to slightly different ends in Robbins (2012).

7. It would be interesting to consider how one might analyze the cases in this issue presented by Gregory, Iteanu, and Willerslev in terms of the different configurations of monism and pluralism they represent.