The sociological theory of modernization propounds a dichotomy between tradition and modernity that has diminished the relevance and explanatory power of tradition and devalued it as a key concept in the discipline. This article seeks to revitalize the concept of tradition by defining it as “an assigned temporal meaning”, i.e., a socio-cultural arena in which temporal identity is constructed by various social groups. Further, it will argue that the conceptual ambiguity of tradition results from scholars’ tendency to define tradition as an ontological reality instead of an assigned meaning.

Sociology has ontologized tradition via two incompatible approaches. The first, which we will call the “anti-modern approach”, saw tradition as the polar opposite of modernity. Detraditionalization – the demise of tradition and the crumbling of traditional society – was seen, for better or for worse, as a necessary corollary, if not the main goal, of the advance of modernization. The second, which we will call the “totality approach”, saw tradition as the be-all and end-all of social and individual life: a central source of knowledge, institutions, and values. The extreme version of this approach practically equated tradition with concepts such as, “culture” or even “society”.

The lack of dialogue between these two schools diminished the explanatory power of the concept of tradition, which was dismissed as a pre-modern residue or turned into the essence of society itself. It is no wonder, then, that since the “reflexive turn” of the 1980s (Turner Reference Turner1991), scholars have called tradition “a misnomer, sending all the wrong signals” (Scott Reference Scott1998, p. 331) and have discarded it.

Sociology’s polar approaches to tradition as either a useless relic or the essence of society stand in sharp contrast to the concept’s pervasiveness in modern life. Tradition plays a significant role in the institutionalization and routinization of cultural practices, the production of cultural and social capital (the rise and fall of fashions, genres, etc.), the consolidation of collective memory, and the identity construction, in the broad sense of the term, of social groups, including nations (Mosse Reference Mosse1975; Herzfeld Reference Herzfeld1982; Wilson Reference Wilson1976; Hobsbawm and Ranger Reference Hobsbawm and Ranger1983; Ozouf Reference Ozouf and Sheridan1988; Nora Reference Nora and Goldhammer1997; Smith Reference Smith2000, pp. 53-60), empires (Cannadine Reference Cannadine2001), ethnoses (Handler Reference Handler1988, pp. 52-80; Povinelli Reference Povinelli1999), classes (Hobsbawm and Ranger Reference Hobsbawm and Ranger1983, pp. 283-291), cities (Burke Reference Burke1987, pp. 244-250), regions (Linnekin Reference Linnekin1983; Becker Reference Becker1998), universities (Shils Reference Shils1981, pp. 179-185; Jacobs Reference Jacobs2007), factories and commercial organizations (Feldman Reference Feldman2006), families (Pleck Reference Pleck2000), and, to some extent, even religions (Wertheimer Reference Wertheimer1992; Graham Reference Graham1993; Devji Reference Devji2007; Nyiri Reference Nyiri and Nyiri1995, p. 20; Sagi Reference Sagi and Stein2008). Modern societies react to the concept of tradition with “sensations; desires; professional, personal and national optimism; and anxieties” (Povinelli Reference Povinelli1999, p. 22).

Each of these social bodies is impacted by tradition in four central aspects: the hermeneutical aspect, or interpretation of the world; the normative aspect; legitimizing authority; and identity construction (Thompson, in Heelas, Lash and Morris Reference Heelas, Lash and Morris1996, pp. 91-93). Classic sociology, on the other hand, saw tradition as either inferior to modernity or identical with society. Thus the need to invigorate the concept of tradition cannot be overestimated.

A relevant theory of tradition can be built from the perceptive insight of Richard Handler and Jocelyn Linnekin, folklorists who see tradition “not [as] an objective property of phenomena but [as] an assigned meaning” (Handler and Linnekin Reference Handler and Linnekin1984, p. 286). This means that tradition, first and foremost, is a symbolic activity that attributes traditional qualities to certain sectors of life that are understood as binding together different times (Smith Reference Smith2007). I propose the following definition: Tradition is a socio-cultural practice that assigns temporal meaning.

Tradition is a cultural practice referred to as “traditional”. Although this may sound tautological, it may not be so self-evident in sociology. It discursively binds a social entity – large or small collectives or even individuals (Hall Reference Hall and Rutherford1990, p. 223; Calhoun Reference Calhoun1994, pp. 9-36) – through different time dimensions (past, present, future) to create a continuous identity. It is never the totality of society, since temporality is only one dimension of human existence.

It should be emphasized that the proposed definition may not be universally valid; different cultures see temporal continuity differently. Rather, it is a working tool to be used by sociologists in their study of traditions in modern and modernized societies. Nonetheless, this definition of tradition can be applied to long-ranged traditions, such as Buddhism or the Catholic carnival, or to short-ranged ones, such as, say, making annual parties on a specific date about a specific theme in a night-club, or a family tradition to wave one’s hand when leaving the house, because “it’s tradition”. Traditionality may be assigned either to beliefs, such as monotheism, nationalism, human rights, hard-working; or to practices, such as voting, making birthday cakes or working hard in the mines. However, the combination of beliefs and practices is most common.

Handler and Linnekin emphasize that the sense of temporality does not necessarily imply total continuity but can also function as conceptual frame for grasping changes and discontinuities. When stories are told, for example, about the foundation of a city, they create a sense of temporal continuity – the name of the city, its location, its infra-structure, and often its “character” or “spirit”. Alongside this, however, the story can also generate a sense of change – the city’s development since its inception until now and into the future. The combination of the story with the particular geographical reference depicts the existence of the city as coming from the past and continuing into the future, i.e. as stretched between past, present and future (e.g. Muir Reference Muir1981, pp. 70-74; Shoham, forthcoming). The listener of the story – whether this is a habitant of the city, an incidental visitor or a tourist who seeks these stories and unwittingly encourages their creation – is getting the impression that the city is set in time no less than in space, i.e., that the city is actually more than what one can see through one’s eyes in the present. This sense of temporality is what we call here “temporally assigned meaning”, and it can be built through multiple cultural sites: objects, rituals, religious dogmas, geographical configurations, oral stories, texts, cultural performances, and leisure activities.

Because social science has been so judgmental of tradition, it is important to see this article as descriptive, not normative. As early as the 18^th century, philosophers of ethics, politics, and science debated whether to reject or accept tradition as a source of norms and knowledge and whether tradition is “rational”.Footnote ¹ Disappointingly, tradition as an analytical category and a tool for sociological research has received far less attention. Here we will examine tradition not as an ideology but as an analytical category, although there will inevitably be some overlap.

We shall begin with a broad overview of a theory that has dominated the field of sociology for decades: the idea that tradition is anti-modern. Here, our focus will be on the legacy of methodological problems that it passed on to contemporary sociology. Next, we will analyze the secondary and less theoretically developed approach that sees tradition as society itself, along with a few middle-ground options that support the theory of tradition as assigned meaning.

Weber and Tradition as the Antithesis of Modernity

According to classical sociology, which saw tradition as an ontological reality, tradition implies the continuity of society through time. To what extent can society experience changes in its institutions, values, landscapes, and population and still maintain its identity? One would expect the concept of tradition, as a way to conceptualize both change and continuity, to occupy a central place in sociological thought. But as Edward Shils argued, “Tradition is a dimension of social structure which is lost or hidden by the atemporal conceptions which now prevail in the social sciences” (Shils Reference Shils1981, p. 7). Did social science indeed lose sight of tradition?

There is no clear answer to this question: tradition has always concerned social scientists, whether as the “Other” of modernity or as the essence of a social system. In both cases, it was basically a transparent concept with no analytical implications.

But Shils inaccurately accused sociology of being atemporal. Indeed, there has been considerable attention in sociology to one temporal concept: modernity. This notion, however, assumed a specific mode of temporality: presentism. Presentism connotes an approach to identity construction that focuses on the present as the dynamic center of individual and collective identity, while eliminating past and future as sources for identity construction.Footnote ² At the same time, it devalued other modes of temporality, making tradition a-temporal by assuming that a past-oriented traditional society is stagnant. In the anti-modern approach, tradition is the diametric opposite of presentist and dynamic modernity. In the totality approach, tradition is the unchanging essence of the changing society. The notion of presentism affected both approaches we have mentioned and, as a result, the anti-modern approach became much more dominant.

The classic thesis of modernization has two essential components: (1) the conceptual opposition of tradition and modernity, and (2) the inevitable triumph of modernity over tradition. Modernization thesis was roundly criticized, but it was only the second statement that was rejected, while the first statement proved tenable and above the fray. Taking tradition as an assigned temporal meaning allows it to have various temporal modes and implies that there is no necessary logical connection between tradition and modernity: the concept of tradition neither necessitates nor negates modernity.

Nevertheless, the conceptual opposition between tradition and modernity has a long and enduring intellectual heritage that dismissed and devalued the concept of tradition. That concept goes back to the Latin word traditio: “handing down” or “delivering”. For the Romans, tradition was the legacy of Greek culture. Later, the Church fathers came to see it as the ultimate authority. During the Reformation, tradition was given the negative connotation of a burdensome mediator between humans and God, biblical texts, and the Greco-Roman classics. These connotations persisted well into the Enlightenment, which saw tradition as the opposite of reason; the ensuing Romantic Age contrasted it to individual creativity. Although Romanticism rejected reason as the exclusive source of judgment, it did not challenge the conceptual dichotomy between reason and tradition. The latter was seen as instinctive; many Romanticist intellectuals relegated it to “non-civilized” (and later unindustrialized) cultures. Although the judgmental tone of discourse has changed, the concept has not.Footnote ³

Social science turned the noun “tradition” into the adjective “traditional” and has always had a deep interest in the difference between the traditional and the modern. Anthropology explained the cultural difference between Western and non-Western societies as a function of temporal difference (Fabian Reference Fabian1983). Sociology was mainly interested in the nature of modernization and the inevitable transformation of a traditional society into a modern one (Eisenstadt Reference Eisenstadt1983). It was Weber who defined the traditional and the modern as two ultimately opposed ideal-types for social action. At the time, “traditional” seemed like a non-judgmental designation for non-Western societies, more appropriate than “primitive”, “uncivilized”, and the like. Weber connected tradition to the habitual mode, which, along with the affective and the rational, comprises the three main types of social action.

Importantly, despite the problems with Weber’s theory, his emphasis on the meaning of a social action for its agent provides a sturdy foundation for a theory of tradition as assigned meaning. There is a striking discrepancy, however, between Weber’s stress on tradition as a pivotal mode of orientation of social action or a legitimization for social order, and the fact that he provides no positive definition of the concept. Weber has no theory to explain how a culture is handed down from generation to generation or how a mode of identity is constructed through time. Rather, tradition is a stagnant habit. Tradition legitimizes social practices that have no justification and rationalizes unreasonable ones; it is simply the antithesis of modernity. All we know about tradition is what it is not: anti-rational, anti-autonomic, anti-individualistic, and so forth (Weber Reference Weber, Henderson and Parsons1964, pp. 115-132, pp. 341-345).

The main property of tradition in Weberian and neo-Weberian thought is its superficiality, i.e., its lack of reflection. “Strictly traditional behavior […] lies very close to the borderline of what can justifiably be called meaningfully oriented action, and indeed often on the other side” (Weber Reference Weber, Henderson and Parsons1964, p. 116). Instead of a positive theory of traditional behavior, Weber invoked Tarde’s theory of imitation: but according to Weber, social action must be meaningful for the agent; it must therefore incorporate at least some reflection. Even saying “I do X because I am used to X, because X has existed since time immemorial” is reflection. In other words, even a low level of reflection can exclude a habitual social action from the realm of tradition (Acton Reference Acton1952-1953, pp. 3-4). This was the basis for Mannheim’s distinction between “tradition” and “traditionalism” – because “tradition” becomes a reflective ideology, it is therefore eminently “modern” and thereby excluded from tradition and traditional society (Mannheim in Wolf Reference Wolf1971, pp. 152-153). In this neo-Weberian school, modernity is understood as an unbridgeable and irreversible split with tradition. Traditional societies themselves do not use it as object for study; only their modernization.

It is crucial to remember that, for Weber, tradition is first and foremost an ideal-type for individual social action. The habitual mode of social action is potentially a fruitful topic for research and analysis. A good example of the habitual mode would be the routinization of social and political practices, which greatly occupied Weber and many of his followers. But routinization was defined by them as a rational mode of orientation for social action, precisely since it does involve reflection. As we will see below, routinization as a temporal mode can be part of tradition.

Weber’s many followers deprived tradition of any reflectivity. Any tradition that demonstrated some measure of reflectivity was derided as “inauthentic” or considered to be “contaminated” by modernity and labeled an “ideology”. Moreover, since change was allegedly a threat to the social order, tradition was characterized by political and cultural stagnancy. Any visible change in a tradition was defined as a deviation from the realm of tradition itself, towards “ideology” in the Mannheimian sense. A notable example is Geertz’s proposal that an “ideology” develops when there is “lack of meaning” due to the crumbling of tradition, with no theoretical account for the concepts of modernity, tradition, and modernization (Geertz Reference Geertz1973, p. 219 and note 41).

The main methodological problem with defining tradition as the “anti-modern” is the problem of discourse. The Weberian conceptualization assumes that, at least as an ideal-type, tradition cannot be self-aware and make sense on its own terms. In other words, tradition can be defined and studied only when there is no discourse about tradition in the society under consideration. Such explicit discourse about tradition puts the society being studied into the mode of reflection. Even if it is only partially reflective, such a society no longer uses tradition and habit as its organizing principles. How can a scholar approach tradition, if discourse about tradition is itself a deviation from tradition and evidence of its decline? If there was ever a real traditional society, how would we know about it? For years, sociologists assumed that there was no longer such a thing as “traditional society”, only nostalgic or residual discourse about tradition, or traditional societies that were already in the process of modernization (Eisenstadt Reference Eisenstadt1983). In other words, discourse about tradition is possible precisely because tradition no longer exists (Bauman, in Heelas, Lash and Morris Reference Heelas, Lash and Morris1996).

Although the debates about the conception of modernity are beyond the scope of this paper, it is important to note that modernity is too often defined as a post-traditional society (Eisenstadt Reference Eisenstadt1972; Giddens Reference Giddens, Beck, Giddens and Lash1994). Ironically, the fact that there is discourse about tradition is understood as a sign of modernization. Modernity was defined as the opposite of tradition, while tradition remained the unanalyzed empty signifier of the anti-modern. Tradition has been continually “crumbling” ever since Weber – otherwise, there would be no indication of modernity.

Methodologically, the problem of reflection pushed the study of tradition towards the “hermeneutics of suspicion”, which assumes that discourse both reveals and conceals something about the nature of things (Ricœur Reference Ricoeur and Savage1970). Sociologists who follow this approach – mostly neo-Marxists – see social life as driven by hidden undercurrents, unbeknownst to its actors. Weber himself by no means adhered to the hermeneutics of suspicion; quite the contrary. Nevertheless, his studies of traditional societies often led him to talk about their “double-standards” of traditional and rational (Weber Reference Weber and Fischoff1963, pp. 108-117).

One school that took this line to its logical extreme was psychoanalysis. In some of his post-World War I works, most notably Moses and Monotheism, Freud defined tradition as a set of repressed religious sentiments, paralleling between universal and individual history (according to the Comtian paradigm of Magic-Religion-Science). Religious traditions are thus repressed in modern civilization, but they may surface as the result of some recurring trauma that disrupts the adolescence of both the individual and the collective. Connecting tradition with repression keeps it temporally continuous and explains why it surfaces from time to time, despite its alleged disappearance from modern life (Freud Reference Freud and Jones1939).

Another approach that adopted the “hermeneutics of suspicion” perspective of tradition is neo-Marxism, which analyzed the Western canon (examined below) as the outcome of the “history of victors”. Walter Benjamin, a neo-Marxist thinker who was acutely sensitive to time, thought that the scholar’s task is to reverse the classical dichotomy and undermine mainstream tradition. They must replace it by developing an alternative view: a tradition of the repressed classes that would connect the repressed of the past to those of the present (Benjamin Reference Benjamin and Zohn1969, p. 259). At the same time, he believed that new means of mechanical reproduction would remove the aura of tradition (Benjamin Reference Benjamin and Zohn1969, pp. 219-254). Both the psychoanalytic and neo-Marxist theories felt no need to incorporate the transmission of practices and ideas between generations and epochs. The temporal continuity of repressed childhood emotions or catastrophic events throughout generations is an almost ahistorical given. By contrast, such temporal continuity is intentionally repressed and concealed by “civilization”. The intergenerational link is recreated only as a scholarly abstraction.

One telling example of the linkage between the “hermeneutics of suspicion” and Weber’s negative concept of tradition is the landmark anthology The Invention of Tradition (Hobsbawm and Ranger Reference Hobsbawm and Ranger1983). Besides its major empirical contribution to the study of how concrete traditions developed from the Industrial Era onward, the anthology also made a significant theoretical contribution. In his introductory essay, Hobsbawm defines tradition as a repetitive practice (we will come back to this later). Consistent with the Weberian conception, however, Hobsbawm saw tradition as a socio-cultural practice that relies on the past as a source for authority or inspiration, in contrast with practices created by routinization, which is characteristic of modern societies. The rise of the reflective discourse of tradition, traditionality, and traditionalism in industrialized societies seemed to indicate the collapse of the allegedly unreflective traditional society. Hobsbawm coined the phrase “invented tradition”, as distinguished from “genuine” tradition, as an intentional oxymoron to show the irreconcilability of modernity and tradition.

Hobsbawm’s neglect of the inherent inventiveness of tradition results from his acceptance of the classic dichotomy as formulated by Weber, no less than from his neo-Marxist hostility toward modern nationalism, emphasized by his critics (e.g. Smith Reference Smith1991; Thompson, in Heelas, Lash and Morris Reference Heelas, Lash and Morris1996, pp. 102-103; Bendix Reference Bendix1997, pp. 211-3). Here neo-Marxist suspicion converged with the Weberian school’s suspicion of tradition.

The discourse about tradition and modernity has focused on nationalism as a strange dichotomous phenomenon. On the one hand, it was an important result of modernization and its leading political principle; on the other hand, it implied a socio-political order based on “primitive” and “irrational” loyalties such as language, race, and religion. If tradition and modernity are understood as diametrically opposed ideal-types, then nationalism is such a peculiar mix (Smith Reference Smith1971, pp. 255-256) that its historical appearance requires particular explanation. For decades, the debate about the “modernity” of nationalism divided scholars between the two sides of the classical dichotomy (Brubaker Reference Brubaker1996; Smith Reference Smith2000, pp. 27-51; Pecora Reference Pecora2001). But nationalism is far from the only instance of the two ideal-types mixing in social reality. Hobsbawm saw the appearance of tradition in the heart of the primary modern political order, the nation-state, as a contradiction proving the impossibility of modern nationalism. But if we adopt the theory of tradition as assigned temporal meaning, nationalism incorporates tradition and modernity not as irreconcilable ontological realities but as different types of temporalities, some “modern” and some “pre-modern”. Later we will explain how this is possible.

Another example of the dominance of the anti-modern definition is the status of tradition in theories of “post-modernity”, where tradition is considered to be an empty signifier that negates cultural change. Bauman (in Heelas, Lash and Morris Reference Heelas, Lash and Morris1996, pp. 49-58) sees the “post-modern” era and its allegedly cosmopolitan political structures (a mostly European perspective) as the diminishing of tradition: but in this case, the traditions referred to are liberal and democratic. In an attempt to devalue modernity, he assigns it traditional characteristics, such as solidarity, necessity, ritualism, and security. “Post-modernity” is assigned the opposite characteristics – alienation, contingency, experimentalism, and risk – which were commonly assigned to modernity.

For too long, the Weberian dichotomy has turned traditional society into a single entity, lumping “together Pygmy hunters and Chinese mandarins, Zande herdsmen and Hindu priests” (Boyer Reference Boyer1990, p. 113). This falsely assumes that tradition could possibly mean one thing in all languages and societies and that the mechanisms of its transmission are universal. Indeed, Boyer (Reference Boyer1990, p. 114) concluded that the prevalence of the classical dichotomy between tradition and modernity has “more to do with an historically dated obsession with the difference between ‘us’ and ‘them’, than with any serious scientific argument” (see also Rudolph and Rudolph Reference Rudolph and Rudolph1967, p. 7; Linnekin Reference Linnekin1991; Adam in Heelas, Lash and Morris Reference Heelas, Lash and Morris1996). For Weber, on the other hand, the anti-definition of tradition is not so much a result of Eurocentrism but of the understanding of tradition as habitual imitation. The Eurocentrism here is more an effect than a cause.

To be sure, the opposition of tradition and modernity is widespread in the literature. It was always prevalent in folklore studies and cultural anthropology, which defined their object as the Other of the modern West, and assumed that the practices of European peasantry and non-European cultures had been handed down as they were from time immemorial. Ben-Amos (Reference Ben-Amos1984, p. 98) contended that “folklore, unlike sociology, has never contemplated replacing tradition with modernity”. Rather, it saw tradition as an authentic mode that was threatened or “contaminated” by the advance of industrialization and literacy (Dorson Reference Dorson1976, pp. 33-34; Burke Reference Burke1987, pp. 15-16). The same held true for anthropology. This fit with classical sociology, which distinguished gemeinschaft from gesellschaft and understood modernization as disenchantment of the world, alienation, and displacement (e.g. Weber Reference Weber, Gerth and Mills1948, pp. 129-156). Folklorists and cultural anthropologists saw tradition as a form of social and cultural backwardness that was far removed from modern life. It was associated with illiteracy and thought to pertain only to small non-industrial communities, like European rural villages or non-European tribes.

Tradition was thus extensively used in these disciplines, but not always reflected upon. It was “a concept to think with, not to think about” (Ben-Amos Reference Ben-Amos1984, p. 97). Rather than being a conceptualization of mechanisms of transmission, it was reified and understood as preserving cultural conventions “from time immemorial” (Pocock Reference Pocock and King1968, p. 112). A few folklorists and anthropologists did notice that traditions do not really stay the same over generations (Ben-Amos Reference Ben-Amos1984, p. 99; Boyer Reference Boyer1990, vii, pp. 8-9). Generally, however, tradition was the definite Other of the modern West.

Co-existence Theory

Weber’s negative definition of tradition as the anti-modern still dominates the field of sociology. But this is not for lack of criticism. The dichotomy between tradition and modernity was first questioned by Shils as early as 1958 (Shils Reference Shils1958). After decades of dominance, modernization thesis provoked a growing discomfort since the 1960s, criticized by many social scientists as too rude, firm, abstract, and Eurocentric.Footnote ⁴ Others rejected the “before-and-after” teleological or even determinist narrative, and instead posited tradition and modernity as dialectical entities that “infiltrate and transform each other” (Rudolph and Rudolph Reference Rudolph and Rudolph1967, p. 3). This led to a search for middle options, dynamic processes, and mixed situations. The theory of “detraditionalization” was gradually replaced by the “coexistence thesis”, which explained the persistence of tradition in the face of modernization (Heelas, Luke and Adam in Heelas, Lash and Morris Reference Heelas, Lash and Morris1996). But while critics disputed the inevitable triumph of modernity, the conceptual opposition between modernity and tradition remained in place.

Shils’ monograph Tradition (Shils Reference Shils1981) was both the most comprehensive articulation of the coexistence thesis and the most developed neo-Weberian theoreticization of tradition as a temporal concept. Shils assumes that no society, no matter how “rational”, can build a knowledge base from scratch; this is why tradition has universal significance (pp. 38-40). Following Weber’s emphasis on meaning, Shils suggests that “tradition is transmission”: i.e., a representation of the past (pp. 12-13). The book mainly suggests an extended and nuanced analysis of various modes of time-binding communication and temporal identity construction. The transmission of tradition may be composed of a variety of elements – emotions, words, structures, perceptions, norms, objects, knowledge, values, or actions – that can be traditionalized and encoded as a transmission from the past (p. 16, pp. 31-32). But to be defined as a tradition, a transmission must endure through at least three generations or two transmissions (pp. 15-16). Tradition is thus dynamic: it depends on contemporaneous means of communication. There is no “traditional society”, only concrete traditions with various statuses in society.

Since tradition has no required content, modernity can also be a tradition. Shils proposes several ironic formulations, such as “traditionality of reason” (p. 21), “tradition of emancipation from traditions” (p. 324) and even “antitraditional traditions” such as “originality, scientism, and progressivism” (p. 235). Nevertheless, he adds that the modern era is the triumph of these “rational traditions” over all the others, which he calls “substantive traditions” (p. 21 ff). Shils does accept the inevitability of modernization, but argues that it entails the decline of “substantive traditions”, while rational traditions prevail. The temporal mechanism of modern tradition is different from that of substantive traditions, since it uses institutional apparatuses of self-criticism. While substantive traditions subjugate the present to the past and assume that the elderly have the most life wisdom, rational traditions tend to prefer the youngsters’ point of view and maintain that the transmissions from previous generations should be constantly re-evaluated.

Arguably, ranking traditions as more or less “substantive”, i.e., traditional, derives from the Weberian dichotomy. Unlike other critics of the modernization thesis, Shils challenged the conceptual opposition of tradition and modernity but accepted the triumph of modernity over all other traditions.

Moreover, Shils’ generational definition of tradition is too narrow, since it excludes many other temporal modes – notably, repetition, as highlighted by Hobsbawm. Shils chooses to overlook the normative and authoritative aspects of tradition and focus on identity and interpretation. For him, the normative power of tradition does not exceed the power of nostalgic memory. There is no society without a past; but in the modern era, the past is losing its power.

Nevertheless, Shils’ amended modernization thesis makes one key contribution to a theory of tradition as temporality. The Shilsian distinction between rational and substantive tradition can be de-ontologized. The distinction between ontological traditions can be converted into a distinction between the two modes of temporality that were empirically revealed by Hobsbawm, but mis-conceptualized as “invented” vs. “genuine”. This distinction has two temporal aspects: (1) Modern invented traditions were based primarily on the calendar and quantitative chronology (jubilees, anniversaries, etc.) rather than on intergenerational transmission.Footnote ⁵ Hence, repetition became a more significant component of the concept of tradition as experienced by modern agents, while the intergenerational component diminished. (2) The temporal frames of meaning of modern traditions widened to include the future dimension. Modern traditions are often invented not only out of an imagined continuity with the past, but also for the sake of an imagined continuity with the future. This was the case, for instance, with the Queen’s Cup football games in 1870s England (Hobsbawm and Ranger Reference Hobsbawm and Ranger1983, pp. 263-307). These annual games were canonized as a “tradition” a few years after their inception not because someone thought football was played in the Middle Ages, but because someone wanted these games to continue well into the future.

The coexistence thesis has moved towards the sociological mainstream. At the same time, there has recently been more recognition of the partiality and/or plurality of modernity in different places, times, life dimensions, and historical contexts. Instead of modernity in the singular, it has become common to talk about “reflexive modernities” (Beck Reference Beck, Beck, Giddens and Lash1994), “alternative modernities” (Taylor Reference Taylor1999), or most frequently, “multiple modernities” (Eisenstadt Reference Eisenstadt2000). These theories challenged the triumphalist version of detraditionalization, pointing to the tenacity of various traditions well into the late-modern era. Nonetheless, they did not propose an alternative conceptualization of the relationship between tradition and modernity; they merely stated that some modernities are less “traditional” than others (Eisenstadt Reference Eisenstadt2000; Wittroc Reference Wittroc2000; Kaviraj Reference Kaviraj2005). This tendency was also evident in the anthology Detraditionalization (Heelas, Lash and Morris Reference Heelas, Lash and Morris1996). Although some contributors criticized the classical dichotomy by pointing out its limits, analyzing the context of its creation, and proposing alternative conceptions, most simply discussed tradition as a vanishing phenomenon (as the book’s title suggests), even while criticizing the classical dichotomy.Footnote ⁶ Research today focuses mainly on the middle ground or dynamic scales between tradition and modernity. Tradition itself, however, is rarely defined as anything but anti-modern.

Since Weber explicitly defined the traditional and the rational as ideal-types, one might say (if such classifications matter) that the idea of multiple modernities is distinctly Weberian. The fact that there is a wide middle ground between, say, East and West, does not mean that the concepts of East and West no longer guide our perception – quite the opposite. The Weberian conception has hardly been challenged.

One of Weber’s useful contributions is that tradition is a mode of social action. But the theory of assigned temporal meaning differs from Weber’s because it sees the traditional mode as meaningful for the agent: it indicates temporality, rather than atemporality. Tradition exists in places where there is discourse about it, rather than the other way around.

Tradition as Totality

But the dominance of the anti-modern approach was not absolute. I turn now to the polar approach of tradition as totality, as found primarily in Durkheimian sociology. Although theoreticians who followed this approach did not always focus on tradition, their approach is more useful for constructing a relevant theory of tradition than that of the anti-modernists. As we shall see, the totality approach allows for tradition as a non-ontological reality. According to this school, modern and non-modern societies adhere to tradition in a similar way. Nevertheless, here as well, modernization and modernity serve as key reference points for writing about tradition.

Durkheim was not particularly troubled by the concept of tradition. He believed that the goal of the newborn science of sociology was to propound static models, rather than to study the dynamics of a society’s movement from one model to another – to focus on the synchronic rather than the diachronic axis (Durkheim Reference Durkheim and Yash1980, pp. 47-57). Durkheim thought that traditional society was simpler than modern society; the former was characterized by a mechanical solidarity, the latter by an organic one (Durkheim Reference Durkheim and Simpson1933). But in his late works, he innovatively suggested a similarity between modern societies and “lower societies” in which “Everything is common to all” (Durkheim [Reference Durkheim and Cosman1912] 2001, p. 7). This abstract similarity was cited to support his theory that everything is social. The common basis of traditional and modern societies are non-temporal concepts such as “culture” or “religion,” rather than “tradition”. Like tradition for primitive societies, where the “individual type almost overlaps with the general type” (ibid.), modern culture provides little if any space for agency and change, a quality shared by both traditional and modern societies.

This Durkheimian overstatement found successors in sociology and anthropology in the United States, a country where the philosophical heritage of pragmatism had already assigned tradition a central place in society. Humanist anthropologists such as Benedict (1959) understood the concept of tradition as synonymous with the concept of culture, and argued that just as everything is culture, so is everything tradition or custom. This assertion was a form of the cultural relativism that was already latent in Durkheim’s The Elementary Forms of Religious Life.

System theory in North American sociology developed a somewhat more sophisticated conception. Parsons argued that any society, including the modern, has a cultural tradition that assures social cohesiveness by functioning as a supreme value-system, and thus limits the possibility of radical change. Moreover, Parsons defined a “self-subsistent social system” as a system that has

Without mentioning the word “tradition”, Parsons anticipated the aforementioned Shilsian inter-generational definition of tradition and made a distinction between “societies” – which operate through the temporal mechanisms of intergenerational communication – and “partial social systems”, which do not. Although clearly influenced by Durkheim, Parsons deviated from him by establishing temporal criteria for social systems. But, once again, the concept itself could not be used as an object of study and research.

The most comprehensive conceptualization of tradition within the early Durkheimian school was that developed by Maurice Halbwachs. Though he wrote in the interwar era, his work was little read as long as functionalism was dominant and did not attract serious attention until the cultural turn. Among the theoreticians surveyed here, Halbwachs is the farthest from defining tradition as an ontological reality.

In his investigations of memory, Halbwachs, following Durkheim, suggested that it be seen as a social category (as opposed to individualistic conceptions in the psychology of his time). Consequently, he suggested focusing on the act of remembering, that is, the social construction of the past. Halbwachs defined Memory as a phenomenon that:

Halbwachs draws a firm distinction between the present and the past, which nevertheless complement each other, like form and matter, to construct collective memory. However, the conventions of the present sift past materials, so that most are forgotten. The dough of tradition is constantly kneaded and reshaped by the requirements of the present. It is not really handed down from the past, but constructed, destructed, and reconstructed in the present.

Nonetheless, the raw materials for assembling the collective memory can be found only in the past. If a tradition does not suit the present need, “principles can be replaced only by other principles, and traditions by other traditions” (p. 185). In Halbwachs’ description, the group can widen the frame of group identity and reconstruct a broader collective memory. In other words, the sense that a particular tradition is inadequate to the present needs of society is derived, not from “the present”, but from the synchronic encounter with another tradition, which is more cosmopolitan and consequently perceived as being more rational: “Reason is opposed to tradition as an extended society is to a narrow one” (p. 184). Halbwachs suggests that we understand modernization as a broadening of social space; today we would speak of “globalization” and the unification of different traditions. We are dealing here with a spatialized version of the modernization thesis.

This neo-Durkheimian framework suggests a nominal answer to the question posed above: how does a society keep a sense of continuity while changing? The answer is: selectively. But it erects too rigid a distinction between past and present, while failing to account for the temporal dynamics that turn a present into a past. In other words, it does not explain how this change is possible in terms of tradition itself. It seems that traditions can change only as a result of encounters with other traditions. Within itself, a tradition lacks internal transformative power. Moreover, it is not clear why some traditions are discarded while others are preserved. Why are certain traditions perceived as Other, while others are adopted?

Halbwachs offers an overly presentist approach that understands tradition as the social construction of the past, and bypasses the issue of continuity and discontinuity by treating tradition as nominal. Instead of studying tradition, he would study the discourse about tradition. To make Halbwachs’ definition more workable, we can break it down into parts, following Douglas, who defined collective memory as a “flexible, articulated set of social segments” (Douglas Reference Douglas1982, p. 268). This segmented definition implies that representations of the past appear in certain cultural sites and can be socially framed. Traditions are understood as incomplete phenomena that occupy a concrete and delineated social space and can be studied as such. The distinction between past and present is understood and studied as socially constructed rather than ontological chronology that is “out there”. This structuralist propensity provided historians and social scientists with theoretical tools for fruitful conversations about memory agents and memory sites (e.g. Nora Reference Nora and Goldhammer1997; Olick Reference Olick2003; Olick, Vinitzky-Seroussi and Levy 2011), in which the harmonizing tendency of Durkheim and Halbwachs was moderated and tradition studied as an arena for negotiations about identity.

So far we have depicted the anti-modern as the principal approach to tradition, even though it mainly hindered the study of traditions, with tradition as totality as the subordinate approach (at least after functionalism lost its dominant position in sociology). We have seen that the second approach would see tradition as socially constructed and without a sufficient account for the temporal mechanisms of transmission. In order to arrive at a theory of tradition as an assigned temporal meaning, we now turn to theories that saw tradition as canon and will present analytical models for analyzing encounters between different types of traditions, perceived as “Self” and “Other”.

Tradition as Canon

In the humanities, tradition has had a different meaning from that of the social sciences: it is neither estranged nor totalized. Western intellectuals revered the humanistic tradition as the most significant source of their own socialization into the world of knowledge, but as by no means synonymous with society. The emphasis in their definition of tradition was more on long-persisting tradition as a significant accumulation of knowledge that cannot be ignored.

The approach of tradition as canon was imported into the social sciences by Robert Redfield in the 1950s, when he developed a new dual model for studying what he called the “great tradition” and the “little tradition”. The great tradition is synonymous with civilization and defined as comprising urban, literary, learned, elite, and specialized bodies of knowledge. The little tradition, by contrast, encompassed local traditions that are mainly rural, oral or unwritten, and are “for the most part taken for granted and not submitted to much scrutiny or considered refinement and improvement” (Redfield Reference Redfield1960, pp. 41-42). The great tradition is by nature urban, literate, and elitist.

Redfield reformulated the classic dichotomy of modernity and tradition as a conceptual opposition between a great tradition and a little tradition (although he thought that capitalist industrialization threatened great traditions as well). The former consisted of a literary canon that encompassed accumulated knowledge and conscious transmission; the latter was an oral tradition that covered lesser social terrains. To further sharpen the link between tradition and civilization, Redfield turned the classical image upside down and held that it is precisely “primitive society” (to which he attaches all the atemporal stereotypes, Redfield Reference Redfield1953, pp. 14-15) that lacks tradition, because it is illiterate and has no history. By contrast, a civilization has a history (though not necessarily in the Western sense), which is played out over a broad region (Redfield Reference Redfield1960, p. 59). The little tradition can survive in the realm of the great tradition, and interact with it, only with the support of the latter.

Redfield’s theory is of particular interest for us here, because he proposed a model that integrates tradition as the “Self” and tradition as the “Other”. His understanding of great traditions has a surprising resemblance to the classical Western notion of tradition, while his concept of little traditions relies on the folkloric understanding of tradition as the “Other” of modernity. He suggested that these two types of traditions conduct a dialogue and interchange, borrowing and influencing each other in various ways.

However, as a member of the Chicago School, which was mainly interested in urban life, he focused on great traditions (Redfield and Singer Reference Redfield, Singer and Sennett1969) and suggested a rather naïve understanding of tradition as a linear accumulation of knowledge (Smith Reference Smith1962). This linear notion was challenged as early as the 1960s, by the studies of the humanistic and scientific traditions by Gadamer ([Reference Gadamer, Weinsheimer and Marshal1959] 2002) and Kuhn (Reference Kuhn1962), respectively. Working independently, both Gadamer and Kuhn suggested that agents always begin their interpretation of the world from inside a tradition that provides them with the linguistic and cognitive tools required for their hermeneutic or scientific project. Both of these thinkers defined tradition as a chain of cultural products. They parted company with the traditionalist position, however, by taking the meanings of the traditional canon as open and dynamic, varying as a function of changes in the interpreter’s or scientist’s historical context, and not as just an accumulation of knowledge.Footnote ⁷

It is unclear whether these more sophisticated conceptions of tradition qua canon as an open and dynamic debate, rather than as a linear accumulation of knowledge, influenced Redfield’s student and colleague, Milton Singer, when he conducted fieldwork in the city of Madras, in southern India, on the encounter of the great tradition with modernity. Discovering that the image of the Hindu tradition as stagnant is wrong, he provided an elaborate description of the rich mechanisms of cultural metabolism whereby Hinduism traditionalizes cultural change, including the encounter with modernity (Singer Reference Singer1972, e.g., pp. 389-390, p. 404).

Redfield’s integrated model influenced sociologists and political scientists, on the one hand, and folklorists and cultural anthropologists, on the other, in different ways.

1. Sociologists and political scientists developed the axial civilizational approach to explain the tenacity of traditions in the face of modernization processes all over the world (Eisenstadt Reference Eisenstadt1986). Echoing Parsons’ definition of a social system, Arjomand defined civilizations as “coherent units through time because they relate to their past by means of continuous interpretation and codification, through their historical memory, and by the canonization of certain texts – in short, through the formation and transformation of tradition” (Arjomand Reference Arjomand2010, p. 372). And one of these civilizations was “the civilization of modernity” (Eisenstadt Reference Eisenstadt2001).

Within this paradigm, the axial civilizations – Hinduism, Confucianism, Islam, Christianity, Buddhism, and perhaps several othersFootnote ⁸ – are described as structurally similar to the Western intellectual tradition, self-consciously and dynamically cultivated by intellectual elites, like the humanistic and scientific traditions analyzed by Gadamer and Kuhn. But the tradition/civilization is also taken here as an entire social system, that is, as a cohesive object that can be studied from the axial age until the present, with the assumption that there is a nuclear canon that remains the socio-cultural core of civilizational identity through the generations. Whereas Shils defined a minimum of at least three generations for there to be a “tradition”, Eisenstadt and his followers set an infinite maximum and contemplated that a tradition might endure for almost all of known human history.

This mode of analysis – which gained renewed popularity with Huntington’s (Reference Huntington1996) controversial thesis of The Clash of Civilizations – excludes most traditions, great and small, oral traditions and newly invented traditions, which do not satisfy the criteria required of the axial civilizations. Is there really only one sub-Saharan African civilization, as Huntington suggests, with a unified value-system, let alone an institutionalized cultural canon and common political structures? Using the civilizational approach, only the five to ten axial traditions, of all those in the world, are qualified to deal with modernity. Eisenstadt later elaborated the civilizational approach into the model of multiple modernities, but in fact they were not as multiple as they might have been. The civilizational approach carried the notion of tradition as a historical-ontological reality, with a stable core enduring throughout the generations, to its logical extreme.

2. At the same time, and separately, folklorists and cultural anthropologists focused on little traditions and their dynamics when they come into contact with larger social bodies. Little traditions were increasingly studied in urban contexts (Paredes and Stekert Reference Paredes and Stekert1971), a tendency in line with the recent trend of “backyard anthropology”. It was found that there is no transmission of tradition without modification, reconstruction, and invention (Honko and Laaksonen Reference Honko and Laaksonen1983; Handler and Linnekin Reference Handler and Linnekin1984; Hofer Reference Hofer1984; Boyer Reference Boyer1990). No longer are the commercial aspects of folklore dismissed as contaminating its authenticity; instead, they are considered to be integral aspects of this phenomenon, no different than any other (Becker Reference Becker1998).

More interestingly, conspicuous aspects of modern life came to be viewed as “folklore”, giving rise to genres such as “urban folklore” and “office folklore”. Children’s songs and games, too, were studied in this light. Scholars found a “folk” character in the mechanisms of their transmission, including oral delivery and the absence of an author, even in the heart of industrialized and literate society (Dundes and Pagter Reference Dundes and Pagter1975). This approach to tradition protected against the reification of tradition. Undoubtedly modern notions such as the “social contract” or the “civilization versus wilderness” are morphing into myths and stories that have been handed down from the past to the present and traditionalized (Shoham, forthcoming). In other words, there is no necessary logical connection between tradition and modernity – neither negation nor affirmation.

Although Redfield’s theory adumbrated the interconnections between great traditions and little traditions, it defined two very different life forms as traditions without asking why the same term is used to designate both of them. What do great and little traditions have in common, if their mechanisms of transmission are so different? The study of tradition as assigned meaning, rather than as ontological reality, encourages further investigation of the conceptual links between the two. To the best of my knowledge, no one has as yet picked up the virtual gauntlet hurled by Redfield.

Epilogue

To date, sociology has tended to understand tradition as an ontological reality and has focused on tradition’s self-identity over time; that is, how much can tradition change and still remain tradition? This question led to an analytical dead end, because social systems are always changing but staying the same. In its ontological version, tradition is taken to be the stagnant core of society, immune to time.

The alternative paradigm of tradition as assigned meaning does not imply that traditions do not really exist. Rather, it allows sociologists who are interested in the social construction of time (Durkheim Reference Durkheim and Cosman2001, pp. 10-12; Zerubavel Reference Zerubavel2003) to make good use of the concept by studying traditions as various modes of temporality. These traditions, however, are only segments – such as rituals, stories, texts, and landscapes – that are set into a temporal relationship with the past or the future. They may be salient (Boyer Reference Boyer1990), but they are not all-encompassing.

In addition, different cultures have different perceptions of transmission and assign different temporal meanings to it. The intergenerational communication between past and present, which receives too much emphasis in the literature about tradition, is far from being the only possible mode of temporality. Other modes are just as relevant.

First, tradition is not embedded in any particular perception of time. It is not exclusively linear, as in Redfield’s definition, or cyclical, as implied by the definition of tradition as the antithesis of the modern. Tradition can be either of them, combine the two, or even integrate other temporal modes, such as nostalgia, futurism, and decline (Koselleck Reference Koselleck and Presner2002, pp. 148-149; Zerubavel Reference Zerubavel2003, pp. 46-48). Revolution, too, can be a mode of tradition, as in Shils’ “tradition of liberation from tradition” and in revolutionary traditions such as radical socialism and certain streams of radical nationalism. Moreover, as I have argued elsewhere, even presentism can turn into a traditional paradigm (Shoham Reference Shoham2009).

One mode of temporality that is too often disregarded in the literature about tradition is quantitative temporality, characteristic of the invented traditions of modern industrialized societies that mainly work with micro-temporal tools such as calendars, timetables, and cumulative sequences of years, days, or practices (Young Reference Young1988). Many of these traditions were anniversaries of places and institutions, birthdays or death days of important figures, or practices that are counted in linear sequence – such as volumes of academic journals, Super Bowls or Eurovision competitions. This punctuated temporality emphasizes the repetitive element of tradition, whose power is expanded, by means of ritual, from the realm of identity to the level of institutional authority and, to some extent (especially in modern nationalist traditions), to norms.

Second, the communication of tradition across the different time dimensions can work in both directions. Some traditions bind past and present, others bind present and future; but most traditions bind all three together (Giddens Reference Giddens, Beck, Giddens and Lash1994, p. 62; Smith Reference Smith2007, p. 111). Most notably, many modern practices, such as annual academic and political conferences, are future-oriented repetitive invented traditions.

Handler and Linnekin’s definition unlinks the discussion about tradition from the problem of authenticity, by accepting that every tradition must have been invented at some time and by someone – whether great minds in the Axial Age or the tourist industry of the 20^th century; what is more, tradition is constantly being reinvented (Handler and Linnekin Reference Handler and Linnekin1984). Here we have tried to further liberate the concept of tradition from its entanglement with the West and its Others. Some of the scholars mentioned above defined tradition as the ultimate negation of modernity; others assigned it the qualities of the Western canonical tradition and then defined non-Western civilizations as traditions in this sense. Redfield produced a theory that incorporates both senses of tradition, but with only partial success, because he gave scant attention to the transformations of little traditions within modernity. This gap has been filled mainly by folklorists, whose suggestion that tradition be treated as assigned meaning should be adopted by sociologists who are interested in the temporality of organizations, states, families and individuals.

Working from the definition of tradition as assigned meaning, I suggest that the focus be shifted from the processes of the integration of great and little traditions to the integration of different meanings of tradition. That is, tradition as assigned meaning is not just an interpretive framework to which an agent is socialized, as in the scientific or humanistic tradition from which civilizational analysis implicitly borrowed its notion of tradition. In modern life, it is often also an object that an agent encounters synchronically as the Other qua Other. It is precisely within a presentist attitude, or a “dynamism and temporalization of the experienced world” embedded in modernity (Koselleck Reference Koselleck and Presner2002: 168), that tradition may be socially produced as an immigrant from “the foreign country” of the past (Lowenthal Reference Lowenthal1985). The demarcation of tradition within modern life – in both the First World and the Third World – constructs ostensibly rigid borders between past and present, in order to mark artifacts, objects, and practices as originating in the past. Most notably, it is in the tourist and souvenir industries that tradition often appears as the ultimate negation of the hectic modern lifestyle. It is precisely this exoticization that makes both tradition and objects designated as “traditional” attractive (Becker Reference Becker1998). The same qualities attracted social scientists from all disciplines, including folklore studies, to tradition, even when they assumed that it was about to disappear or already had.

Traditions are essential to social life, which has a temporal dimension. Traditions dialogue with each other, borrow materials, ideas, and institutions, and often develop through mutual interaction. Such cultural encounters are often conceptualized as meetings of different traditions: oral and literate, great and little, civilizational and folk, modern and non-modern. The paradigm of assigned temporal meaning suggests viewing them as encounters between different modes of temporality – long-range and short-range, past-oriented and future-oriented, cyclical and linear, fragmented and continuous – in all their possible permutations and combinations, along with those of different social units. For example, there are conceptual interconnections between the spread of birthday rituals in the Industrial Age and the simultaneous dissemination of invented national traditions (Pleck Reference Pleck2000). Thus the sociological discussion about tradition should not invoke either “Self” or “Other”. Rather, it should acknowledge the complexity of the links among all these temporal configurations, so that it can productively conceptualize the persistence of tradition in modern life.