Authors –Marx, Durkheim, Weber, and so on– represent landmarks which structure our theoretical space and our perception of this space. The difficulty of sociological writing stems from the fact that you have to struggle against the constraints inscribed in the theoretical space at a given moment – and especially, in my case, against the false incompatibilities that they tend to produce (Bourdieu 1990a: 30).
One way of describing Cultural Theory is to say that it is a programme for the extermination of dualisms (Thompson 1996: 49).
My hope in writing this book is to show that social scientists have independently discovered bits and pieces of something that can only be understood when we bring the fragments back together and reconstruct the whole from which they come (Fiske 1991a: vii).

2.2 Although each scholar goes about this in his or her own way (with Bourdieu taking relatively more from the works of Marx, Douglas and Thompson seeking to build more on the ideas of Durkheim and Evans-Pritchard, while Fiske puts somewhat more emphasis on Weber and Piaget), this shared ambition to overcome the ancient dilemmas and divisions of social science is the main source of the many similarities between their theories.

2.3 A first such commonality is the attempt to provide a synthesis of the long-standing opposition between objectivism and subjectivism. Social scientists have often been split between those who have held that social life is governed by the law-like interaction of objective (that is, non-mental) forces, and those who have assumed that social life is instead determined by people’s ideas, values and perceptions (i.e., their subjective states) that defy any attempt at generalisation. In other words, a common assumption has been that if one wanted to generate valid, scientific knowledge about social life, then it would be fruitless to focus on “subjective” factors, such as perceptions, norms, and ideas. Not so, according to our authors:

Of all the oppositions that artificially divide the social sciences, the most fundamental, and most ruinous, is the one set up between subjectivism and objectivism (Bourdieu 1990b: 26).
Our view is that this rigid dichotomy between interpretation of meaning and scientific explanation is unjustified. It is true that human beings create meaning. But it is also true that it is possible to make statements of regularities that help in explaining and even predicting (or retrodicting) the human construction of meaning. Subjectivity need not rule out regularity as long as different sorts of people feel subjective in similar ways regarding similar objects (Thompson, Ellis and Wildavsky 1990: xiii).
[My Structures of Social Life] was also implicitly an argument with the relativistic cultural constructionists… I hoped (futilely, no doubt) to convince them that it was possible for universal evolved mechanisms to generate cultural diversity and uniqueness – and to show why (Fiske 2004a: 19).

2.4 All four theorists attempt to resolve this old conundrum by spelling out ways in which people’s thoughts, assumptions and perceptions are linked with, and are constrained by, ways of structuring human interaction. In the theory of practice (Bourdieu 1977: 72), a particular “social structure” gives rise to a specific “habitus” (i.e., “a system of durable, transposable dispositions”), which serves to fortify that social structure. In cultural theory, five “ways of organising social relations” induce four “cultural biases” (including views of nature, human nature, time, space, risk, blame and justice, besides many other predispositions), which justify and underpin those ways of organising social relations. In relational models theory, four “mods” (fundamental models for social interaction) are complemented by “preos” (consisting of “paradigms, parameters, precepts, prescriptions, propositions, and proscriptions” – Fiske 2004a: 4). In insisting that the fundamental categories of human thought are shaped by –and, in their turn, shape– different social structures, all the authors follow the main thesis of Durkheim’s The Elementary Forms of Religious Life (1912/1985).

2.5 Another social science-dualism that these authors strive to overcome concerns the question whether people’s behaviour is determined by collective forces or by personal decisions: is society the outcome, or the source, of individual predispositions and actions? Again, our four authors argue that this is a conceptual non-starter. In their view, any social theory worth its salt needs to make room for both social determination and individual initiative, and needs to spell out their possible mutual relationships. Essentially, the authors make two claims. First, they assert that social structures never fully determine people’s choices, but merely constrain them. They compare social structures to the rules of a game, which can be bent, disputed and followed in different ways, and which leave plenty of room for alternative game plans. In addition, they argue that although people’s social relations provide them with their basic categories of thought (such as views of time, space, human nature, causation), individuals remain capable of critically reflecting upon, and comparing, their social conditions and the empirical validity and practical value of their own assumptions – and thus of bringing about social change. Therefore, as one cultural theorist puts it:

It is individuals as social creatures that, not only being moulded by but actively moulding their social context –shaping the maze as well as running it– that are the focus of cultural theory (Wildavsky 1987: 7).

2.6 Compare this to:

Habitus is not the fate that some people read into it. Being the product of history, it is an open system of dispositions that is constantly subjected to experiences, and therefore constantly affected by them in a way that either reinforces or modifies its structures. It is durable but not eternal! (Bourdieu and Wacquant 1992: 133).

2.7 And to:

The relational models do not fully specify behaviour in any interaction or situation, but they comprise a set of rules that strongly constrain the possibilities and that organize responses to violations of the rules (Fiske 1991a: 21).
Relational models theory argues that [..] people are thinking about the structures and processes of interaction. People are intensely concerned and deeply knowledgeable about relationships in their own right [..]. We care about and cognize equality, balance, hierarchy, shared communal identity, and so on (Fiske and Haslam 2005: 282).

2.8 The next red herring across which Bourdieu, Douglas and Thompson, and Fiske collectively leap concerns the debate whether either “values” or “self-interests” are the prime movers of social life.

2.9 They have all pointed out that these concept are often left undefined in social and political science, merely to be used in a tautological manner (Bourdieu and Wacquant 1992: 116-17; Douglas 1987: 24-25; Schwarz and Thompson 1990: 49; Fiske 1991a: 384-7). That is to say, by not defining and delimiting such notions as norms, values and interests, it becomes possible to always claim that whatever actions people have engaged in was the result of their adherence to some norm, value or perceived self-interest (cf., Sen 1977; Hirschman 1992). Rather than confining oneself to this distressingly small circular loop, the authors point out that it makes more sense to conceptualise the origins, and content, of people’s norms and perceived interests. In their view, as I already mentioned, social structures determine to a significant degree people’s fundamental categories of thought, such as their notions of time, space, human nature, justice, causation, honor, etc. Taken together, these predispositions shape what people perceive to be the overriding goals (or interests) that they should be striving for, the legitimate and illegitimate strategies for doing so (i.e., the norms and values that they should keep in mind), and what constitutes and confers status and power in particular situations. Thus, in their view, people’s values and norms are just different elements of their habitus (to use Bourdieu’s concept), cultural bias or cosmology (to invoke the terms of cultural theory), or interaction models (in Fiske’s phrase).

2.10 Yet another dichotomy that is conspicuously absent in the theory of practice, cultural theory and relational models theory is the one between modernity and pre-modernity. Mary Douglas’s classic precursor to cultural theory, Purity and Danger (1966), is explicitly written against the old division of the world in parts that are primitive, pre-modern, magical, superstitious, religious, irrational and pre-scientific, and those that are advanced, modern, secular, rational and scientific. (She blames an earlier social science-classic, The Golden Bough: A Study in Magic and Religion by the anthropologist Sir James George Frazer (1922/1995), for popularising this distinction within academia.) Indeed, it is possible to characterise the entire research of Douglas as an effort to show that the actions, emotions and perceptions of human beings have always been shaped by social structures and cultural biases – no matter where and when their lives unfolded. Thompson followed up on her lead, showing –in articles such as “Fission and Fusion in Nuclear Society’ (1980); “Among the Energy Tribes” (1984); and “How to Save the Himalayas When You Cannot Find out What’s Wrong with Them” (1985) – that the policy – making and scientific elites in the fields of energy provision and development assistance can be as (if not more) blinkered as anyone, including the Nepali villagers they are supposed to advise. Very much the same applies to the frameworks of Bourdieu and Fiske. In their view, too, all people are rational, creative and reflective individuals that are constrained, motivated and directed by their social relations:

Social agents, in archaic societies as well as in ours, are not automata regulated like clocks, in accordance with laws which they do not understand. In the most complex games, matrimonial exchange for instance, or ritual practices, they put into action the incorporated principles of a generative habitus (Bourdieu 1990a: 9; emphasis added).
The broadly consistent pattern of results across American, Bengali, Chinese, Korean, and Vai subjects suggests that the four relational models are not etnosociological folk concepts derived from any particular culture. They are implicit schemata people use in diverse and widely dispersed cultures – perhaps all cultures. Different cultures use the same models in different domains and implement them according to different parameters, but the underlying models are the same (Fiske 1993: 491).

2.11 It is no coincidence that the first, rudimentary insights of the three theoretical frameworks were formulated during anthropological fieldwork in various parts of Africa (Bourdieu 1958; Douglas 1963; Fiske 1985), and where then later transposed, and generalised, to (dare I say?) “more industrialised” societies.

2.12 Another fork in the road of the social sciences that the four scholars refuse to take consists of the usual split between those who assume that functional forces are at work within societies and those who assert that societies are conflict-ridden. The theory of practice, cultural theory and relational models theory all have a functionalist streak, in that these frameworks assume that social structures tend to produce the systems of collective thought that justify these very social structures (of course, with all the provisos for individual creativity, reflection and critique that I mentioned above). Yet these theories also view social life as characterised by an unceasing struggle between different groups of actors, each of which is trying to impose its preferred way of organising (and, therefore, of interpreting) society onto everyone else. Again, the language is different, but the message is quite similar:

Each state of the social world is thus no more than a temporary equilibrium, a moment in the dynamics through which the adjustment between distributions and incorporated or institutionalized classifications is constantly broken or restored. The struggle which is the very principle of the distributions is inextricably a struggle to appropriate rare goods and a struggle to impose the legitimate way of perceiving the power relations manifested by the distributions (Bourdieu 1990b: 41).
Cultural transmission is absolutely not a game of pass-the-parcel. Culture is a lively and responsive thing, more like a cat fight than a single solitary cat, a coherent flurry of contradictory bundles that are continually being negotiated and renegotiated in order to sustain and justify the preferred patterns of social relations that go along with them. And if, as we argue, social, political, and economic life is about the clash of these contradictory preferred patterns of relations –preferred ways of life– then there will be more than one cultural bias in any one society. The continuity of the whole (like the cat-fight) is sustained by the perpetual divergence of the parts. A consensus around customs does indeed exist –the Brits eat roast beef and the French eat frogs’s legs– but it is neither very interesting nor very instructive until and unless we connect these customs to continuity or change in ways of life. Cultural dissensus –the clash of cultural biases– is what we should be looking at and analyzing policies in terms of (Thompson and Wildavsky 1986: 170-71).
People are disposed to get angry and punish those who violate the models that they themselves are using, but the targets of such sanctions often do not acknowledge that that particular model applies, or that their acts were transgressions, so they perceive the intended sanctions as illegitimate aggression. This may launch rounds of mutual recrimination or violence (Fiske 2004a: 21).

2.13 And this immediately reveals the way in which these authors are able to reconcile a functionalist account with the assumption that social life is essentially antagonistic. Rather than assuming that the whole of society is dominated by a single way of organising and interpreting (as was often the case in previous social theories – see Thompson, Ellis and Wildavsky, part II), the authors assume that social situations typically comprise several antagonistic groups of actors adhering to rivalling ways of organising and perceiving.

2.14 Besides their shared ambition to not get impaled on the horns of a variety of long-standing social science-dilemmas, Bourdieu, Douglas and Thompson, and Fiske have developed other ideas that are strikingly similar as well. They all stress what Bourdieu (1977: 109-114) has labelled the “economy of logic”, i.e., people’s tendency to use a small number of cognitive schemes and templates to make sense of different parts of their lives. As cultural theorist Aaron Wildavsky (1987: 8) put it:

[Social filters] enable people who possess only inches of facts to generate miles of preferences.

2.15 This is not far removed from Bourdieu’s (1977: 110) assertion that:

Practical logic –practical in both senses of the word– is able to organize the totality of an agent’s thoughts, perceptions, and actions by means of a few generative principles…

2.16 Or from the following observation by Alan Fiske:

[A] few universal models can generate the great cultural diversity that we see around the world and through history (Fiske 1991a: 139).

2.17 In addition, all these authors dwell upon the symbolic roles that the human body plays in shaping society (Bourdieu 1977: 114-24; Douglas 1970; Fiske 1991a: 391-400), as well as chide neo-classical economics for acting as a scientific front for a particular ideology by presenting a socially constructed reality (namely, that of competitive markets and people) as a “natural”, inevitable and all-pervasive state of affairs (Bourdieu 1977: 177; Douglas and Isherwood 1979; Fiske 1991b). Last, all these authors assert that the proper unit of analysis for social science is not made up of individual persons or organisations, but rather of systems of social relations. On this, once again, they seem to almost echo each other:

To speak like Cassirer (1923) in Substance and Function: one must think relationally. Now, it is easier to think in terms of realities that can be “touched with the fingers,” in a sense, such as groups or individuals, than in terms of relations…. It is for the purpose of breaking with this substantialist mode of thinking, and not for the thrill of sticking a new label on old theoretical wineskins, that I speak of the “field of power” rather than of the dominant class, the latter being a realist concept designating an actual population of holders of this tangible reality that we call power. By field of power, I mean the relations of force that obtain between the social positions which guarantee their occupants a quantum of social force (Bourdieu in Bourdieu and Wacquant 1992: 228-30; emphasis added).
A final distinctive feature of our relational models theory approach to interpersonal cognition and behaviour is that its level of analysis and emphasis is relational rather than individual. (Fiske and Haslam 2005: 282; emphasis added).
We have now come to the most difficult, and most contentious, part of cultural theory: the counter-intuitive turning-away from the individual as the fundamental unit of attention.… The trick is to see the various forms of solidarity as the ‘prime movers’, each all the time trying to strengthen itself by encouraging more people to place more of their transactions, more of the time, on that pattern of social relationships and its sustaining cultural bias, while doing everything it can to put spokes in the wheels of the others. Preferences, in other words, are formed between people rather than within each of us, and values are stabilised not in the hearts and minds of individuals but by the kind of solidarity that they are developing at any one time. Individuals, however, do not disappear from the picture, nor do we have to see them as preprogrammed automata unquestioningly carrying out the actions that are specified by the solidarities of which they are the zombie-like components. Cultural theory does not deny individuality; all it does is systematically open up the implications of the unremarkable observation that individuality (as Elster nicely puts it) is, to a considerable extent, ‘inherently relational” (Thompson, Grendstad and Selle 1999b: 12; emphasis added).

2.18 In sum, the theory of practice, cultural theory and relational models theory should be counted among the most exciting theoretical enterprises of the last three decades as they represent well-developed syntheses of many previously unsettled debates within the social sciences. These three syntheses are intriguingly similar in a variety of ways. This is not to argue, of course, that no differences exist between the frameworks. Their differences will come out in a discussion of their (relative) weaknesses and strengths.


3.1 In particular in the light of cultural theory and relational models theory, the main weakness of Bourdieu’s theory of practice seems to me to be clear: it lacks an explicit typology of basic, alternative forms of organising, justifying and perceiving social relations. Bourdieu’s rejection of the need for, and possibility of, a classification of elementary social forms (e.g. Bourdieu 1984: 470) constitutes the main difference between his work and that of Douglas and Thompson, as well as that of Fiske. In cultural theory, five basic ways of organising social relations –each with distinctive perceptions of nature, human nature, risk, time, space, causality, leadership, etc.– are identified (Douglas 1978): individualism, egalitarianism, hierarchy, fatalism and the hermit. In relational models theory, four elementary structures for coordinating social relations are distinguished (Fiske 1991a): communal sharing, authority ranking, equality matching, and market pricing. In both these theories, the infinite social and cultural variety of the past and present is explained as the endless reinvention, and recombination, of a small set of fundamental forms of human interaction. In proposing these (quite similar) typologies, Douglas and Thompson, and Fiske, are again inspired by the canonical works of social and political science. Marx (1845/1972), Maine (1861/1986), Tönnies (1887/1991), Durkheim (1912/1985), Weber (1922/1980), Piaget (1932/1997), Benedict (1934/1989), among many others, were all convinced of the view that explanation and theory-building require typologies. Bourdieu (1984: 479), of course, acknowledges that:
What is at stake in the struggles [of the social world] is power over the classificatory schemes and systems which are the basis of the representations of the groups and therefore of their mobilization and demobilization.

3.2 However, he refuses to set out a clear classification of such groups – and their competing classificatory schemes. Thus, Bourdieu gets himself into various conceptual troubles.

3.3 The first of these troubles is that without a typology of alternative ways of organising human relations it becomes very difficult to explain social change. A classification enables one to conceive of social transformation as the waxing and waning, as well as splitting and merging, of alternative organisational and cognitive structures. (This would also open the possibility to combine social theory with complexity theory; for cultural theory, this is currently being done by Paul Tayler and Michael Thompson forthcoming). Without a taxonomy, it is hard to understand how social dynamics could be conceived of. As Bourdieu stresses the possibility for critical thinking, resistance and self-reflexivity, he leaves the door open for social change, but without any typology of ways of organising it remains unclear in which direction any change might then be headed.

3.4 In the absence of a clear typology of organisational and cognitive structures, it also becomes much more difficult to achieve one of Bourdieu’s main theoretical aims, namely to facilitate self-reflexivity among social scientists. It is always difficult to recognise –let alone remove– the blinkers that impair one’s vision (if that weren’t the case, then these wouldn’t be blinkers). But if one is armed with a typology of the ways in which people tend to think and act, then it becomes easier to realise from which perspective one is arguing, and on which fundamental assumptions one’s perspective rests – and thus to start taking other viewpoints and concerns into account. For instance, if one applied the cultural theory to which political scientist Aaron Wildavsky has made significant contributions (Douglas and Wildavsky 1982; Thompson, Ellis and Wildavsky 1990) to his own efforts to influence specific public policies (e.g., Wildavsky 1988; Singer and Wildavsky 1993; Wildavsky 1995), then it becomes clear that he was almost invariably advocating individualistic policies. (In terms of relational models theory, Wildavsky was forever favouring more market pricing over other models of human interaction). However, in the theory of practice, each social field is a battleground for the struggle “to determine the conditions and criteria of legitimate membership and legitimate hierarchy” (Bourdieu 1988: 11). In other words, although Bourdieu portrays social life as a battle between groups that are adhering to alternative principles of legitimacy and classificatory schemes, he also depicts each of these alternatives as just another variant of an exploitative, exclusionary and hierarchical way of organising, perceiving and justifying social relations. In his view, each of the competing social groups or classes attempts to impose a particular moral and classificatory system that would ensure its own ascendancy in terms of status, material benefits and influence – at the expense of all others. As he puts it in Outline of a Theory of Practice (Bourdieu 1977: 164-65):

Every established order tends to produce (to very different degrees and with different means) the naturalization of its own arbitrariness. Of all the mechanisms tending to produce this effect, the most important and the best concealed is undoubtedly … the correspondence between the objective classes and the internalized classes, social structures and mental structures, which is the basis of the most ineradicable adherence to the established order. Systems of classification which reproduce, in their own specific logic, the objective classes, i.e. the divisions by sex, age, or position in the relations of production, make their specific contribution to the reproduction of the power relations of which they are the product.
The taxonomies of the mythico-ritual system at once divide and unify, legitimating unity in division, that is to say, hierarchy.
Social representations of the different ages of life, and the properties attached by definition to them, express, in their own logic, the power relations between the age-classes, helping to reproduce at once the union and the division of those classes by means of temporal divisions tending to produce both continuity and rupture. They thereby rank among the institutionalized instruments of the symbolic order, and hence among the mechanisms of the reproduction of the social order whose very functioning serves the interests of those occupying a dominant position in the social structure.

3.5 And, therefore, in Homo Academicus (Bourdieu 1984: 11; emphasis added):

The university field is, like any other field, the locus of a struggle to determine the conditions and the criteria of legitimate membership and legitimate hierarchy, that is, to determine which properties are pertinent, effective and liable to function as capital so as to generate the specific profits guaranteed by the field. … Thus there exists quite objectively a plurality of rival principles of hierarchisation, and the values, which they determine are incommensurate, or even incompatible, because they are associated with mutually conflicting interests.

3.6 From this perspective, seemingly benevolent, generous acts are not to be trusted, as these only serve to shore up the particular form of hierarchical oppression (i.e., the prevailing distribution of economic, symbolic, religious and cultural capital) that happens to dominate in a time and place. Even donations to hospitals and support for welfare policies are not to be taken at face value (Bourdieu 1990b: 133):

It is equally clear why the development of the capacity for subversion and critique that the most brutal forms of ‘economic’ exploitation have aroused, and the uncovering of the ideological and practical effects of the mechanisms ensuring the reproduction of the relations of domination, should bring about a return to modes of accumulation based on the conversion of economic capital into symbolic capital, with all the forms of legitimizing redistribution, public (‘social’ policies) and private (financing of ‘disinterested’ foundations, donations to hospitals, academic and cultural institutions, etc.), through which the dominant groups secure a capital of ‘credit’, which seems to owe nothing to the logic of exploitation.

3.7 In terms of cultural theory, Bourdieu perceives social life mainly as an eternal struggle between different versions of hierarchy – and only does so from the critical vantage point of egalitarianism. In the words of relational models theory, Bourdieu depicts the social world as a series of battles between groups favouring different forms of authority ranking. For instance, when Bourdieu writes of capitalism, education or the arts (Bourdieu et al. 1999; Bourdieu 1999; Bourdieu and Passeron 1977; Bourdieu 1984), then he merely attempts to uncover the systems of domination and oppression (the self-perpetuating distributions of economic, symbolic, cultural and religious capital) that underpin these forms of social life – and he remains silent on their many other conditions and characteristics (such as individual creativity, entrepreneurship, vibrancy, stewardship, or concern for others). At one point in his work, Bourdieu (1984: 470) appears to come close to recognising alternative forms of organising, perceiving and justifying social relations – only to argue again that these could not be anything else but different variations on domination and hierarchy:

And the recurrence of the triadic structure studied by Georges Dumézil, which Georges Duby shows in the case of feudal society to be rooted in the social structures it legitimates, may well be, like the invariant oppositions in which the relationship of domination is expressed, simply a necessary outcome of the intersection of the two principles of division which are at work in class-divided societies – the division between the dominant and the dominated and the division between the different fractions competing for dominance in the name of different principles.

3.8 In thus truncating social life to the struggle between alternative forms of repressive hierarchy, Bourdieu is unable both to recognise the possible useful functions that hierarchical ways of organising may have to fulfil in a peaceful, democratic and plural society (cf. Douglas 2005) and to explore the disadvantages and weaknesses of non-hierarchical forms of organising. Ironically, although Bourdieu sees his call for more self-reflexivity as his most important contribution to the social sciences (Bourdieu 1990b: 15; Bourdieu and Wacquant 1992), his own analysis does not allow for much self-reflexivity. This second line of criticism underscores my earlier point as well: as Bourdieu conceptualises social life as a never-ending struggle between different versions of repressive hierarchy, it is hard to see how fundamental social change could come about or could be explained.

3.9 The absence of an explicit typology of ways of organising human relations in the work of Bourdieu causes two additional problems. This lack of concreteness makes it quite difficult to falsify any part of the theory of practice – and this sits uneasily with Bourdieu’s repeated calls for an objective, scientific analysis of social life. For example, how could one go about falsifying the claim that –in general– classificatory systems matter to the distribution of material goods and influence in society? Furthermore, the lack of concreteness renders the theory less useful as a basis for understanding how pressing social problems can be practically resolved. If seemingly genuine acts of solidarity (such as private gifts to hospitals or social welfare-programs) are to be understood merely as attempts to placate the dominated classes, then how to give concrete policy advice concerning public health, social security, environmental protection, and other pressing problems? How, to be very concrete, to judge for example the current efforts of Bill Gates –the wealthiest person on earth– to eradicate a variety of terrible diseases in the poorest parts of the world by founding a charity dedicated to these causes, and then donating 750 million U.S. dollars to it?

3.10 In my view, the Achilles heel of Alan Fiske’s relational models theory concerns his typology of elementary forms of organising social relations.[5] I am not convinced that it meets the standard demands for a useful classification, namely that the categories of the classification be jointly exhaustive as well as mutually exclusive. In particular, his fundamental modes of human interaction (communal sharing, authority ranking, equality matching, and market pricing) appear to be somewhat overlapping. Communal sharing is “characterized by the fact that people attend to group membership and have a sense of common identity, while the individuality of separate persons is not marked” (Fiske 1991a: 13), while people who engage in authority ranking “construe each other as differing in social importance or status” (Fiske 1991a: 14). So far, so good – these two “mods” appear clearly different. I believe that the overlap begins with the other two mods. Equality matching is “an egalitarian relationship among peers who are distinct but coequal individuals” (Fiske 1991a: 14-15), whereas market pricing denotes “a relationship mediated by values determined by a market system” (Fiske 1991a: 15). To me at least, it appears difficult to always unequivocally separate equality matching and authority ranking on the one hand, from communal sharing and market pricing on the other. For instance, in competitive markets (i.e., transactions that resemble the “ideal markets” of neo-classical economics), actors are certainly “equal, but distinct peers”: none of them has more market power than any of the others, and none of them can hope to earn a higher wage, or make a larger profit, than the others. Yet, these markets not only fulfil the conditions of equality matching, but also of market pricing. Similarly, in many Japanese markets (Dore 1987), each major company is closely related –in an almost paternalistic way– to a set of smaller competitors, suppliers and retail firms. When profits and prices fall or cheaper alternatives become available, the large company at the centre of the network will not abandon the peripheral enterprises, but instead offer them advice as to how to reorganise their companies and even forms of temporary financial support. These practices constitute authority ranking and market pricing at the same time. In addition, whenever communal sharing results in a more or less equal allocation of goods, then it will be hard to differentiate this mod from that of equality matching. Last, as Edward Banfield (1958) has beautifully illustrated (cf., Putnam 1993), a particular form of equality matching is also prevalent when social relations are characterised by extreme levels of inter-personal distrust and jealousy. Under those circumstances –labelled “amoral familism” by Banfield– no cooperation will take place if anyone benefits more from the collective effort than the others. Fiske calls such relations “asocial relationships”, and does not include them among his four models of human interaction (as they are more a model of obstructing human interaction). Still, a perverse form of equality matching is characteristic of such null relationships.

3.11 Fiske does not deny that his categories are partly overlapping. And he has proposed an elegant way out of this predicament by arguing that his four relational models should not be seen as mutually exclusive types, but rather as categories that are nested inside each other: each category is supposed to represent an increasingly complex form of organising human relations. The simplest way of interacting is communal sharing. A bit harder to do is authority ranking, while equality matching is even more difficult. And the most complicated form of organising social relations is represented by market pricing. This increasing complexity is reflected in the different types of measurement scales that are supposed to guide the various relational models (Fiske 1991a: 210-230). In communal sharing, the appropriate measurement scale is categorical/nominal. In authority ranking, decisions are made with the help of an ordinal scale, and in equality matching an interval scale is used. In market pricing, ratios guide decisions. These measurement scales are of course of increasing complexity – with the more complex scale incorporating all of the simpler ones. The idea that the mods are of increasing complexity also reappears in other parts of Fiske’s theory. He argues (Fiske 1991a: 402-4) that children learn each of these forms of interacting consecutively, starting early on with communal sharing and ending relatively late in their childhood with market pricing. Moreover, he asserts (Fiske 2000) that the capacity to engage in these ways of coordinating has become ingrained in different stages of the evolution of human life. However interesting all this is, it also reinforces the criticism that Fiske’s four relational models are not mutually exclusive. In particular in light of the fact that all four mods are supposed to be present within any social situation, this appears to reduce the applicability and clarity of relational models theory.

3.12 Cultural theory has its own drawbacks. One of these also pertains to its classification of fundamental ways of organising, justifying and perceiving social relations. This classification is clearly mutually exclusive and jointly exhaustive, as it is derived (Douglas 1978) from two independent dimensions of social interaction, usually called “grid” (the degree to which social relations are ranked and stratified) and “group” (the extent to which people feel bound to a larger unit than the individual). Assigning two values (high and low) to these two dimensions results in four distinct ways of organising (and justifying and perceiving) human relations: hierarchy (high grid, high group), individualism (low grid, low group), egalitarianism (low grid, high group) and fatalism (high grid, low group). A fifth way of life, that of the hermit, is then added by noting that while the other ways of life are all means of engaging in social interaction, people can also decide to withdraw from social life (Thompson 1982). Each social field is then supposed to consist of a never-ending waxing and waning, and merging and splitting, of these five “social solidarities” (to invoke Durkheim’s old term). Although this typology has been remarkably successful –in particular in setting out the disagreements among people regarding a wide variety of public issues, from safety seat belts-legislation (Adams 1995) to the governance of the Internet (Tranvik and Thompson 2005) – it remains problematic that two continuous dimensions (grid and group) are used to produce four distinct ways of organising (and one way of avoiding) social relations. It would be more elegant for these two dimensions to be dichotomous. Hence, the search is still on for a proper definition of cultural theory’s two central terms.[6]

3.13 In addition, it is still not clear why, as cultural theory states and as has been confirmed by quite a bit of empirical research (e.g. Coyle and Ellis 1994), public debates are typically polarised between groups of people adhering to quite distinct and pure versions of the ways of organising and perceiving (whereas the social domains in which these people have to operate typically consist of ever-changing mixes of these ways of organising). How does this process of polarisation of opinion along the four lines set out by cultural theory take place? Why it is so easy to recognise individualistic, egalitarian, hierarchical and fatalistic voices in public debates – in particular when it seems clear that people’s lives take place in many social domains that are typically mixes of the various ways of organising and thinking?

3.14 Last, I believe that cultural theory’s initial “promise” –the very reason for which Douglas (1970; 1975; 1978; 1982) set out to develop the approach– has not yet been fully delivered upon. In developing cultural theory, Douglas (1978: 15) explicitly aimed to construct a method that would enable comparisons across time and space, which would then facilitate the formulation and testing of novel explanations of such important social phenomena as nationalism, war, religion, poverty, genocide, etc. In my view, this challenge has only been partially taken up thus far. One interesting application of cultural theory has been to the field of leadership (Wildavsky 1984; Ellis and Wildavsky 1989), where it has deepened our understanding of which types of political and religious leadership are acceptable, and therefore successful, in different social settings. But perhaps the most impressive applications have been to the field of governance (Schwarz and Thompson 1991), where it has been established that public policies and business plans tend to fail if they are not based on a full consideration of all ways of organising and perceiving, as well as to the analysis of risk (Douglas and Wildavsky 1982; Thompson and Rayner 1998), where it has been shown that people’s perceptions of risk are best predicted by their preferred ways of organising. Yet, the bulk of applications of cultural theory have remained confined to showing that it is possible to distinguish between egalitarian, hierarchical, individualistic and (at times) fatalistic voices in discourses on social issues. Douglas’s clarion call –to construct, on the basis of the theory, alternative explanations of such vital social phenomena as nationalism, religion, revolution, famines, terrorism, etc.– has thus far not been sufficiently heard, or acted upon. Much remains to be done in this regard (Coyle 1994).


4.1 Each of these three theories, therefore, has its own, specific weaknesses. It is equally important to note that they all have their comparative strengths as well. In my view, Bourdieu’s work distinguishes itself from the other two approaches in at least two ways. First, it sets itself apart by its insistence on the need for social science to be studied in a self-reflexive manner (i.e., to reflect critically and rigorously on both the social conditions for, and the social consequences of, social theorising itself). This, to me, remains a vital insight and practice that the social sciences should always incorporate. Second, Bourdieu’s research stands out for having been applied to elucidate whole social fields, such as the French university system or the French cultural field. Such an ambitious endeavour has not been undertaken on the basis of cultural theory or relational models theory, and Bourdieu’s work in this regard should function as an inspiration to do so.

4.2 Alan Fiske’s relational models theory has been ground-breaking in forging links with a variety of sciences that are not usually considered when building social theory. Together with some of his UCLA colleagues (Iacoboni et. al. 2004; Fiske 2005), he has been at the forefront of the efforts to bridge the gap between the social sciences and brain research. He has attempted to deepen our understanding of the human brain with the help of relational models theory, and, vice versa, to fortify his social theory with novel insights from neuroscience. Among other things, Fiske and his colleagues have suggested –on the basis of solid neuroscientific research– that being aware of, and cognising, social relations may well be the “default state” of the human brain, thus providing further ammunition to the social scientists who have sided with Aristotle in arguing that humans are first and foremost social beings. Closely related to these efforts, Fiske and his collaborators have begun to search for fruitful ways of linking social theory with evolutionary theory and studies of animal social complexity (Haslam 1997; Fiske and Fiske 2005). They have suggested that the human ability to engage in communal sharing may have been an extension of an early evolutionary trait, namely the bond between parents and offspring, whereas the human ability to engage in authority ranking (and other relational models) may have emerged at a later, more complex evolutionary stage. With the help of evolutionary theory, Fiske (2005) has begun to build the case for why people may be able to co-ordinate their mutual relations in not more than a handful of basic ways. Fiske has also pushed the envelope in having been the first one to link his social theory to childhood studies (arguing that children learn to interact according to different models at consecutive phases) and to the study of emotions (Fiske 2002). In addition to all this, Fiske has been one of the very few social scientists to confront one of the most basic, and perhaps most difficult, questions in the social science (Fiske 2004b): what is a “social relationship”? How do social relationships get constituted?

4.3 Specific praise can also be heaped upon the cultural theory developed by Mary Douglas and Michael Thompson. Apart from its more useful typology, it appears the one theory that has the clearest and most concrete implications for collective efforts to alleviate human suffering. As it argues and shows in detail (Schwarz and Thompson 1991; Hood 1998; Thompson, Grendstand and Selle 1999a) why policies that are not based on a full consideration of all the points of view that are advocated by the people and organisations involved tend to be self-defeating and to increase human misery, it offers yet another form of support for establishing human rights, democracy and deliberative decision-making. In addition, in particular Michael Thompson’s efforts have linked cultural theory in elegant ways to the study of human complex systems (Tayler and Thompson 1986; Tayler and Thompson forthcoming), ecology (Thompson 1996) and technology studies (Tranvik, Selle and Thompson 2000). Thus, he has been able to sketch the contours of a theory in which social, ecological and technological forces constantly interact in unpredictable, though not random, ways (Thompson 2002).

Summing Up

5.1 I believe that by combining these three frameworks, we may be able to arrive at a social theory that is cumulative, parsimonious, relevant to collective efforts to alleviate pressing social problems, in support of human rights, and linked in interesting ways to relatively novel fields in the natural sciences (including social neuroscience, the study of animal social complexity, ecology and non-linear mathematics). This may seem a very distant goal, but I think that the analyses of Pierre Bourdieu, Mary Douglas and Michael Thompson, and Alan Fiske have already brought us a considerable distance. In showing that there are solutions to many of the long-standing disputes in the social sciences, they have helped us overcome a number of the obstacles that have tripped up previous attempts at building social and political theory. Moreover, by picking and mixing the most appealing features of these three approaches, we can get ourselves even further on our way. I will conclude by sketching my favourite conceptual brew, which I propose to call “a theory of constrained relativism”.

5.2 Besides all the elements that are shared by the theory of practice, cultural theory and relational models theory, I would advocate selecting the five-fold typology that has been developed by Mary Douglas and Michael Thompson – with the proviso that the underlying dimensions would have to be further clarified. Without a typology there can be no theory. Although this statement appears a commonplace, banality even, among natural scientists, and was also unquestioned by the “grand” social scientists of yesteryear, in my experience it often comes as a surprise to today’s sociologists, anthropologists, economists and political scientists. Yet, consider the following description of theory by Milton Friedman (1953: 53). In his view, theory shows:

that there is a way of looking at or interpreting or organizing the evidence that will reveal superficially disconnected and diverse phenomena to be manifestations of a more fundamental and simple structure.