Rule Enforcement Without Visible Means: Christmas Gift Giving in Middletown Author(s): Theodore Caplow Source: American Journal of Sociology, Vol. 89, No. 6 (May, 1984), pp. 1306-1323 Published by: The University of Chicago Press Stable URL: http://www.jstor.org/stable/2779184 . Accessed: 23/12/2014 15:50 Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at . http://www.jstor.org/page/info/about/policies/terms.jsp . JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org. . The University of Chicago Press is collaborating with JSTOR to digitize, preserve and extend access to American Journal of Sociology. http://www.jstor.org This content downloaded from 147.251.68.31 on Tue, 23 Dec 2014 15:50:47 PM All use subject to JSTOR Terms and Conditions Rule Enforcement without Visible Means: Christmas Gift Giving in Middletown' Theodore Caplow University of Virginia As part of a much larger study of social change in Middletown (Muncie, Ind.), a random sample of adult residents was interviewed early in 1979 about celebrations of the previous Christmas. This paper describes the unwritten and largely unrecognized rules that regulate Christmas gift giving and associated rituals in this community and the effective enforcement of those rules without visible means. A theoretical explanation is proposed. The Middletown III study is a systematic replication of the well-known study of a midwestern industrial city conducted by Robert and Helen Lynd in the 1920s (Lynd and Lynd [1929] 1959) and partially replicated by them in the 1930s (Lynd and Lynd [1937] 1963). The fieldwork for Middletown III was conducted in 1976-79;2 its resultshave been reported in Middletown Families (Caplow et al. 1982)and in 38 published papers3 by various authors; additional volumes and papers are in preparation. Nearly all this materialis an assessmentof the social changes that occurred between the 1920s and the 1970s in this one community, which is, so far, the only place in the United States that provides such long-term comprehensive sociological data. The Middletown III research focused on those aspects of social structuredescribed by the Lynds in orderto utilize the opportunities for longitudinal comparison their data afforded, but there was one important exception. The Lynds had given little attention to the annual cycle of religious-civic-family festivals (therewere only two inconsequential references to Christmas in Middletown and none at all to Thanksgiving or Easter), but we found this cycle too important to ignore. The celebrationof Christmas, the high point of the cycle, mobilizes 1 I am greatly indebted to Cleva Maggio, Carmen Matarazza, Steven Nock, and Margaret Williamson for able and imaginative assistance with the Christmas survey. Requests for reprints should be sent to Theodore Caplow, Department of Sociology, University of Virginia, Charlottesville, Virginia 22903. 2 Supported by the National Science Foundation grant SOC 75-13580. Investigators were Theodore Caplow, Howard H. Bahr, and Bruce A. Chadwick. 3A complete list of these publications is available on request from any of the investigators. (C 1984 by The University of Chicago. All rights reserved. 0002-9602/84/8906-0002$01 .50 1306 AJS Volume 89 Number 6 This content downloaded from 147.251.68.31 on Tue, 23 Dec 2014 15:50:47 PM All use subject to JSTOR Terms and Conditions Middletown almost the entire population for several weeks, accounts for about 4% of its total annual expenditures, and takes precedence over ordinary forms of work and leisure. In order to include this large phenomenon, we interviewed a random sample of 110 Middletown adults early in 1979 to discover how they and their families had celebrated Christmas in 1978. The survey included an inventory of all Christmasgifts given and received by these respondents. Although the sample included a few very isolated individuals, all of these had participated in Christmas giving in the previous year. The total number of gifts inventoried was 4,347, a mean of 39.5 per respondent. The distribution of this sample of gifts by type and value, by the age and sex of givers and receivers, and by gift-giving configurationshas been reported elsewhere (Caplow 1982). The following were among the findings: (1) Four out of five Christmas gifts went to kin, and four out of five of these to close kin. (2) Fifty-seven percent of all gifts were a part of a multiple gift, that is, two or more gifts from the same giver(s) to the same receiver(s), and 59% of all gifts werejoint, that is, from more than one giver or to morethan one receiver. (3)The proportionof each class of kin relationshipsmarked by Christmas gifts and the value of those gifts were roughly proportionateto the closeness of the kin relationship. (4) Women were much more active as gift givers than men; they selected most of the gifts given jointly by couples, gave more gifts singly than men, and did nearly all of the gift wrapping. (5) Although married women were largely responsible for Christmas gift giving, they did not favor their own relatives over their husbands'. Gifts to maternal relatives did not differ significantly in number or value from gifts to paternal relatives. (6) In gift giving, close affinal relatives were equated with the linking consanguineous relative. For example, gifts to daughters-in-lawwere as numerous and valuable as gifts to marriedsons. (7)The flow of gifts between adults and childrenwas heavily unbalanced. The respondents, all adult, gave about seven times as many gifts to children as they received in return. (8) Residential distance, which has a major effect on most forms of contact between kin, has only a minor influence on Christmas gift giving. In an earlier paper I undertook to account for those features of Middletown's Christmas gift-giving system that seemed most distinctive: the heavily unbalanced gift giving from parents to children, which does not change when the children are grown; the equal treatment of affinal and consanguinous relatives; and the lack of interest in exact reciprocity in gift giving between kin. Drawing on the ethnographic literature, I suggested that ritualized gift giving, in any society, is a method of dealing with important but insecure relationships, whereby gifts are offered to persons or collectivities whose goodwill is needed but cannot be taken for granted; and I showed how this formula seems to explain the features mentioned above. 1307 This content downloaded from 147.251.68.31 on Tue, 23 Dec 2014 15:50:47 PM All use subject to JSTOR Terms and Conditions American Journal of Sociology In this paper, I discuss a quite different problem: How are the rules that appearto govern Christmasgift giving in Middletown communicated and enforced? There are no enforcement agents and little indignation against violators. Nevertheless, the level of participation is very high. Here are some typical gift-giving rules that are enforced effectively in Middletown without visible means of enforcement and indeed without any widespread awareness of their existence: THE TREE RULE Married couples with children of any age should put up Christmas trees in their homes. Unmarried persons with no living children should not put up Christmas trees. Unmarried parents (widowed, divorced, or adoptive) may put up trees but are not required to do so. Conformity with the Tree Rule in our survey sample may be fairly described as spectacular. Table 1 shows the distribution of Christmas trees by family situation in our respondents' households. Of the 45 married respondents with children under 18, only two had no tree. One was a newly married woman who had spent the entire Christmas season with her husband's parents in another state. The other was a recent immigrant from Venezuela who omitted the tree to demonstrate her refusal to be assimilated: "Wetry to keep our own culture,"she told the interviewer in explaining why she and herhusband had set up a nativity scene instead. Of the 36 married respondents with children who were adults, only three lacked a tree. Two were away from home for the entire Christmas season; the third, a 69-year-old woman whose husband had been hospitalized recently, had broken up house-keeping and was living with a married daughter who had her own tree. Two of the six married and childless respondents had trees. They were a 23-year-old man and woman, each recently married, and presumably planning to have children. The other four married and childless respondents were much older, and in each case, the wives were beyond childbearing age. They had no trees. Of the 13 unmarried parents in the sample (one was raising two adopted grandchildren, another had an illegitimate child, and the others were widowed or divorced), eight had trees, five did not. Of the nine unmarried and childless respondents in the sample, none had individual trees, although most had put up Christmas decorations, and one young woman had decorated a potted begonia with Christmas lights. The two unmarried childless respondents who reported trees at home were young people still living with their parents; these apparent exceptions support the rule. 1308 This content downloaded from 147.251.68.31 on Tue, 23 Dec 2014 15:50:47 PM All use subject to JSTOR Terms and Conditions clC'3 C'3 r. 0C13 U 'nacd >, cd C132X z C13 cn C.-I C13 z 0 C,3r. Cd 4, C,4 C's cli u Cld Cld Cld $. E-;, 2 Cd Cd C-d Cd Cld4. >C'd cn U- Cd cd cda OC oo 1-4 a Cd C'd U U r. r. cd cli4.j Cj --Z 0 cli C'd>cli cli E-4.- U a Cld Cld-C This content downloaded from 147.251.68.31 on Tue, 23 Dec 2014 15:50:47 PM All use subject to JSTOR Terms and Conditions American Journal of Sociology Nobody in Middletown seems to be consciously aware of the norm that requires married couples with children of any age to put up a Christmas tree, yet the obligation is so compelling that, of the 77 respondents in this categorywho were at home for Christmas 1978, only one-the Venezuelan woman previously mentioned-failed to do so. Few of the written laws that agents of the state attempt to enforce with endless paperwork and threats of violence are so well obeyed as this unwritten rule that is promulgated by no identifiable authority and backed by no evident threat. Indeed, the existence of the rule goes unnoticed. People in Middletown think that putting up a Christmas tree is an entirely voluntary act. They know that it has some connection with children, but they do not understand that married couples with children of any age are effectively required to have trees and that childless unmarried people are somehow prevented from having them. Middletown people do not consciously perceive the Christmastreeas a symbol of the complete nuclearfamily (father, mother, and one or more children). Those to whom we suggested that possibility seemed to resent it. Ethnographershave debated at some length whether the symbolic connections they detect in tribal cultures need to be verified by the testimony of participants (Foster and Brandes 1980). In this exceptionally clear instance, we infer that Middletown people sense the symbolic meaning of the Christmastree because, otherwise, the consistency of theirbehavior with respect to it would be inexplicable, but there is direct evidence that they themselves do not translate the symbol. THE WRAPPING RULE Christmas gifts must be wrapped before they are presented. A subsidiary rule requires that the wrapping be appropriate, that is, emblematic, and another subsidiary rule says that wrapped gifts are appropriatelydisplayed as a set but that unwrapped gifts should not be so displayed. Conformity with these rules is exceedingly high. An unwrapped object is so clearly excluded as a Christmas gift that Middletown people who wish to give something at that season without defining it as a Christmas gift have only to leave the object unwrapped. Difficult-to-wrap Christmas gifts, like a pony or a piano, are wrapped symbolically by adding a ribbon or bow or card and are hidden until presentation. Nowadays, in Middletown, it is not sufficient to wrap a Christmas gift in ordinary paper. Nearly all gifts are wrapped in special paper, most emblematically colored red, green, or white with graphic emblems that include Santa Claus, the Christmas tree, bells, candles, holly, mistletoe, wreaths, or carolers-a whole lexicon of familiar images. Christmas packages are decorated furtherwith ribbons, bows, and stick- 1310 This content downloaded from 147.251.68.31 on Tue, 23 Dec 2014 15:50:47 PM All use subject to JSTOR Terms and Conditions Middletown ers. Many of these packages are made up in the stores, but a greater number are wrapped at home. Women wrap far more gifts than men. Almost half the male respondents had someone else wrap their gifts, and wives were much more likely to wrap their husbands' gifts than were the husbands. Of the women in the sample, 57% wrapped all their gifts without help, compared with 16%of the men. In nearly every Middletown household, the wrapped presents are displayed under or around the Christmas tree as a glittering monument to the family's affluence and mutual affection. Picture taking at Christmas gatherings is clearly a part of the ritual; photographs were taken at 65% of the recorded gatherings. In nearly all instances, the pile of wrapped gifts was photographed; and individual participants were photographed opening a gift, ideally at the moment of "surprise."Although the pile of wrapped gifts is almost invariably photographed, a heap of unwrapped gifts is not a suitable subject for the Christmas photographer.Among the 366 gatherings we recorded, there was a single instance in which a participant, a small boy, was photographed with all his unwrapped gifts. To display unwrapped gifts as a set seems to invite the invidious comparison of gifts-and of the relationships they represent. THE DECORATION RULE Any room where Christmas gifts are distributed should be decorated by affixing Christmas emblems to the walls, the ceiling, or the furniture. This is done even in nondomestic places, like offices or restaurantdining rooms, if gifts are to be distributed there. Conformity to this rule was perfect in our sample of 366 gatherings at which gifts were distributed, although, once again, the existence of the rule was not recognized by the people who obeyed it. The same lack of recognition applies to the interesting subsidiary rule that a Christmas tree should not be put up in an undecorated place, although a decorated place need not have a tree. Unmarried, childless persons normally decorate their homes, although they have no trees, and decorations without a tree are common in public places, but a Christmas tree in an undecorated room would be unseemly. The decorations are often elaborate: We had lights outside and around the front door. There was a wreath on the front door and over the fireplace, candles around, the large and the small Christmas angel, a mistletoe ball, Christmas salt and pepper shakers, and Christmas plates for cookies, the creche on the television, card holders for the Christmas cards. Every room in the house had Christmas decorations. I have outside lights on the outdoor tree and on the garage door, electric candles in the window, 1311 This content downloaded from 147.251.68.31 on Tue, 23 Dec 2014 15:50:47 PM All use subject to JSTOR Terms and Conditions AmericanJournal of Sociology we decorate the outdoor pole lamp. On the mantel in the family room I have two latch hook stockings for my granddaughters. The mantel is decorated with angels and Rudolph. I decorate the bulletin board to look like a package. There is a wreath on the door, and the nativity scene my mother made. A Holly Hobby ball in the master bath, mistletoe in the entry way, an artificialtree and lights around the mirrorin the bathroom. My Christmas angel collection was out. It goes without saying that Christmas decorations must be temporary, installed for the season and removed afterward(with the partialexception of outdoor wreaths, which are sometimes left to wither on the door.) A room painted in red and green, or with a frieze of plaster wreaths, would not be decorated within the meaning of the rule. THE GATHERING RULE Christmasgifts should be distributedat gatheringswhere every persongives and receives gifts. Compliancewith this rule is very high. More than nine-tenths of the 1,378 gifts our respondents received, and of the 2,969 they gave, were distributed in gatherings, more than three-quartersof which were family gatherings. Most gifts mailed or shipped by friends and relatives living at a distance were double wrapped, so that the outer unceremonious wrappings could be removed and the inner packages could be placed with other gifts to be opened at a gathering. In the typical family gathering, a number of related persons assemble by prearrangementat the home of one of them where a feast is served; the adults engage in conversation; the childrenplay;someonetakes photographs;gifts are distributed,opened, and admired;and the company then disperses. The average Middletown adult fits more than three of these occasions into a 24-hour period beginning at Christmas Eve, often driving long distances and eating several large dinners during that time. THE DINNER RULE Family gatherings at which gifts are distributed include a "traditional Christmas dinner." This is a rule that participants in Middletown's Christmas ritual may disregard if they wish, but it is no less interesting because compliance is only partial. Presumably,this rule acquired its elective characterbecause the pattern of multiple gatherings described above requires many gatherings to be scheduled at odd hours when dinner either would be inappropriateor, if the dinner rule were inflexible, would requireparticipants to overeat beyond the normal expectations of the season. However, 65% of the survey respondents had eaten at least one traditional Christmas dinner the previous year. 1312 This content downloaded from 147.251.68.31 on Tue, 23 Dec 2014 15:50:47 PM All use subject to JSTOR Terms and Conditions Middletown The term "traditional Christmas dinner" was used by respondents themselves to describe a meal pattern with all or most of the following elements: (1) turkey or ham, preferably turkey and ham; (2) dressing; (3) white potatoes, preferably mashed; (4) sweet potatoes in some form; (5) cranberry sauce or salad; (6) green beans, baked beans, or bean salad; and (7) pumpkin pie and other pies. There appearsto be a subsidiaryrulethat traditionalChristmasdinners served in homes should be prepared exclusively by women. There was not a single reported instance in this survey of a traditional Christmas dinner prepared by a man. THE GIFT SELECTION RULES A Christmas gift should (a) demonstrate the giver's familiarity with the receiver's preferences; (b) surprise the receiver, either by expressing more affection-measured by the aesthetic or practical value of the gift-than the receiver might reasonably anticipate or more knowledge than the giver might reasonably be expected to have; (c) be scaled in economic value to the emotional value of the relationship. The economic values of any giver's gifts are supposed to be sufficiently scaled to the emotional values of relationshipsthat, when they are opened in the bright glare of the family circle, the donor will not appear to have disregarded either the legitimate inequality of some relationships by, for example, giving a more valuable gift to a nephew than to a son, or the legitimate equality of other relationships by, for example, giving conspicuously unequal gifts to two sons. Individuals participating in these rituals are not free to improvise their own scales of emotional value for relationships. The scale they are supposed to use, together with its permissible variations, is not written down anywhere but is thoroughly familiar to participants. From analysis of the gifts given and received by our survey respondents, we infer the following rules for scaling the emotional value of relationships. THE SCALING RULES (a) A spousal relationship should be more valuable than any other for both husband and wife, but the husband may set a higher value on it than the wife. (b) A parent-child relationship should be less valuable than a spousal relationship but more valuable than any other relationship. The parent may set a higher value on it than the child does. (c) The spouse of a married close relative should be valued as much as the linking relative. (d) Parents with several children should value them equally throughout their lives. (e) Children with both parents still living, and still married to each other, may value them equally or may value their mothers somewhat more than their fathers. A married couple with two pairs of living, still-married parents 1313 This content downloaded from 147.251.68.31 on Tue, 23 Dec 2014 15:50:47 PM All use subject to JSTOR Terms and Conditions American Journal of Sociology should value each pair equally. Children of any age with divorced, separated, or remarried parents may value them unequally. (f) Siblings should be valued equally in childhood but not later. Adult siblings who live close by and are part of one's active network should be equally valued, along with their respective spouses, but siblings who live farther away may be valued unequally. (g) Friends of either sex, aside from sexual partners treated as quasi-spouses, may be valued as much as siblings but should not be valued as much as spouses, parents, or children. (h) More distant relatives-like aunts orcousins-may be valued as much as siblings but should not be valued as much as spouses, parents, or children. It is a formidable task to balance these ratios every year and to come up with a set of Christmas gifts that satisfies them. Small wonder that Middletown people complain that Christmas shopping is difficult and fatiguing. But although they complain, they persist in it year after year without interruption. People who are away from home for Christmas arrange in advance to have their gifts distributed to the usual receivers and to open their own gifts ceremoniously. People confined by severe illnessdelegate othersto do shopping and wrapping. Although ourrandom sample of Middletown adults included several socially isolated persons, even the single most isolated respondent happened to have an old friend with whom he exchanged expensive gifts. Given the complexity of the rules, errors and failures in gift selection can be expected to occur, and they frequently do. Indeed, the four or five shopping days immediately after Christmas are set aside in Middletown stores for return or exchange of badly selected gifts. A number of respondents described relatives who make a point of being impossible to please, like the grandfather in Renata Adler's story: The grandfather, who pretended not to care about the holiday, every year, until the precisemoment when the doorto the study, where the piano stood, was opened and the presents were revealed, became every year, at that moment, hopeful, eager, even zealous and then dejected utterly. No one had ever found a present that actually pleased him. "Verynice,"he would say, in a tight voice, as he unwrapped one thing after another. "Verynice. Now I'll just put that away."The year his sons gave him an electric razor, he said, "Verynice. Of course I'll never use it. I'm too old to change the way I shave." When they asked him at least to try it, he said "No, I'm sorry. It's very nice. No I'll just put that away."[Adler 1978, pp. 136-37] The standard disappointing gift is an article of clothing in the wrong size. Womenare particularlyresentful of oversized items that seem to say the giver perceives them as "fat."Childrenareoften insulted by inattentive relatives who gve them toys that are too "young."The spouse's or lover's gift that is disliked by the receiver is a sign of alienation. Two of the five couples in our sample for whom such gifts were reported at Christmas 1978 had separated by the time of the interview several weeks later. The rigor of the Selection Rules is softened by several devices-joint 1314 This content downloaded from 147.251.68.31 on Tue, 23 Dec 2014 15:50:47 PM All use subject to JSTOR Terms and Conditions Middletown gifts from and to married couples, from children to parents, and from two or three siblings to another are common. Such arrangements make it difficult to determine whether the comparative value of relationships has been correctly translated into gifts, and that is the more or less conscious intention. Two families in our sample drew lots for their gifts. That practice is nearly standardat nonfamilial Christmasgatherings, like ward parties for hospitalized children, where presents are distributed without any attempt to particularize relationships. FITNESS RULES Rules about the fitness of gifts (e.g., women should not give cut flowers to men) are too numerous to specify, but one deserves passing attention. Money is an appropriate gift from senior to junior kin, but an inappropriate gift from junior to senior kin, regardless of the relative affluence of the parties. This is another rule which appears to be unknown to the people who obey it. Of 144 gifts of money given by persons in our sample to those in other generations, 94%went to junior kin, and of the 73 money gifts respondents received from persons in other generations, 93% were from seniorkin. A gift certificatemay be given to a parent or grandparent to whom an outright gift of money would be improper, but we did not record a single instance of a gift certificate having been given to a child or a grandchild, no substitution being called for. THE RECIPROCITY RULE Participants in this gift system should give (individually or jointly) at least one Christmas gift every year to their mothers, fathers, sons, daughters; to the current spouses of these persons; and to their own spouses. By the operation of this rule, participants expect to receive at least one gift in return from each of these persons excepting infants. Conformity runs about 90%for each relationship separately and for the aggregate of all such relationships. Gifts to grandparents and grandchildren seem to be equally obligatory if these live in the.same community or nearby, but not at greater distances (see Caplow 1982, table 6). Christmas gifts to siblings are not required. Only about one-third of the 274 sibling relationships reported by the sample were marked by Christmas gifts. The proportion was no higher for siblings living close than for those farther away. However, gifts to siblings do call for a return gift; this obligation is seldom scanted. Gift giving to siblings' children, and parents' siblings and their respective spouses, appears to be entirely elective; fewer than half of these are reciprocated. We have no way of knowing whether such gifts may be reciprocated at another Christmas, but there were no references to deferred reciprocation in the interviews. 1315 This content downloaded from 147.251.68.31 on Tue, 23 Dec 2014 15:50:47 PM All use subject to JSTOR Terms and Conditions American Journal of Sociology The Reciprocity Rule does not require reciprocated gifts to be of equal value. Parents expect to give more valuable and more numerous gifts to their minor children and to their adult children living at home than they receive in return. This imbalance is central to the entire ritual. The iconography of Middletown's secular Christmas emphasizes unreciprocated giving to children by the emblematic figure of Santa Claus, and the theme of unreciprocated giving provides one of the few connections between the secular and religious iconography of the festival-the Three Wise Men coming from a distant land to bring unreciprocated gifts to a child. Equivalence of value tends to be disregarded in gift giving between husbands and wives and between parents and their adult children. Husbands often give more valuable gifts to wives than they receive from them. The gifts of parents to adult children are approximately balanced in the aggregate-about the same number of substantial gifts are given in each direction-but there is no insistence on equivalence in particular cases, and when we examine such relationships one by one, we discover many unbalanced exchanges, which seem to be taken for granted. Only in the relationship between siblings and sibling couples do we findany active concern that the gifts exchanged be of approximatelyequal value, and even there it is more important to give gifts of approximately equal value to several siblings than to exchange gifts of equal value with each of them. Empirically, the gift giving between adults and children in our sample was highly unbalanced, in both quantity and value. Respondents gave 946 gifts to persons under 18 and received 145 in return;89 of these were of substantial value and six of the return gifts were. In about one-third of these relationships, no gift was returnedto the adult either by the child or in the child's name. In most of the remaining relationships, the child returned a single gift of token or modest value. There is little reciprocity in the gift giving between non-kin. A large number of the gifts in this category are addressed to persons who provide minor services; reciprocation in those cases would be bizarre. Gifts from employers to employees, from grateful patients to physicians, and from pupils to teachers do not call for reciprocation. The Christmas gifts exchanged en masse at club meetings and office parties are reciprocal to the extent that each participant gives and receives some small gift, but there is no direct exchange between giver and receiver. DISCUSSION Since the problem is to account for the uniformitiesof gift-giving behavior revealed by the data, speaking of rules begs the question to some extent. 1316 This content downloaded from 147.251.68.31 on Tue, 23 Dec 2014 15:50:47 PM All use subject to JSTOR Terms and Conditions Middletown Although we infer from the uniformitiesobserved in Middletown's Christmas gift giving that, somewhere in the culture, there must be statements to which the observed behavior is a response, the crucial point is that we cannot find those statements in any explicit form. Indeed, they are not recognized by participants in the system. In effect, the rules of the game are unfamiliar to the players, even though they can be observed to play meticulously by the rules. Instructions for Christmas gift giving are not found in administrative regulations or popular maxims or books of etiquette; they are not promulgated. Neither do they seem to be enforced by what Durkheim called "thepublic conscience"(Durkheim [1895] 1964, pp. 2-3). People who scanted their Christmas obligations would not be disapproved of by the public conscience in Middletown because Christmas gift giving is visualized there as both a private and a voluntary activity. We never heard anyone make an even indirect reference to community opinion in connection with Christmas gift giving. As far as we can tell, there are no customary forms of moral disapproval reserved for persons who neglect their Christmas duties (which are not, of course, considered to be duties). The moral drift goes the other way. Among Middletown's Protestant fundamentalists there are still vestiges of the violent Puritan objection to the celebration of Christmas as a "wanton Bacchanalian feast"(Barnett 1954, pp. 1-23), which is commonly expressed in sermons about the "degradation"and "commercialization"of the festival. Since the rules of Christmas gift giving have no explicit form or institutional backing or moral support, they seem to escape the dichotomous classification of "summary rules" and "rules of practice" proposed by TABLE 2 GIFT GIVING IN MIDDLETOWN BY RELATIONSHIP AND RESIDENTIAL DISTANCE, CHRISTMAS, 1978 RELATIONSHIPS MARKED BY GIFTS (%) RELATIONSHIP Within 50 Miles Beyond 50 Miles Fathers ............. ....... 100 (29) 85 (26) Mothers .................... 98 (41) 90 (29) Children ........... ........ 96 (191) 95 (61) Childrens' spouses ........... 92 (52) 94 (38) Grandparents ............... 96 (25) 50 (26) Grandchildren ....... ....... 90 (95) 77 (63) Siblings .................... 32 (122) 35 (152) Siblings' spouses .......... .. 24 (84) 24 (113) Siblings' children ............ 19 (254) 15 (384) Parents' siblings ....... ...... 15 (86) 10 (200) NOTE.-Numbers in parentheses are N's 1317 This content downloaded from 147.251.68.31 on Tue, 23 Dec 2014 15:50:47 PM All use subject to JSTOR Terms and Conditions American Journal of Sociology Rawls (1967)and elaboratedsubsequentlyby Giddens(1976)and Mulligan and Lederman (1977). They cannot be summary rules because they are not formulated on the basis of past decisions, and they cannot be rules of practice because they do not refer exclusively to a closed situation created by the rules themselves. It is tempting to abandon the concept of rulesaltogether,but if we do that, how arewe to explain the extraordinary uniformity of behavior demonstrated, for example, in tables 1 and 2? Substituting another word like "custom"or "norm"would merely introduce additional complications. What are we to make of unrecognized customs and amoral norms? As Collins remarked, Why do people repay a gift? Self-interest is not a sufficient explanation, as an exchange is rewarding only to the extent that individuals already know there will be reciprocity. Hence theorists have felt it necessary to fall back on such claims as "what is customary becomes obligatory"(Blau 1964), or to invokean alleged"normof reciprocity"(Gouldner1960;see alsoHeath 1976). Both formulations beg the explanatory question: in both cases, the customariness of the behavior is just what remains to be explained, and to call this customariness a "norm"is merely to describe it. [1981, p. 1006] Both functionalist and social exchange theories provide convincing accounts of this gift-giving system (or any other) taken as a whole. Functionalism, with its realist conception of society, looks to the contribution that a given system of activity makes to the maintenance of some larger and more durable system: the institution, the culture, the society. Exchange theory, implicitly nominalist, looks to identities in human nature to account for observed uniformities in social behavior. Some recent investigators, notably in ethnomethodology, have modified this view without attenuating its nominalism by asserting that exchange transactions seldom involve rationalcalculations of self-interestbut aretypically based on tacit understandings rooted in previous experience (see Cicourel 1973; Deutscher 1973; Cancian 1975). Ekeh (1974)distinguishes between "collectivisticorientations"in social exchangetheory,exemplifiedby Durkheim, Mauss, and particularlyLeviStrauss, and "individualisticorientations,"whose principal spokesmen he identifies as Homans and Blau, although very similar positions were articulatedmuch earlier by Frazer (1919) and by the forgotten American sociologistAlbert Chavannes, who was rediscoveredby Knox (1963).The collectivistic orientations emphasize systems of exchange and their contributionsto social solidarity.The individualistic orientationsproposethat the self-interest of individual participants provides sufficient explanation for particular transactions and ipso facto accounts for any similarities displayed by a plurality of transactions. But the two approaches, when applied to particular cases, are not as contradictoryas their protagonists claim, since the collectivists cannot demonstrate that individual trans- 1318 This content downloaded from 147.251.68.31 on Tue, 23 Dec 2014 15:50:47 PM All use subject to JSTOR Terms and Conditions Middletown actions do not satisfy the self-interest of participants in an exchange system, while the individualists have never, to my knowledge, attempted to show that the repetition of similar transactions does not contribute to social solidarity.Indeed, it is very easy to cross from one side of this street to the other when working with empirical material. An earlier report of this study proposed a functional explanation of Christmas gift giving in Middletown as serving to reinforce group solidarity (Caplow and Williamson 1980), and another report of the same study suggests that individuals are persuaded by self-interest to concentrate their gift giving on persons whose goodwill is wanted but cannot be taken for granted (Caplow 1982)-an individualistic account with no reference to group solidarity. But while it may be possible, if not prudent, to use opposing theories of social exchange to illuminate different facets of the same data, it must be admitted, nevertheless, that neither theory directly explains cultural uniformities in gift giving. In the collectivist perspective, any type of exchange transaction, whether it is cross-cousin marriagein New Guinea or Christmasgiving in Middletown, binds the entire community together. But what invisible hand accomplishes this result? It cannot be the community as a whole, because that is incapable of such concerted action, or any of its components, since none of those takes particular cognizance of the uniformities we are trying to explain. The individualistic account is incomplete in a different way. Rational self-interest may explain why Middletown people give their affinal relatives so many Christmas gifts, but the explanation is plausible only because of other uniformities in the relationshipsof Middletown people with their affines. If the uniformity of gift-giving behavioris attributedto these other uniformities, these will then, in turn, call for explanation, and we will not have made much progress with the problem. At some point in the sequence, we must face the fact that Christmas gift giving in Middletown is a customary pattern of ritual behavior. "The customariness of thebehavior"to quoteCollinsagain, "isjust what remainsto be explained." Recent work by ethnographershas carriedthe analysis of gift exchange in simple societies beyond the classic formulation of Marcel Mauss ([1925] 1967).Schieffelin (1980), summarizingthe researchof otherethnographers working in New Guinea (Wagner1969;Kapferer1976;Rubel and Rosman 1978;Ernst 1978)together with his own, concludes that "allthese studies contain the idea that exchange, as a system of meanings, is involved in the shaping or construction of particular cultural realities. They do this by focusing on the act of prestation as a rhetorical gesture of social communication, stressing the symbolism of the objects exchanged, and viewing transactionsas expressivestatementsor movementsin the management of meaning" (p. 503). 1319 This content downloaded from 147.251.68.31 on Tue, 23 Dec 2014 15:50:47 PM All use subject to JSTOR Terms and Conditions American Journal of Sociology Gift exchange, in effect, is a language that employs objects instead of words as its lexical elements. In this perspective, every culture (theremay be exotic exceptions, but I am unaware of them) has a language of prestation to express important interpersonal relationships on special occasions, just as it has a verbal language to create and manage meaning for other purposes. The language of prestation, like the verbal language, begins to be learned in early childhood and is used with increasing assurance as the individual matures and acquires social understanding. These "natal"languages are seldom completely forgotten, although new languages may be learned by translation and practice. The problem of accounting for the enforcement of gift-giving rules without visible means is simplified if we take them to be linguistic rules, or at least as similar to them, because linguistic rules, for the most part, are enforced among native speakers of a language without visible means and without being recognized explicitly. It may be objected that school teachers do make linguistic rules explicit and then enforcethem by reward and punishment, but that is a rather special case of learning a new language or relearning a natal language in more elegant form. The acquisition of language does not depend on schooling, and the grammaticalrulesthat aremade explicit in school are only a small fraction of the rules that native speakers obey without being aware of their existence. The process whereby grammatical rules acquire consensual support is partly instinctual, partly cultural, and partly social. The tendency to follow linguistic rules without explicit awareness appears to be innate in the construction of new verbal combinations: young children acquire the language of the people who raise them along with other elements of the ambient culture; and linguistic rules are self-enforcing insofar as the effective transmission of messages rewards both senders and receivers. Visualizing Christmas gift giving as a language-or, more precisely, as a dialect or code (Douglas 1972, 1979)-helps to explain, among other matters, the insistence on wrapping and other signs to identify the objects designated for lexical use and the preference for the simultaneous exchange of gifts at family gatherings rather than in private. In most cases such a gathering is composed of a parent-child unit containing one or two parents and one or more children together with other persons who are tied to that unit by shared membership in another parent-child unit, such as children'schildren, children'sspouses, parents' siblings, or parents'parents. Although there is room at a family gathering for a friend or distant relative who otherwise might be solitary at Christmas, there is no convenient way of including any large number of persons to whom no gift messages are owed. Under the Scaling Rules, gift messages are due from every person in a parent-child relationship to every other. The individual message says "I value you according to the degree of our relationship"and anticipates 1320 This content downloaded from 147.251.68.31 on Tue, 23 Dec 2014 15:50:47 PM All use subject to JSTOR Terms and Conditions Middletown the response "I value you in the same way."But the compound message that emerges from the unwrapping of gifts in the presence of the whole gathering allows more subtle meanings to be conveyed. It permits the husband to say to the wife "I value you more than my parents" or the mother to say to the daughter-in-law "I value you as much as my son so long as you are married to him" or the brother to say to the brother "I value you more than our absent brothers, but less than our parents and much less than my children." These statements, taken together, would define and sustain a social structure, if only because, by their gift messages, both parties to each dyadic relationship confirmthat they have the same understanding of the relationship and the bystanders, who are interested parties, endorse that understanding by tacit approval. The compound messages would have a powerful influence even if they were idiosyncratic and each parent-child unit had its own method of scaling relationships. In fact, there are some observable differences in scaling from one Middletown family to another and from one subcultural group to another, but the similarities are much more striking than the differences. We attribute this commonality to the shared dialect of Christmas gift giving, hyperdeveloped in Middletown and elsewhere in the United States in response to commercial promotion, stresses in the family institution, and constant reiteration by the mass media. Once the dialect is reasonably well known, these factors continue to enlarge its vocabulary and its domain. Another circumstance facilitating the standardization of the dialect is that nearly every individual in this population belongs to more than one parent-child unit for Christmas gift-giving purposes. Because these units are linked and cross-linked to other units in a network that ultimately includes the larger part of the community, they would probably tend to develop a common set of understandings about appropriate kinship behavior, even without the reinforcement provided by domestic rituals. The most powerful reinforcement remains to be mentioned. In the dialect of Christmas gift giving, the absence of a gift is also a lexical sign, signifying either the absence of a close relationship, as in the Christmas contact of cousins, or the desire to terminate a close relationship, as when a husband gives no gift to his wife. People who have once learned the dialect cannot choose to forget it, nor can they pretend to ignore messages they understand. Thus, without any complicated normative machinery, Middletown people find themselves compelled to give Christmas gifts to their close relatives, lest they inadvertently send them messages of hostility. In this community, where most people depend on their relatives for emotional and social support, the consequences of accidentally sending them a hostile message are too serious to contemplate, and few are willing to run the risk. In sum, we discover that the participants in this gift-giving system are 1321 This content downloaded from 147.251.68.31 on Tue, 23 Dec 2014 15:50:47 PM All use subject to JSTOR Terms and Conditions American Journal of Sociology themselves the agents who enforce its complex rules, although they do so unknowingly and without conscious referenceto a system. The dialect, once learned, imposes itself by linguistic necessity, and the enforcement of its rules is the more effective for being unplanned. REFERENCES Adler, Renata. 1978. Speedboat. New York: Popular Library. Barnett, James H. 1954. The American Christmas: A Study of National Culture. New York: Macmillan. Blau, Peter M. 1964. Exchange and Power in Social Life. New York: Wiley. Cancian, Francesca. 1975. What Are Norms? London: Cambridge University Press. Caplow, Theodore. 1982. "Christmas Gifts and Kin Networks." American Sociological Review 47:383-92. Caplow, Theodore, Howard H. Bahr, Bruce A. Chadwick, Reuben Hill, and Margaret Holmes Williamson. 1982. Middletown Families: Fifty Years of Change and Continuity. Minneapolis: University of Minnesota Press. Caplow, Theodore, and Margaret Holmes Williamson. 1980. "Decoding Middletown's Easter Bunny: A Study in American Iconography." Semiotica 32:408-29. Cicourel, Aaron V. 1973. Cognitive Sociology. Baltimore: Penguin. Collins, Randall. 1981. "On the Microfoundations of Macro-Sociology." American Journal of Sociology 86:984-1014. Deutscher, Irwin. 1973. What We Say, What WeDo: Sentiments and Acts. Glenview, Ill.: Scott Forseman. Douglas, Mary. 1972. "Deciphering a Meal." Daedalus 101:31-81. . 1979. The World of Goods. New York: Basic. Durkheim, Emile. (1895) 1964. Rules of Sociological Method. New York: Free Press. Ekeh, Peter P. 1974. Social Exchange Theory: The Two Traditions. Cambridge, Mass.: Harvard University Press. Ernst, Peter T. 1978. "Aspects of Meaning of Exchanges and Exchange Items among the Onabasalu of the Great Papuan Plateau." Pp. 161-78, in Trade and Exchange in Oceania and Australia, edited by J. Specht and J. P White. Sydney: Mankind. Foster, Mary, and Stanley H. Brandes, eds. 1980. Symbol as Sense: New Approaches to the Analysis of Meaning. New York: Academic Press. Frazer, Sir James G. 1919. Folklore in the Old Testament, vol. 2. London. Macmillan. Giddens, Anthony. 1976. New Rules of Sociological Method. London: Hutchinson. Gouldner, Alvin W. 1960. "The Norm of Reciprocity: A Preliminary Study." American Sociological Review 25:161-78. Heath, Anthony. 1976. Rational Choice and Social Exchange: A Critique of Exchange Theory. New York: Cambridge University Press. Kapferer, Bruce, ed. 1976. Transaction and Meaning. Philadelphia: Institute for the Study of Human Issues. Knox, John B. 1963. "The Concept of Exchange in Sociological Theory." Social Forces 41:341-46. Lynd, Robert, and Helen Merell Lynd. (1929) 1959. Middletown: A Study in American Culture. New York: Harcourt Brace. . (1937) 1963. Middletown in Transition: A Study in Cultural Conflicts. New York: Harcourt Brace. Mauss, Marcel. (1925) 1967. Essay on the Gift: An Archaic Form of Exchange. Translated by I. Cunnison. London: Cohen & West. Mulligan, Glenn, and Bobby Lederman. 1977. "Social Facts and Rules of Practice." American Journal of Sociology 83:539-50. 1322 This content downloaded from 147.251.68.31 on Tue, 23 Dec 2014 15:50:47 PM All use subject to JSTOR Terms and Conditions Middletown Rawls, John. 1967. "Two Concepts of Rules." In Theories of Ethics, edited by Philippa Foote. Oxford: Oxford University Press. Rubel, Paula, and Abraham Rosman. 1978. YourOwn Pigs YouMay Not Eat. Chicago: University of Chicago Press. Schieffelin, Edward L. 1980. "Reciprocity and the Construction of Reality." Man: Journal of the Royal Anthropological Institute 16:502-17. Wagner, Roy. 1969. The Curse of Souw. Chicago: University of Chicago Press. 1323 This content downloaded from 147.251.68.31 on Tue, 23 Dec 2014 15:50:47 PM All use subject to JSTOR Terms and Conditions