Toward a Theory of Public Ritual

March 13, 2000

275. "Toward a Theory of Public Ritual," Sociological Theory, Vol. 18, No. 1 (March 2000), pp. 40-59.

Given that holidays both reflect a society's attributes and serve to modify these attributes, they are a valuable tool for a macro-sociological analysis. This paper proceeds by examining Durkheim's well-known contributions on rituals and advancing theoretical ideas on how these might be modified, seeking to develop a theory of holidays. 


The article concerns the role of holidays in managing tensions and recommitment to values; their role in relating communities to the society at large; their effect on gender roles, and the theoretical issues concerning holiday cycles and holiday-engineering efforts by religious authorities and states that have endeavored to adapt holidays for their own purpose. The article relies on public accounts, personal observations, and findings culled from a few studies by contemporary social scientists.


Sociologists have long paid much attention to the family as the unit that initiates the socialization process and, in this context, to the effects that changes in family composition and structure and the rise of other child care institutions have had on the outcomes of socialization. Considerable attention has also been accorded to schools as agencies of socialization. In the last few decades, leaders and members of society at large have addressed these matters using code words such as "family values" and "character education." In contrast, little consideration has been given by sociologists and the public to the role of holidays and rituals as significant elements of the institutional foundations that undergird socialization and resocialization. I do not mean to suggest that holidays and rituals are as pivotal as families and other sources of child care including educational institutions, only that their role seems greater than the attention accorded to them in recent decades. For instance, the term "holiday" does not even appear in the index of the 16 volumes of the otherwise rich and elaborate International Encyclopedia of Social Sciences (1968); it is not listed in the index of the American Sociological Review nor the American Journal of Sociology from 1975 to 1995, nor in the Encyclopedia of Sociology (1992). I am of course not arguing that there have been no sociological studies of holidays; indeed, several rather fine ones are cited below. I am only pointing out that they have been relatively few in number compared to other areas of study.

This article attempts to lay foundations for a sociological theory of holidays; our state of knowledge may not suffice to advance an actual theory at this point. Holidays are defined as days on which custom or the law dictates a halting of general business activity to commemorate or celebrate a particular event (American Heritage Dictionary 1996:862).(1) For the purposes of this paper, "holidays" will be used to imply both ritual and holidays because it has the same basic similarities to rituals in the role they play in society.

The work proceeds by examining Durkheim's well-known contributions (1995) and suggesting ways these may need to be modified to develop the theory. In addition, the article raises theoretical issues not directly addressed by Durkheim, particularly those that concern the prevalence of a holiday cycle (annual, repeated sequence) and holiday-engineering efforts by religious authorities and states that have sought to modify holidays to their purpose but without undermining their "sacred" legitimation. In the process, the article draws on public accounts, personal observations, and findings culled from a few studies by contemporary social scientists. (Durkheim and others frequently refer to rituals rather than to holidays. However, their observations and findings apply to holidays because they are a form of rituals.)

Because Durkheim's work has been systematically analyzed, well reviewed and effectively summarized (Lukes 1972, 1975; Miller 1996; Parsons 1937; Wallace 1986), very little is to be gained by a re-analysis of Durkheim here. The main relevant points, following Durkheim's functional approach, are briefly:


(a) Profane (secular), routine, daily life, the conduct of instrumental activities at work and carrying out household chores, tend to weaken shared commitments to shared beliefs and social bonds, and enhance centrifugal individualism. For societies to survive [these] centrifugal, individualistic tendencies, they must continuously "recreate" themselves, by shoring up commitments to one shared ("common") set of beliefs and practices.

(b) Rituals provide one major mechanism for the recreation of society, one in which the members of a society worship shared objects and in which they share experiences that help form and sustain deep emotional bonds among the members.

(c) The specifics of the rituals, and the objects that are being worshiped or celebrated in these rituals, be they colored stones or woodcuts or practically anything else, have no intrinsic value or meaning. It is the society that imbues these objects with significance, and, thus endowed, they become the cornerstones of the integrative rituals built around them.

From this viewpoint, religious services during weekends serve to reinforce the commitments that have been diluted during weekdays. Holidays, in this context, are seen as supra-weekends, as especially strong boosters of commitments and bonds.

To put it somewhat more formally, Durkheim hypothesizes that rituals, holidays included, correlate negatively with societal disintegration (defined as excessive individualism) (Lukes 1975:292).(2)

Before I can both build on this theoretical starting point and seek to modify it, a methodological digression is needed.


Holidays have a special methodological merit that makes them particularly attractive to students of societies: They provide indicators of the states and attributes of large collectivities. In studying societies social scientists often rely on measurements based on aggregate data about myriads of individuals, objects or transactions such as public opinion polls, economic statistics, and census data (Lehman 1977). For various reasons not explored here, such aggregated data are best supplemented with indicators that tap directly into collective attributes of the macro social system under study, sometimes referred to as "global" measurements (Etzioni and Lehman 1967). For instance, to characterize a polity it is useful to determine not merely the voting and alienation rates of millions of citizens, but also to establish whether there are two dominant political parties or numerous small ones and whether there is a written constitution. These are significant macro (or "system") attributes not derivable from data about the attributes of citizens or from other aggregates.

In studying the belief systems of a society by using global indicators, researchers have drawn on various cultural products, such as books considered to "typify" a particular group (e.g., novels by William Faulkner as reflecting the "mind" of the South), select movies (e.g., A Clockwork Orange), and major speeches by public leaders (e.g., Martin Luther King Jr. and Louis Farrakhan). Using such cultural products as indicators seems unavoidable when one deals with earlier periods such as ninth century feudalism in France, from which only a very limited amount of such material is available. However, for contemporary societies such cultural products are available in huge quantities and their content varies greatly. Hence, choosing a particular book, movie, or speech as indicative of a period and system of beliefs is often difficult to justify, and analyzing all of them (or even a random sample) is an onerous task (even when setting aside problems that arise from questions regarding the proper universe to be sampled, for example, whether one should include imported books and movies). In contrast, the ways holidays are celebrated-whether their focus is nationalist, militarist or religious, whether they are dominated by merchandising and conspicuous consumption or dedicated to public service, whether they take place in people's homes or in public spaces, and so on-provide relatively telescoped and hence economic global indicators. Moreover, data about the ways holidays are celebrated in a particular society are often available about countries in which one cannot conduct reliable public opinion polls nor gain many other kinds of aggregate data, Tibet for instance.

Such usages of holidays as indicators might be rather misleading if the focus only on the content of the occasions, for instance how religious versus secular they are. To the extent possible, it is very desirable to determine the rough number of the participants, the extent of their involvement, and any specific attributes that distinguish them from other members of the society. For instance, observation of a Promise Keeper ritual revealed that most who attended were male and white (Escobar and Murphy 1997).

To point to the merits of observing holidays as a source of data and insight into the belief systems of a society is not to argue that such data is fully accurate or can be relied upon as the sole source of evidence. Holidays, like other cultural products, tend to offer a somewhat refractory reading of society (Shils 1953). Thus, one cannot rely on the finding that many in India watch MTV to conclude that they are becoming Americanized. Indeed, studies show that when members of different ethnic groups watch the same TV program, they see rather different things (Katz and Liebes 1990). Nor can one necessarily conclude that because Times Square in New York City, and several other such sites in major cities, are crowded with jubilant people on New Year's Eve-that the country is in a happy mood. I merely suggest that observations about holidays provide one major and relatively accessible source of global data about the beliefs and other attributes of a given society. The resulting readings, like those of other indicators, need to be compared with other sources of data.

Finally, it should be noted that, unlike many other measurements (GDP levels, for instance), which are social science constructs or artifacts, holidays are both a social phenomenon of considerable sociological interest in their own right as well as a source of data on a society at large. Hence, those who study holidays reap a double benefit: They advance our understanding of a set of specific social phenomena, and they cast light on the community or society of which they are a part. All this comes to support the suggestion that holidays deserve more sociological attention than they have received in recent decades, and point to the merit as well as the limits of using holidays as macro-sociological indicators.


To move beyond the Durkheimian starting point one must recognize that Durkheim basically treats all rituals-holidays included-as if they were of one kind, in the sense that they all fulfill one societal "function"; they all foster integration by reinforcing shared beliefs (Durkheim 1995: book III). To put the same proposition in somewhat different sociological terms, holidays serve to socialize members of a society as well as to reaffirm their commitments to values, and as such serve to sustain the integration of society. (Socialization, it is widely recognized, is not limited to the young; adults are continuously socialized in the sense that social processes and resources are dedicated to recommit adults to existing beliefs or introduce them to new ones.)

While it may be true that all holidays serve a socializing function to one extent or another, I suggest that (a) different holidays play different societal roles; indeed no two holidays serve the same societal role, and (b) not all holidays are integrative. (The last proposition assumes that one holds constant one's frame of reference; a holiday may be integrative for one group or another, but not all holidays are integrative for the society at large, as Durkheim suggested.)

To proceed, one must first note that there is no agreed upon typology of holidays to draw on, let alone one based on the societal roles fulfilled by various holidays. Some scholars have arranged holidays by the seasons they mark (Santino 1994); others have called attention to the holiday's role in the lives of the individuals involved (rather than to the societal roles of holidays) (Bellah 1992); still others see holidays as largely historically shaped. I attempt here to provide a typology based on the varying societal roles fulfilled by different holidays.

Probably the most important distinction among holidays from this viewpoint is those that use narratives, drama, and ceremonies to directly enforce commitments to shared beliefs-which I shall refer to as recommitment holidays-and those that fulfill this role indirectly, by releasing tensions that result from the close adherence to beliefs, which I term tension management holidays(cf. Parsons, Bales and Shils 1953:180-181). One and the same holiday may serve in both ways, but a very preliminary and informal survey by the author suggests that in those societies the author is familiar with, each holiday serves in one way more than in the other. (However, such a primacy is changeable over time, as we shall see below.)

Recommitment holidays are most familiar; they are the ones Durkheim had in mind and are commonly associated with his integration thesis. Easter and Passover are typical holidays of recommitment. Easter dramatizes and extols the essential message of Christianity: the resurrection of Christ, the joy and fulfillment of redemption, and rebirth and reaffirmation of faith. The holiday is marked by specific and elaborate rituals, such as services at sunrise and rousing music to celebrate the Resurrection. Passover is built around the reading of the Haggadah, a narrative openly dedicated to socialization with special focus on children. The associated Seder ritual is rich with symbols that entail reaffirmation of one's commitment to the beliefs of political liberty, deliverance by a supra-natural force, and those of a separate cultural identity and tradition. The implied sociological-Durkheimian hypothesis is that those who share the particular Christian or Jewish beliefs at issue and who participate in these holidays will be more committed to the shared beliefs and institutions of their respective communities after such participation than they were before. (To test this proposition by comparisons of intensity of commitments to beliefs of those who participate in the said rituals to those who did not is somewhat more difficult, because of the effects of self-selection).

While holidays of recommitment are expected to directly serve the socialization and hence societal integration, holidays of tension management are expected to serve societal integration indirectly and pose a higher risk of malfunction. Holidays of tension management include New Year's Eve, Mardi Gras, Purim, Oktoberfest and their equivalents in other cultures. (Whether or not Halloween belongs in this category is a matter of considerable conjecture, which cannot be resolved here. On this matter, see Best and Horiuchi 1985; Best 1985; and Williamson 1982. April Fools Day also belongs in this category although it is hardly a holiday.) During these holidays, mores that are upheld during the rest of the year are suspended to allow for indulgence, and some forms of behavior usually considered asocial and hence disintegrative are temporarily adopted. Anthropologists report that in some societies, there are holidays in which daughters-in-law are permitted to insult their mothers-in-law. New Year's office and other parties are known for suspending mores against excessive consumption of alcohol and sexual infidelity (Turner 1985). During Mardi Gras in New Orleans thousands of students expose themselves on the balconies in the French Quarter. Jews, usually warned against interrupting their studies of scriptures for idle chatter and even appreciation of aesthetic beauty, are allowed to play games on Purim.

Tension management holidays are expected to contribute to reinforcement of shared beliefs and institutions indirectly, by releasing tension that results from conformity to societal beliefs and the behavioral prescriptions they entail. The underlying theoretical assumption is that people cannot be fully socialized and that sublimation is not fully successful, and that hence there will be a significant and accumulative residue of alienation to all commitments, even if they are not imposed by some foreign power or cultural or political elite (Freud 1989 and Wrong 1994). Occasional release of this tension is expected to enhance socialization (including resocialization).

The extent to which a tension release holiday, which occurs every few months, actually serves to vent alienation, or at least keep it at a low level, has, as far as I can establish, not been a subject of empirical study. Thus, there seems to be no systematic evidence that people who return to work after New Year's Day have a stronger commitment to their duties than before, though Santino (1994:49) suggests that people do return to work "psychologically refreshed" after the holidays. (He does not suggest how long such "refreshment" lasts). The proposition that holidays fulfill such a role, therefore, should be treated largely as a hypothesis rather than as a finding.

Tension management holidays that set clear time limits are expected to be more integrative than those that do not set such limits. Time limits refer to points in time after which participants are expected to return to the conformist modes of behavior that reflect shared societal beliefs. It is as if society fears that once its members experience the raw satisfaction that results from suspended mores, members might be reluctant to return to the tighter restraints entailed in social roles. (A similar point has been made about people recovering from an illness who are reluctant to return to work because of the "secondary gains" of being sick, being legitimately exempted from numerous duties from working to attending to household chores. Physicians--and work rules about sick pay--are used to curb such tendencies.) Thus, Mardi Gras is followed by a recommitment on Ash Wednesday. New Year's Eve is followed in the United States by one day of vacation (New Year's Day) but there is a rather clear expectation that on the following day people must return to work. Several religions set a clear time limit on mourning and for the conclusion of other rituals, as Judaism does at the end of the Sabbath. In contrast, for secular mourning there is no clear time limit.

In testing the preceding hypothesis and others that follow, one must take into account that while holidays are largely of one kind or the other, they are not pure types. A recommitment holiday may include some opportunities for indulgence and suspension of mores (for instance, those entailed in work), and tension management holidays may include some recommitment (for instance, prayers) but the dominant activities tend to be of one kind or the other.

Critics may question whether a functional analysis of holidays biases the analysis in a conservative direction. Indeed, the analysis so far has focused on one societal role, that of evolving and sustaining commitments to prevailing societal beliefs and the behavioral expectations they contain. We shall see below that holidays also serve as opportunities for societal change, by providing occasions to symbolize and embody new conceptions of social relations and entities. This is illustrated later in the development of new roles for women in religious holidays, especially for rituals in the home.

I turn next to show that holidays can and are employed to strengthen the commitments to a great variety of societal entities rather than to uphold the dominant society; these entities include social movements, ethnic or racial groups, deviant religious denominations and still other societal entities that challenge established regimes or seek to form new societies. Like pipelines, holidays are largely indifferent to the content of the normative "flows" they facilitate.


While as a preliminary approximation it might suffice to propose that various attributes of holidays (such as their prevalence, number of participants, the strength of recommitment they evoke) correlates negatively with societal disintegration, this proposition must be further specified. Because many, if not all, holidays may have some kind of an integrative function, I hypothesize thatnumerous holidays that help integrate some societal entities have the opposite effect on others. Returning to my starting point, to the work of Durkheim, one notes that he focused on the integrative function of holidays for whole societiesand for societies that did not have significant internal divisions, at least none that he recognized. This approach is best understood if one takes into account not only the utilitarian and individualistic bodies of thought Durkheim was challenging, but also that his empirical base observations were drawn from studying Australian totemism. That is, he was dealing with a very small society with a rather high level of homogeneity and integration.

A theory of holidays applicable to complex societies needs to abandon the assumption that holidays are necessarily unifiers of societies and specify which social entity they are serving. Instead, it should be assumed that: (a) while holidays do provide an integrative mechanism, this mechanism may work to solidify member groups and not necessarily a whole society (indeed, an examination of the holidays of a given society can serve as an indicator of the level of unity of that society); (b) the integrative effects of holidays on the society will depend on the relationships among such groups and the society-at-large, which can vary from confrontational to complementary, and that (c) these relationships can in turn be changed, and tensions "worked through" (on this concept see Yankelovich 1991) during holidays. Some illustrative observations follow which give some preliminary and limited support for the preceding propositions.

An example of a holiday that encompasses and unifies most members of a community is provided by observations of the celebrations of Passover in early small and homogenous Israeli kibbutzim. In contrast, in contemporary American society, one finds, in addition to some holidays that are widely, although not universally, shared on a national level (for instance, Memorial Day), numerous holidays that are specific to one religious group or another (e.g., Christmas and Hanukkah, studied by Waits 1993 and Wilde 1995), or to one ethnic group or another (e.g., St. Patrick's day, Kwanza, Cinco de Mayo, Chinese New Year), and-much less often-to one class (May Day).

If there were systematic evidence indicating whether or not all these various differences in the ways holidays are practiced by various social entities have increased or decreased over a given period (for instance, during the second half of the 20th century) in a given society, this would provide a global indicator of the extent to which that society, in that period, had become more fractionated or more unified. For instance, one would expect that the rising diversity and intergroup tensions in American society between 1960 and 1990 would be reflected in a rising diversity in the way holidays are celebrated by various ethnic and racial groups. Lacking systematic evidence to this effect, one must either rely on rather casual and informal observations to this effect, or regard the preceding statements as strictly theoretical propositions.

Holidays can serve to modify the relationships between societal parts and the whole, to the extent that such a whole exists and is recognized. In some ceremonies, conscious and systematic efforts are made to ensure that a particular group holiday (one observed only by a given group) does not undermine the commitment to the whole. One symbolic expression of the commitment to the whole is that displaying flags during parades, ceremonial speeches, and prayers. When a group in the United States seeks to reinforce its members' commitments not only to the group but also to the society at large (on layered loyalties see Etzioni 1996: 202-3), the celebrants often display both ethnic flags (e.g., Italian) and the Stars and Stripes. Most speakers, religious functionaries, and community leaders--if the said disposition is sought--are careful to include ". . . and God bless America" or some other such expression, to indicate their loyalty to the whole, stressing that their group upholds dual loyalties, and hence, its particularist commitments do not conflict with its commitment to the national society.

Clearly not all particularistic holidays are celebrated in such a dual (group andsociety building) way. Some group celebrations are disintegrative for the society as a whole, are openly oppositional and challenge the societal mores and symbols, or even are outright expressions of a breakaway from the societal whole or from some other group. Native Americans, especially the Wampanoag, used Thanksgiving as a protest holiday, a day of fasting and mourning (National Public Radio 1993). Several New York City schools also adopted this protest perspective, teaching children that Thanksgiving marks a day on which ". . . 'strange looking' people . . . landed in the family's backyard and proceeded to ransack their homes, cut down their trees, kill their pets, and take tomatoes from their gardens" (Belluck 1995; Mack 1996). Intense debates about Columbus Day mark similar societal divisions, with some seeing the day as a recognition of a great discovery of the man who opened the door for the creation of the American society, while others view the day as celebration of a brutal killer and conqueror who should not be lionized (Belluck 1996).

Finally, holidays can be used to work out a new relationship between society and a member group, and in the process advance and ritualize a change in the beliefs of those involved. That is, integration is achieved as Durkheim would expect, but in a way he did not expect: by changing the beliefs around which society congeals. (The point is not that Durkheim did not recognize changes in symbols and rituals, but that his theory did not allow for changes in the level of societal integration that are achieved, in part, by changing the content of holidays, as is hypothesized here.)

Some illustrative examples follow. Initially, the idea that there should be a holiday marking Martin Luther King, Jr.'s birthday met with considerable opposition, especially from conservative Americans, including President Ronald Reagan. After 15 years of considerable lobbying by civil rights groups, a federal holiday was finally declared in 1986. Several states continued to oppose the declaration of this new national holiday, and were only gradually won over. Conservative Arizona was particularly slow to join the other states in this regard. New Hampshire chose to follow its individualistic streak and instead declared a Civil Rights Day.

A much more successful working through of an intergroup difference helped by and reflected in a holiday, can be seen in the participation of Vietnam veterans in parades and other activities on Veteran's Day. Initially, the refusal of large segments of society to treat Vietnam vets as returning heroes, as individuals who served their country at par with those who fought in the First and Second World Wars, caused many to be reluctant to participate in these parades. Nor could they expect a warm welcome from other veterans who did march (Estes 1994). However, as the rift over Vietnam gradually healed in the early 1990s, Vietnam veterans participated much more often in these parades, were more warmly received by the public at large and by other veterans, and felt more positive about participating in these national ceremonial occasions (The New York Times 1995:32). In Chicago in 1994, Mayor Richard Daley conducted a ceremony that officially honored Vietnam veterans (Caro and Mixon 1994). In 1995, the city council of Berkeley, California, once a center of anti-Vietnam War protests, held a commemoration on Veteran's Day honoring those who fought in the Vietnam War (The New York Times 1995:32).

All this suggests that, to advance a theory of holidays, one must strongly modify Durkheim's notion that holidays serve to integrate a society and recognize that, while they may solidify member groups, the relationships of these groups to the whole may vary from being highly complementary and hence integrative to rather conflictive and disintegrative. The relationship may range from one of a group that is self-aware and well-defined yet deeply integrated into a more encompassing whole, to hostile relations or even civil war (Kurds in Turkey, for instance). Additionally, groups that started as conformist may draw on reinterpretation of holidays to aggrandize their distinctiveness, while hostile subgroups--and the societies in which they are situated--may reformulate holidays as part of the reconstruction of society, its core beliefs, and its relations to the member groups.

In short, (a) whether or not a given holiday is integrative cannot be assumed by merely observing that a holiday is celebrated, even if it is well attended and celebrants are deeply involved. In addition, one needs to specify the reference unit: Which entity is being integrated, the society as a whole or some other social entity, or both; (b) the nature of the relationship between this entity and the society needs to be examined, as reflected in the given holiday and otherwise; and (c) one needs to introduce a dynamic perspective that calls attention to the fact that holidays that originally integrated a group into society as a whole may change to undermine that bond, or--that holidays that originally served to divide, may help work through differences to enhance not only the internal integration of a particular group but also of the society as a whole. For instance to illustrate the latter point, before 1917 Christmas seems to have been, relatively speaking, an integrated holiday for Russian society. It ceased to be so after the Soviet Revolution, although it's clearly far from vanished.


Durkheim assumes that holidays are public events, in the sense that members of tribes assemble as one group in one space, and the rituals involved are shared by and visible to one and all. It hence follows that this integrative function cannot be served, at least not as well, if holidays are celebrated privately by individuals (or their families) in their huts, shanties, or suburban homes rather than in public squares, parade grounds, or other points of assembly. Parsons suggested a different interpretation of Durkheim in this regard, suggesting that holidays can be integrative even if celebrated privately as long as the commitment to the society at large is recognized, a point discussed later (Parsons 1937:438-441).

Durkheim's hypothesis finds some initial support in numerous reports of communal rather than individualized celebration of holidays in non-literate tribes as well as in Western societies, at least in the preindustrial eras. For instance, in 18th century rural America, when society was more homogenous and most communities were much smaller than contemporary ones, holidays were often celebrated in public spaces. For example, typical Fourth of July celebration centered around a parade that ended up in a church, in which the crowd was blessed and shared a communal meal (Appelbaum 1989).

In contrast, in recent decades, many holidays have been celebrated in private homes, by families or other small groups (Nissenbaum 1996; Caplow et al. 1982). The Fourth of July is now often played out at private picnics, outings with friends or co-workers, or in backyard barbecues. And it has become less committing. It is popularly reported that Thanksgiving initially was a communal holiday which, it is reported, Pilgrims and Native Americans celebrated together, but long ago has become mainly a family holiday. "By the mid 19th-century, Thanksgiving had become associated with homecoming . . . Returning home for Thanksgiving was both a metaphor and a ritual performance of solidarity, renewing or validating family ties" (Siskind 1992:175).

Still, it is far from established that the privatization of holidays is necessarily disintegrative. There seems to be some evidence to suggest that private rituals can engender recommitment to the society at large as do public, shared rituals. Religions, for instance, differ in the extent to which they rely on rituals that take place in shared and public spaces, such as Sunday rituals in churches, as compared to those that build on family-based rituals that take place in private homes, such as the kind that play a major role in Judaism. There is, however, no immediate evidence that Jewish holidays are less integrative than other ones, although to the extent that I could find out, this topic has not been subjected to empirical research.

True, holidays are rarely if ever completely private. One tends to mention to coworkers and friends, even shop keepers, about the occasion and its special meaning. But this of course cannot make up for the decline in participation in shared communal events.

A related question is whether the increased privatization of holidays reflects an adaptation of holidays to a decline in the level of integration of society or vice versa, whether a privatization of holidays helped cause a decline in societal integration as Durkheim would have it.(3) So far, most observers have depicted the increase in privatization as driven by societal factors, as opposed to the other way around. Among the factors cited are a decline in public safety and an increase in the level of heterogeneity (Rested 1995; Waits 1993) and the decline of close (or "primary") relations in growing cities (Schmidt 1995: 30). That is, the increase in private holiday celebrations are said to reflect a general decline in bonding that ties small social entities into a society and in the level of individual commitments to this society. As well, they reflect a growing loyalty to particularistic groups (ethnic, religious, etc.), small intimate social circles, and emphasis on personal achievement and interests. While there is little solid evidence to support these hypotheses, it should be noted that according to several accounts, individualism rose between 1960 and 1990 in American society, the same years holidays have become less public.

It should also be noted that since 1990, following increased concern about the rising level of individualism and the decline of public spaces, limited attempts have been made to restore the public nature of holidays. For instance, a community in Ontario, California annually sets up a two-mile long table, around which many members of the community, from divergent social backgrounds, are reported to picnic together on the Fourth of July (Ontario, California Chamber of Commerce 1996). Furthermore, there are some reports about a rise in efforts to celebrate holidays in "artificial" extended families for single parents, other singles, gays and lesbians, and still others who seek communities that are more extensive than their households, especially to share holidays (Stronghart 1994). All this provides some very tentative support for the hypothesis that societal changes drive holiday changes rather than holidays changing society.


To advance the sociological theory beyond Durkheim's work, one must determine the sociological significance of the fact that in most, if not all societies holidays are repeated over time and in the same sequence. While I was unable to find a comparative study directly focused on this point, an informal survey of numerous cultures suggests that a fixed and annual sequence of holidays is one of the most robust observations one can make based on cross-cultural comparisons, in the sense that it can be seen in societies that differ a great deal on numerous other accounts.(4)

The question for the sociologist is: What is the societal significance of the particular sequence in which holidays are arranged? To suggest that holidays follow the climatic seasons per se may be true, but casts little sociological light on the matter, unless one uncovers the social reasons some seasons are ritualized while others are ignored.

One interesting sociological hypothesis that has been advanced in this context deals with what might be called holiday sub-cycles. For instance, Williamson (1982:235) noted that holidays focused on children--Christmas for instance--is preceded and followed by festivities that are built around aggressive, sexual, adult themes (e.g., Christmas is preceded by the Christmas office party and followed by New Year's Eve). We have already noted the difference between recommitment and tension management holidays; the hypothesis should be added here that these two kinds of holidays will "alternate," rather than holidays of one kind being followed by more of the same kind.

The underlying assumption of both Williamson's observation and our hypothesis is that the serving of various societal needs is arranged in a sequence so that after one need is attended to, attention to the other gains predominance. The question of this hypothesis, developed on the basis of informally observing a few holidays, can be extended to all holidays, requiring a macro analysis of societal needs to determine if they are all served by one holiday or another. This is a task that remains to be faced. Only after this is completed can one seek to determine whether societies whose holidays are "properly" sequenced show a lower level of social tension, a higher level of integration, and a greater commitment to their shared values than societies whose holidays are out of sequence for one reason or another, perhaps as a result of the attempts at social engineering introduced by totalitarian governments and more recently by religious fundamentalist ones.

VII. DESIGN AND TRADITION: Holy-Days Compared to Civic Celebrations

Durkheim often treated societies and their holidays as given or as evolving under the impact of unfolding historical and social forces, rather than as subject to deliberate societal change (on the implications of this difference see Etzioni 1968). Modern societies, though, are engineered to some extent, and are subject to public policies that attempt to change the relations among racial, ethnic, and economic groups. As holidays serve to solidify the society at large, the member groups, or both, attempts to change these entities, and above all the values around which they congeal, has led numerous public authorities to deliberately recast holidays in order to change the beliefs of one or more member groups and their orientation to the encompassing society. Given, however, that holidays must build on the legitimation of tradition and affective attachment founded on shared memories and histories (Warner 1959), the question faced by public authorities, and of much interest to the sociological study of change, is whether holidays lose some or all of their power to reinforce commitments to values once they have been openly redesigned. As religions are typically more traditional than secular ideologies (if only because the latter tend to have significantly shorter histories), one promising place to start the examination of the limit and opportunities of social re-engineering of holidays is to ask whether secular holidays are less degraded by being recast than religious ones.

While Durkheim recognized the difference between religious rituals carried out by non-literate tribes and those national holidays that Robert Bellah analyzed as examples of "civil" religion, to Durkheim these were but two forms of religion, fulfilling the same essential integrative function. For Durkheim, while a society could replace God as the source of sacredness, this did not secularize the icons, cults, and rituals the society endowed in this peculiar manner. They were all sacred (Durkheim 1995; for additional discussion see Schoffeleers and Meijers 1978:35-39).

Contemporary sociologists might well find it necessary to draw a distinction between holidays that are built around sacred religious objects and those that surround revered secular objects, including days that mark national liberations, independence, armistices, or civic occasions such as New Year's Eve. Like many other typologies, reference here is not to pure types but to the dominant elements. Thus, a secular holiday may include some prayers, but the main focus of the objects held in special awe, the values dramatized and ritualized, are secular. In the same vein, a largely religious holiday may include some secular elements. Once this is taken into account, one can formulate the hypothesis that, counter to common sense, religious holidays can be more readily redesigned, without losing their legitimacy, than secular ones.

In the Soviet Union, continuous, systematic, and deliberate efforts were made to secularize holidays. Christmas and Easter were abolished in 1920, and November 7 (the anniversary of the founding of the Bolshevik government) and May 1 (the day celebrating the unity of labor) were introduced as holidays. In 1929 Sunday was abolished to make for a six day work week. Gift exchanging was moved to New Year's Eve, and a secular "Father Frost" replaced Santa Claus (Stanley 1996). These efforts were widely resisted. Religious holidays continued to be observed by millions, often although not always in secrecy. Sunday was restored in 1940. And while the Soviet regime did not survive, the celebration of religious holidays did.

A case in point would be the Catholic Church shift from the use of Latin to the use of the vernacular, symbolizing a move from strong universalism to greater openness to local cultural differences, as a part of a much larger attempt to downgrade the central role of authority and place greater emphasis on communal elements of the church.

Traditional Jewish holidays were profoundly changed by Conservative and Reform Jews, attempting to make the religious rituals more accessible to congregations whose members by and large do not understand Hebrew, have been less willing to participate in prolonged rituals, insist on some measures of gender equality, and otherwise seek to reconcile their Jewish commitments with other normative ones, including feminist and gay agendas (Convey 1999). (Also most modern versions of the Haggadah drop or reinterpret a line that calls on God to wreak his vengeance on "other" people.)

While debates continue within these and other religious groups about the appropriateness of deliberate efforts to redesign holidays, and there have been some efforts to restore traditional features in recent years, many of the new modes of celebrations are very widely followed. And, while there has been a loss in the intensity of commitment, it is far from clear that the changes in the way holidays are celebrated contributed to, rather than helped stem, this loss.

Establishing the extent to which holidays can be deliberately recast, and the comparative effectiveness of various ways and means of such endeavors, is particularly significant in light of a challenge posed by Gillis (1996). He argues that American holidays, rituals, and myths are lagging behind reality: that they represent a distorted view of a society that is long gone, especially the notion that there was and ought to be one "traditional" kind of family. He argues-drawing on Marxist and Freudian ideas about the possibility of developing a higher level of consciousness-for a profound change of our core myth (and hence holidays and rituals), toward a formulation that provides a higher level of reality-testing and a more genuine expression of our true feelings and psychic needs. One may argue that this is what is taking place, under the influence of various consciousness-raising and reeducation drives, despite opposition by traditional groups. However, until more research is conducted on these questions, such a conclusion is at best tentative.


Durkheim, dealing with small homogeneous societies, assumed that they were rather monolithic; he characterized whole societies either as well integrated or as suffering from lack of integration. However, it should be hypothesized for complex societies that change in some sectors will lag after developments in others. And, based on some very preliminary and informal observations to follow immediately, a plausible hypothesis is that holidays tend to lag rather than lead societal change, and the more they lag the more they hinder rather than enhance societal integration. The reason is that some members of society are likely to be more involved in the leading segment of society (for instance, those younger than 50) while others might be more involved in the lagging segments (those 65 or older). Hence, the greater the sectorial lag, the more tensions one would expect between the social groups involved.

This hypothesis can be illustrated by changes in women's roles in the preparation for and celebration of holidays. Women's roles in holidays seem to have been akin to their roles in other parts of the socialization infrastructure. Thus, it is hardly surprising to learn that in traditional American society, for instance in the 1950s, women were charged with preparing the celebratory meals, shopping for gifts, promoting the holiday spirit and so on (Schmidt 1995). Similar accounts are available from some earlier periods; for instance, the Pilgrim dinner, to which native Americans were invited and which laid the foundation to the first Thanksgiving in 1623, is reported to have been prepared exclusively by women (Linton 1949).

As the feminist movement started to challenge women's roles in the 1960s, their holiday roles were also recast, but these changes seem to have lagged behind other changes in society, and/or there has been a significant measure of regression toward traditional mores during holidays. While there seems to be no systematic evidence to support the preceding statements, informal observations suggest that even in those households where women work outside the home and husbands assume some household and child care responsibilities, women still seem to do a disproportionate share of the inviting, planning and preparing, cooking and serving of holiday meals, and above all are expected to be the ones to ensure the warm glow of the holiday spirit (Williamson 1982).

All this is not to suggest that changes in women's societal roles have not been reflected in the internal structure of holidays. Some women lead religious services in both communal and home-based rituals in several religious denominations and some have been ordained for posts from which they previously had been excluded: a fair number of Jews have added Bat Mitzvah (for girls) to Bar Mitzvah, and rituals for naming girls to parallel the Bris; there have been several attempts to edit sexist language out of traditional texts, for instance by reading "she" (or alternating between "she" and "he") whenever the Lord is referred to as masculine. But these changes often followed rather than led role changes in other societal areas and seem less extensive than changes in work roles, for instance.

One of the least noted but rather indicative changes in the ways holidays are celebrated and rituals are performed is the splitting of traditional gender roles, reflecting changes in the structures of the nuclear family and more widely, in the kinship structures. Where formerly only one father-role and one mother-role existed, there are now several versions of each, variously referred to as "natural" versus "social" fathers, parents versus stepparents, "biological grandparents" versus step-grandparents and so on. The increasing complexity of roles and relations is reflected in the inclusion in family holidays of a large number of people who are related neither by blood nor marriage, but by former marriages, what one sociologist half-jokingly calls the "x-kinship structure." The parents of former spouses (if they are grandparents of one or more of the stepchildren) are sometimes included. (One student reported that his Christmas celebration as a child included both his divorced parents and his father's "new wife"). A considerable etiquette has been developed about who "gives away" The bride (e.g., The natural father or the stepfather?) and the place of the natural mother during the wedding ceremony and in rituals such as confirmations and Bar and Bat Mitzvahs when there is a stepmother. These examples indicate role changes, as the society is struggling to come to terms with new gender and related family structures. But again, these changes are not widely recognized, codified, or institutionalized, lending further support to the hypothesis that the internal restructuring of holidays lags after major societal changes, and hence in this regard hinders rather than serves integration, as the adaptation of one part of the society--holidays--lags behind the others.


Holidays provide a valuable tool for a macro-sociological analysis of societies because they both reflect various attributes of societies and their major member units (i.e., are effective macro indicators), as well as serve to modify these attributes (i.e., constitute forces of societal change). To develop such an analysis suggestions are provided for the foundations of a sociological theory of holidays that, in several instances, significantly modify Durkheim's hypotheses, and add factors he has not considered.

The main propositions that arise from the discussion are:

I. If one uses the society as the frame of reference, tension management holidays are expected to be less integrative then holidays of recommitment. While both might contribute to the integrative state of a societal entity, tension management holidays are more prone than others to foster antisocial behavior.

II. A theory of holidays will benefit from taking into account that one and the same holiday may have differential effects on the integration of the society at large as compared to that of some member units. For instance, some ethnic holidays strengthen the communal bonds of members units but undermine the societal integration, while other holidays help to reinforce not merely the integration of the member units but also their relations to the society at large. Moreover, a holiday that started as fragmenting may serve as a societal process that helps work through conflicts among the member units and the society at large.

III. The extent to which holidays that have been previously celebrated in public are privatized is expected to correlate with the rise of diversity in society. Moreover, the privatization of holidays that used to be public will undermine societal integration, in conflict with Durkheim's expectation that all holidays will serve to enhance this kind of integration.

IV. The particular reasons holidays are arranged in a given repeated sequence (or cycle) in a given society are not known. A preliminary hypothesis suggests that holidays follow the sequences observed in order to change the societal need various holidays serve, so that sooner or latter most or all needs are served. For this reason, for instance, tension management holidays and holidays of recommitment tend to alternate rather than a set of one kind will be followed by a group of the other.

V. In contrast to common sense intuition, religious authorities seem to be more effective at deliberately changing holidays without alienating large numbers of their followers than civic authorities.

VI. Changes in gender and kinship roles during holidays are expected to lag behind changes in these roles in other institutional sectors, undermining the integrative role of holidays.

In short, once we abandon Durkheim's assumption of a close, positive correlation between the occurrence and participation in holidays and societal integration, we are on our way to laying the foundations for a richer and more realistic sociological theory of holidays.


Abrahams, Roger D. and Richard Bauman. 1978. "Ranges of Festival Behavior," in Barbara A. Babcock ed. The Reversible World: Symbolic Inversion in Art and Society. Ithaca, NY: Cornell University Press.

American Heritage Dictionary of the English Language-3rd edition. 1996. Boston: Houghton Mifflin Company.

Appelbaum, Diana Karter. 1984. Thanksgiving: An American Holiday, an American Tradition. New York: Facts on File.

______. 1989. The Glorious Fourth: An American Holiday, an AmericanTradition. New York: Facts on File.

Bella, Leslie. The Christmas Imperative: Leisure, Family and Women's Work. Halifax, Nova Scotia: Fernwood Publishers, 1992.

Bellah, Robert. 1967. "Civic Religion in America." Daedalus 96:1-21.

Belluck, Pam. 1995. "Pilgrims Wear Different Hats In Recast Thanksgiving Tales." The New York Times, 23 November.

"Berkeley Makes Peace With Its Veterans." 1995. The New York Times, 12 November:32.

Best, Joel. 1985. "The Myth of the Halloween Sadist." Psychology Today 19:14.

Best, Joel and Gerald T. Horiuchi. 1985. "The Razor Blade in The Apple: The Social Construction of Urban Legends." Social Problems 32:488-99.

Borghatta, Edgar F., ed. 1992. Encyclopedia of Sociology. New York: Macmillan.

Caplow, Theodore. 1982. "Christmas Gifts and Kin Networks." American Sociological Review. 47:383-392.

Caplow, Theodore, Howard M. Bahr, Bruce A. Chadwick, Reuben Hill, and Margaret Holmes Williamson. 1982. Middletown Families: Fifty Years of Change and Continuity. Minneapolis: University of Minnesota Press.

Caro, Mark and Bernie Mixon. 1994. "Veterans Get A Big Salute as Holiday is Commemorated Amid Pomp, Protest." The Chicago Tribune, 24 November.

Convey, Eric. 1999. "Spirituality: Passover by the book." The Boston Herald, 26 March:2.

Durkheim, Emile. 1995. The Elementary Forms of Religious Life. Translated and with an introduction by Karen E. Fields. New York: Free Press.

Epstein, Eve and Frank James, "The Native American Perspective of Thanksgiving," interview by Bob Edwards, Morning Edition, National Public Radio, 24 November 1993.

Escobar, Gabriel and Caryle Murphy. 1997. "Promise Keepers Answer the Call." The Washington Post, 5 October:A1.

Estes, Jack. 1994. The Graying of Bitterness and Pain. Los Angeles Times, 30 May.

Etzioni, Amitai. 1968. The Active Society: A Theory of Societal and Political Process. New York: Free Press.

Etzioni, Amitai. 1996. The New Golden Rule: Community and Morality In A Democratic Society. New York: Basic Books.

Etzioni, Amitai and Edward Lehman. 1967. "Some Dangers in 'Valid' Social Measurements: Preliminary Notes." The Annals of the American Academy of Political and Social Science 373:1-15.

Freud, Sigmund. Civilization and Its Discontents. 1989. Translated and edited by James Strachey and with a biographical introduction by Peter Gay. New York: W.W. Norton.

Gillis, John R. 1996. A World of Their Own Making: Myth, Ritual and the Quest for Family Values. New York: Basic Books.

Humphrey, Theodore and Lin T. Humphrey, eds. 1988. We Gather Together: Food and Festival in American Life. Ann Arbor, MI: UMI Research Press.

Johnson, Sheila K. 1971. "Sociology of Christmas Cards."Transactions/Society.

Katz, Elihu and Tamar Liebes. 1990. The Export of Meaning: Cross-Cultural Readings of "Dallas". New York: Oxford University Press.

Kolarz, Walter. 1961. Religion in The Soviet Union. New York: St. Martin's Press.

Lehman, Edward. 1977. Political Society: A Macrosociology of Politics. New York: Columbia University Press.

Linton, R., and W. Linton. 1949. We Gather Together: The Story of Thanksgiving. London: Abelard-Schumann.

Lipset, Seymour Martin. 1996. American Exceptionalism: A Double-Edged Sword. New York: W.W. Norton.

Litwicki, E.M. 1992. Visions of America: Public Holidays and American Culture.Ph.D. diss., University of Virginia.

Lukes, Steven. 1972. Émile Durkheim: His Life and Work. New York: Harper & Row.

_______. 1975. "Political Ritual and Social Integration," Sociology 9:289-308.

Mack, Tara. 1996. Listen Up, Pilgrims: Teachers Modernize Thanksgiving Message. The Washington Post, 27 November.

Malinowski, Bronislaw. 1922. Argonauts of The Western Pacific: An Account of Native Enterprise and Adventure in The Archipelago of Melanesian New Guinea. With a preface by Sir James George Frazer. London: G. Routledge & Sons.

Miller, Watts. 1996. Durkheim, Morals and Modernity. Montreal: McGill-Queen's University Press.

Nissenbaum, Stephen. 1996. The Battle For Christmas. New York: Alfred A. Knopf.

Ontario, California Chamber of Commerce, interview with David Carney of The Communitarian Network, 2 July 1996.

Parsons, Talcott. 1937. The Structure of Social Action. Glencoe, Illinois: The Free Press.

Parsons, Talcott, Robert F. Bales, Edward A. Shils. 1953. Working Papers in the Theory of Action. New York: The Free Press.

Radcliffe-Brown, A.R. 1922. The Andaman Islanders. Glencoe, Illinois: The Free Press.

"Reintroduction of Christmas Hailed." In Daily Report/Soviet Union. [Foreign Broadcast Information Service Database]. New Canaan, CT: DataNews, Inc. - cited 7 January 1991.

Restad, Penne L. 1995. Christmas in America: A History. New York: Oxford University Press.

Santino, Jack. 1994. All Around The Year: Holidays and Celebrations in American Life. Urbana: University of Illinois Press.

Schmidt, Leigh Eric. 1995. Consumer Rites: The Buying and Selling of American Holidays. Princeton, NJ: Princeton University Press.

Schoffeleers, Matthew and Daniel Meijers. 1978. Religion, Nationalism and Economic Action: Critical Questions on Durkheim and Weber. Assen, The Netherlands: Van Gorcum.

Shils, Edward. 1969. "Dreams of Plentitude, Nightmares of Scarcity." in Lipset and Altbach, Students in Revolt. Boston: Houghton Mifflin Company.

Siskind, Janet. 1992. "The Invention of Thanksgiving: A Ritual of American Nationality," Critique of Anthropology 12n2.

Stanley, Alessandra. 1996. December 25 in Russia: The Adoration of the Monetary. The New York Times, 24 December.

Stronghart, Amy Adams Squire. 1994. Developing Your Own Holiday Traditions.St. Louis Post-Dispatch, 27 December.

Turner, Victor. 1985. The Ritual Process: Structure and Anti-Structure. Ithaca, NY: Cornell University Press.

Vidich, Arthur J. and Joseph Bensman. 1958. Small Town in Mass Society.Princeton, NJ: Princeton University Press.

Waits, William B. 1993. The Modern Christmas in America: A Cultural History of Gift-Giving. New York: New York University Press.

Warner, William Lloyd. 1959. The Living and The Dead: A Study of Symbolic Life of Americans. New Haven: Yale University Press.

Wilde, Anna Day. 1995. "Mainstreaming Kwanzaa." The Public Interest 119:68-79.

Williamson, Margaret Holmes. 1982. "Family Symbolism in Festivals." inMiddletown Families. Minneapolis: University of Minnesota Press.

Wrong, Dennis. 1994. The Problem of Order: What Unites and Divides Society. New York: The Free Press.

Yankelovich, Daniel. 1991. Coming to Public Judgment: Making Democracy Work in a Complex World. Syracuse, NY: Syracuse University Press.

1. Steven Lukes defines ritual as "rule-governed activity of a symbolic character which draws the attention of its participants to objects of thought and feeling which they hold to be of special significance" (Lukes 1975:291 Italics in original).

2. Reviewers of a previous draft of this essay pointed out, quite correctly, that Durkheim is subject to different interpretations on the points discussed here. Given that the purpose of the essay is not to sort out the differences among these interpretations but merely to use one of them as a starting point, I will not delve into the question of what is the correct interpretation of Durkheim.

3. For yet another view see Caplow et al. (1982) where the authors argue that the trend towards family at the center of holidays reflects the fact that family is the institution that is most at risk in the Middletown community.

4. To conduct the informal survey I queried six anthropologists about all the cultures with which they are familiar, and additionally, scanned the literature and the cultures regarding the cultures with which I am familiar.