Off the Rack:
Store Bought Emotions and the Presentation of Self

Sarah C Brabant
University of Southwestern Louisiana

The authors would like to thank Jon Beckert for his helpful assistance. Direct all correspondence to Linda A. Mooney

The impact of relational status (husband-wife, parent-child, etc.) and age diversity (same cohort, younger, older) were measured on two of Goffman’s interactional rituals — deference and demeanor. While deference refers to an Actor’s definition o f Other as conveyed through an expression of sentiment, demeanor refers to an Actor’s presentation of self through “deportment, dress, and bearing” (1967: 58,77). In order to measure deference and demeanor, 535 birthday cards were randomly selected from 1 4 retail outlets and a subset of 244 cards in which relational status was known was identified and analyzed. Deference was measured by card content i.e., sentiment conveyed to receiver, and demeanor was measured by such “decorations” as card make, cost, a nd design. Results indicate that while deference is not consistently related to relational status or age diversity, positive demeanor is directly related to intimacy of sender-receiver relationship.

Despite only recent attention to the sociology of emotions, much of Simmel’s and Durkheim’s writings solidly place the study of emotional life within the sociological arena. Although seemingly an unlikely area for scientific inquiry, and in particular for macro- sociological theorizing, many early examinations of emotions concentrated on their role in the emergence and development of modern societies (Gerhards 1986: 903). However, more recent sociological studies are often micro- sociolog ical in nature focusing on the ways in which emotions are socially constructed, “interpreted, propagated, and deployed’ (Jackson 1993: 209).

The social construction of emotions, whether in reference to etiology, content or consequence, is methodologically difficult to investigate. Jackson, in studying what she calls “emotional literacy” states that love, for example, can not be explored (19 93: 204):

except through the ways in which it is talked about and written about. Language itself, moreover, contributes to the cultural construction of emotions and is a means by which we participate in creating a shared sense of what emotions are.

Investigating emotions thus requires an examination of the linguistic tools through which they are conveyed. One such innovative method is an examination of greeting cards. While greeting cards have been used to investigate a number of emotions (cf. Cl ark’s (1989) “store bought sympathy” ), the present study examines not only the sentiments conveyed, but the nature of the relationship between card sender and receiver.

Greeting Cards

The first known greeting card was sent over 150 years ago when Henry Cole, a British civil servant, had copies of a hand-painted Christmas scene lithographed and sent to co-workers, friends, and relatives (Sinha 1995). Since that time, greeting cards h ave become a multi-billion dollar business with Hallmark and American Greetings recording 1995 sales of over three and two billion dollars respectively (AGC 1996; “The Hallmark Story” 1996).

Americans send more greeting cards than any other people in the world (Sinha 1995). In a recent industry-sponsored study of over 1,000 adults, 90 percent of the participants stated that they respond immediately to receiving an unexpected card either by telephoning the sender or by sending a card or letter (Audits and Surveys, Inc. 1996). Further, 92 percent of the respondents reported sending at least one card in the previous year, with an average of 25 cards a year being sent and 27 cards a year being received — excluding Christmas cards.

What is it about receiving a greeting card that makes it a keepsake, elicits an immediate response, or brings people to tears? The answer lies in what greeting cards most often communicate — sentiments. Sentiments are socially significant feelings — each sentiment signifying a pattern of sensations, emotions, actions, and cultural beliefs appropriate to a social relationship (Michner, DeLamater and Schwartz 1991; Gordon 1989).

Greeting cards then are one element of “emotional culture” or the “patterns of meanings embodied in symbols, by which people communicate, perpetuate, and develop their knowledge about and attitudes toward emotions” (Gordon 1989: 115). Emotional culture is conveyed through “language, rituals, art forms, and other publicly available, symbolic vehicles of meaning” (Gordon 1989: 115). Greeting cards, as symbolic vehicles of meaning or what Goffman (1967) calls “ceremonial tokens”, are an unobtrusive yet re adily available measure of emotional culture. The present research examines greeting cards and, specifically, what Clark (1989) calls “off-the-rack sentiments” within the framework of Goffman’s theoretical analysis of deference and demeanor.

Deference And Demeanor

Card giving occasions, like gift giving occasions, are governed by “ceremonial rules,” “rules of etiquette” (Goffman 1967) or “rules of civil propriety” (Denzin 1970). Ceremonial rules have their primary importance “as a means of communication by which the individual expresses his character or conveys his appreciation of the other participants…” (Goffman 1967: 74).

As in all ceremonial activity, conformity to or violation of ceremonial practices convey information about the actor through the two major components of ceremonial idiom, deference and demeanor. Deferential behavior entails actions in which a participa nt “celebrates and confirms his relationship to a recipient” by symbolically conveying “a sentiment of regard” (e.g. appreciation, respect, compliments, etc.) for Other (Goffman 1967: 55-62).

Alternatively, demeanor connotes a “presentation of self” to Other, by an Actor performing or failing to perform in accordance with established ceremonial rules, thereby conveying the image of an individual who has “certain desirable or undesirable qua lities” i.e., is either properly or improperly demeaned (Goffman 1997: 77). It is, thus, through demeanor that an individual “creates an image of self” (Goffman 1967: 77-78).

Deference and demeanor, while theoretically distinct, are often analytically inseparable for the granting or withholding of proper deference may infer good or bad demeanor (Goffman 1967: 81).

It may be illustrated from recent material on doctor-patient relationships, where it is suggested that one complaint a doctor may have against some of his patients is that they do not bathe before coming for an examination; while bathing is a way of paying deference to the doctor it is at the same time a way for the patient to present himself as a clean, well demeaned person (Goffman 1967: 81-82).

Similarly, proper or improper demeanor may not only convey an appreciation of Other, it may demonstrate the Actor’s value as a worthwhile role occupant.

Gifts, for example, are ceremonial tokens that convey information about the giver and the receiver in terms of deference and demeanor. In his classic essay on the social psychology of gifts, Schwartz (1967: 2) argues that gifts “impose an identity upon the giver as well as the receiver” for gifts “are most frequently given which are consonant with the character of the recipient” and are, as well, “self-defining” of the giver. Similarly, as noted by Schwartz (1967), Emerson (1936: 358) writes, “a man’s biography is conveyed in a gift” and Levi-Strauss notes (1965: 77) that:

…the refinement of selection [of Christmas cards], their outstanding designs, their price, the quantity sent or received, give evidence (ritually exhibited on the mantelpiece during the week of celebration) of the recipients social bonds and the degree of his prestige.

Thus, the most seemingly trivial ceremonial acts, giving a gift or sending a greeting card, carry with them information about sender and receiver for cards, as gift giving in general, “not only confirm social bonds but also identities” (Komter 1996: 31 0). The present research analyzes deference and demeanor through the examination of one ceremonial token, birthday cards. While card content is used as an indicator of “a sentiment of regard” for Other, card characteristics (what Goffman [1959; 1979] call s “decorations”) such as style and design as suggested by Levi-Strauss (1965) are used as indicators of a sender’s “presentation of self.” Specifically, the present research examines the two major components of ceremonial activity, deference and demeanor, as a function of relational status and age diversity of card sender and receiver.

Methodology

Greeting cards transmit “meaning” between social actors — meanings which are evaluated by the card sending population in selecting the appropriateness of a card. Cards are defined as appropriate not only because they contain suitable “decorations ” given relational status and age considerations, but because they convey the desired emotion. While it could be argued that greeting cards more accurately reflect the writers’ and artists’ conceptions of appropriateness than the general publics’, the ext ensive consumer testing procedures used by the greeting card industry assure that the cards appearing in retail displays “meet the approval of a significant segment of the card buying and receiving public” (Finn 1980) “reflecting changes in the way people communicate” (The Hallmark Story 1996). Nonetheless, it should be noted that the present research examines deference and demeanor as ostensibly conveyed between send and receiver as measured by one type of ceremonial token — greeting cards.

While some greeting card sending and receiving populations are limited by age (e.g. wedding), sex (e.g. Bar mitzvah), ethnicity (St. Patrick’s Day) and/or relational status (e.g. Mother’s Day), birthday card senders and receivers constitute a culturall y heterogeneous population, 90% of Americans aged 16-69 reporting that they received a card on their last birthday (Brown 1995). Because of their universality, the present analysis was confined to birthday cards.

In order to secure a representative sample of birthday cards, fourteen retail outlets in a mid-size city in southwestern Louisiana were selected. The stores selected represented the variety of retail outlets available to the card-buying population incl uding large discount stores such as K-Mart, as well as drug stores/pharmacies, dime stores, department (e.g. Sears, Penny’s) and grocery stores (e.g. Albertson’s), and small specialty “card shops”. Stores included for sampling purposes represented geograp hically distinct, high traffic areas as well as diverse socioeconomic areas of the community.

Prior to entering a store, a number between one and ten was randomly selected. Upon entering a store, the card display was located and, beginning at the upper-most left corner of the birthday card rack, cards were counted to the predetermined number un til the “start card” was reached. That card was selected and became card number 001. From that card forward, every tenth card was selected for inclusion in the sample. Thus, store sample size varied in proportion to the inventory available to customers an d the total sample represents 10.0% of the cards available in the targeted stores.

A total of 535 cards were collected, assigned identification numbers and coded twice, first by each of the researchers independently and then, for reliability, collectively. When discrepancies in coding existed, a third judge was used to resolve the di fferences. Given the unmistakable nature of many of the variables (e.g. cost of card, manufacturer of card, presence or absence of a word) however, few disagreements emerged.

Cards were then coded as to the intended receiver (e.g. mother, father, aunt, uncle), if known. Of the 535 cards collected, a subsample was identified in which intended receiver, and thus the relationship between sender and receiver, was known (N=244). Since it is more profitable for card manufacturers to produce cards that are not limited by “status identifiers” (Goffman 1967) such as “To a Fine Son” or “To my Nephew on his Birthday”, it is not surprising that over half of the cards were eliminated fr om the present analysis.

Table 1 displays the sample and subsample of cards by type of store. Percentage differences between the two are minimal suggesting that the integrity of the initial sampling procedure was preserved. Cards in the subsample were then classified on the ba sis of the two independent variables and analyzed.

Table One: Distribution of Cards Sampled by the Type of Retail Outlet

 SAMPLE SUBSAMPLE 
 (N=535) (N=244) 
Type of StoreN%N%
Specialty Shops15629.271.029.6
Department Stores15529.066.027.5
Discount Stores79.014.834.014.2
Dime Stores65.012.129.010.4
Grocery Store49.09.227.011.2
Drug Store/Pharmacy31.05.817.07.1

Measurement of the Independent Variables

The first independent variable is relational status i.e., the specific kin relationship between card sender and card receiver. Relational status has been shown to impact gift giving behavior in a variety of ways and settings (cf. Mauss 1989 [1923]); Ma linowski 1922; Sahlins 1972; Komter and Vollebergh 1997; Komter 1996). The six relational status categories include husband-wife (N=25), parent-child (N=113), siblings (N=43), grandparent-grandchild (N=26), uncle/aunt-niece/nephew (N=22), and in-laws (N=1 5).

The second independent variable is age diversity. Goffman theorizes that the rules of deference and demeanor vary between social equals, i.e., those in a symmetrical relationship, and those in an asymmetrical relationship (1967: 78). Cards were thus co ded into three categories: 1) cards likely, on the basis of age, to be between “social equals”, that is, members of the same age cohorts (to husband, wife, sibling, brother-in-law, or sister-in-law, N=74) versus cards intended for receivers likely to be 2 ) younger than the card sender (to son, daughter, granddaughter, grandson, niece, nephew, daughter- in-law, or son-in-law, N=73) or 3) older than the card sender (to mother, father, grandmother, grandfather, aunt, uncle, mother-in-law, or father-in-law, N =97).

Measurement of the Dependent Variables

Goffman’s theoretical constructs of deference and demeanor were operationalized and measured through the analytical technique of content analysis. Deference, or “a sentiment of regard” for others, was measured by card content. Cards were coded in terms of the presence or absence of the expression of: 1) love (e.g “A mother’s love” “To the Man I Love”), 2) praise (e.g. “One in a Million”, “Perfect in Every Way” “To a Good Boy”), 3) gratitude (e.g. “I’m so thankful you’re in our family!”, “You’re always there in a time of need”), and 4) caring (e.g. “You are so dear to me”, “To a Special Aunt”). Cards were also coded as to 5) whether or not the message contained a reference to the sender or receiver being in the others’ thoughts (“I think of you often”, “Thinking of you Grandma on your Special Day”). The sum of the five deference indicators (PRAISE + LOVE + GRATITUDE + CARING + THOUGHT) comprise a summary measure of the expression of positive sentiment. Ranging from a minimum of five to a maximum of ten, the sentiment index has a mean of 7.42, and a mode and median of 7.00. The higher the score on the deference index, the more positive the sentiment conveyed from sender to receiver i.e., the greater the “sentiment of regard”.

The second dependent variable, demeanor, represents the presentation of self that is conveyed by a card to the receiver. It should be remembered that it is “through demeanor the individual creates an image of himself” to others (Goffman 1967: 78) and t he selection and sending of a greeting card is one way in which that image is constructed. As Furnham (1991: 60) writes:

What do cards tell us about the senders? Consider the hidden messages: those not contained in the words in the cards (though these are interesting in their own right), but ones that reflect the values, personality and self-concept of the sende r.

Indeed, capitalizing on the knowledge that cards convey information about the sender, Hallmark’s motto “when you care enough to send the very best” and, more recently, “…go ahead and look” reinforces the idea that if the sender really cared, they’d s end a Hallmark. Hallmark’s 40 percent share of the market and research, confirms the success of the campaign (Woodyard 1997). For example, Andrus, Silver and Johnson (1997) report that buying “status brand” gifts increases the symbolic meaning of the pres ent as well as the positive regard in which the sender is held.

Thus, cards were categorized by card decorations including: 1) whether or not the card was from one of the two major greeting card manufacturer (off-brands v. Hallmark and American Greeting), 2) whether the card was below or above the average card cost , 3) finish (dull v. shiny, metallic, glitter), 4) shape (regular v. irregular, over-sized, cut-a-ways, etc. ), 5) surface (flat/smooth v. raise/ embossed pictures and/or letters) and 6) extra (the absence or presence of some “bonus” [e.g. an enclosure, i nsert sheets, four or more sides with writing, etc.]). The sum of these six indicators (MAKE + COST + FINISH + SHAPE + SURFACE + EXTRA) comprises the decoration index ranging from a low of six to a high of twelve, with a mean, median and mode of 9.00. The higher the indicator or index mean, the more positive, theoretically, the demeanor conveyed. Following Wright (1986: 228-229), differences of ten percent or more were considered meaningful.

Results

Of the five deference indicators, the expression of caring, praise, and love are the most frequently occurring with 83%, 74%, and 52% of the cards respectively expressing these sentiments. Gratitude (21%), as well as remarks indicating that the recipie nt, the sender, or both were “in one’s thoughts”, is less likely with only thirty of the 244 cards (12%) containing this message.

When holding relational status constant, some interesting results are revealed (see Table 2). Eighty percent of cards between husbands and wives contain the expression of love while only 65% of cards between grandparents and grandchildren, and 62% of c ards between parents and children, express love between role occupants.

Table Two: Mean Score for Each of the Difference Indicators by Relational Status and Age Diversity (N=24)

 PraiseLoveGrat.CaringThoughtIndex
Relational Status      
Husband-Wife (N=25)1.561.801.281.881.047.56
Parent-Child (N=113)1.741.621.251.801.107.51
Siblings (N=43)1.701.261.121.791.167.02
Grandparent-      
Grandchild (N=26)1.771.651.351.961.157.89
Uncle/Aunt-      
Niece/Nephew (N=22)1.771.271.051.861.187.14
In-laws (N=15)2.001.201.131.801.137.27
Age Diversity      
Same Cohort (N=74)1.681.431.181.841.127.34
To Younger (N=73)1.801.451.121.841.147.24
To Older (N=97)1.741.641.311.811.117.67
Grand Means1.741.521.211.831.127.42

Grandparent-grandchild cards contain the highest proportion of references to caring (96%) and gratitude (35%), followed by husband-wife cards in which the expression of caring (88%) far exceeded the expression of gratitude (28%). Being in each other’s thoughts, however, is most likely to be expressed between uncles/aunts and nieces/nephews (18%), while 100% of in-law cards contain some reference to praise.

An examination of the mean sentiment index scores indicate that the grandparent-grandchild relationship is characterized by the highest frequency of positive sentiments. The mean sentiment score for this relationship is 7.89 followed by that between hu sband-wife (7.56), parent-child (7.51), in-laws (7.27), uncles-aunts-nieces/ nephews (7.14), and siblings (7.02).

“In thoughts” and the expression of caring reveal few differences in the symmetry of the relationship with means for the three age diversity categories differing little. However, while gratitude is most likely to be expressed upward i.e., to older card recipients (31%) when compared to those of the same age (18%) or younger (12%), praise is more likely to be expressed to younger card recipients (80%).

Love, like gratitude, is most frequently expressed to older recipients. Of all cards to older recipients, 64% contain a reference to love compared to 45% of all cards to younger recipients and 43% of card receivers in the same age cohort. Given these r esults, it is not surprising that the sentiment index scores are highest among younger sender-older recipients (7.67), followed by members of the same cohort (7.34), and the older sender-younger recipient category (7.24).

Demeanor

Table 3 displays the results of the demeanor analysis as indicated by the six measures of card decorations.

Table Three: Mean Score for Each of the Demeanor Indicators by Relational Status and Age Diversity (N= 244)

 Decoration Indicators
 MakeCostFinishShapeSurfaceExtraIndex
Relational Status       
Husband-Wife (N=25)1.881.441.841.001.801.449.40
Parent-Child (N=113)1.891.321.831.071.681.469.26
Siblings (N=43)1.861.301.721.091.721.429.12
Grandparent-       
Grandchild (N=26)1.841.061.811.151.581.128.54
Uncle/Aunt-       
Niece/Nephew (N=22)1.591.051.641.181.451.278.18
In-laws (N=15)2.001.201.871.001.801.409.27
Age Diversity       
Same Cohort (N=74)1.881.341.781.051.761.429.23
To Younger (N=73)1.831.151.741.161.611.358.83
To Older (N=97)1.871.301.851.051.671.419.44
Grand Means1.861.271.681.081.801.399.08

Unlike the sentiment analysis, there appears to be a fairly consistent link between card decoration and relational status. For three of the variables, make, cost and surface, as relational distance increases from husband-wife to uncle/aunt-niece/nephew (Sahlins 1972; Liao and Stevens 1994), each of the dependent indicators decreases. Thus, for example, 44% of cards between husbands and wives are above average in cost while only 5% of cards between uncle/aunt-niece/nephew are above average in cost.

Not surprisingly, the decoration index scores reflect this trend. In general, the greater the likely intimacy between sender and receiver, the higher the card decoration score. Thus, cards between husbands and wives were more ornate, with a mean index score of 9.40, than cards between parents and children (mean=9.26), siblings (mean=9.12), grandparents and grandchildren (mean=8.54), and uncles/aunts-nieces/nephews (mean=8.18).

Cards between in-laws, however, had a higher card decoration index score (mean=9.27) than all other role relationships except husband-wife and thus serves as the only exception to the general trend. No such trend exists for the age diversity or symmetr y analysis. In general, cards to older recipients had the highest number of displays as indicated by a decoration index mean score of 9.44. However, cards to younger receivers where the most likely to be irregular in shape or size, and cards within the sa me age cohort were the most likely to be name brand (88%). Further, percentage differences between age categories are minimal suggesting that, as with deference, the present investigation was unable to uncover any consistent way in which the symmetry (or lack thereof) of a relationship is linked to demeanor.

Discussion

The results of the present research indicate that “a sentiment of regard”, as one measure of an Actor’s deference toward Other, varies between relational status and age diversity although with few consistent and clear patterns. Positive sentiments are most often expressed within the context of three role relationships, in rank order: 1) grandparent-grandchild, 2) husband-wife, and 3) parent-child. Further, the age diversity analysis suggest that, although there are indicator differences, the direction of sentiment expressed is most often from younger sender to older receiver i.e., from grandchild to grandparent and from child to parent. This finding supports what Goffman calls a common sense understanding of deference — it is “something a subordi nate owes to…[a] superordinate” (1967: 59) i.e., it occurs more often in asymmetrical relationships.

Alternatively, demeanor, or the presentation of self to Other as measured by card decoration, is generally linked to the intimacy of the Actor-Other relationship: the closer the role occupants, the more favorable the demeanor or presentation of self. T his finding is consistent with gift giving research results which report that gift exchange occurs more frequently between friends or family members, and that the nature of the gift differs as proximity increases (Komter 1996; Komter and Vollebergh 1997).

How are we to interpret the rather ambiguous findings? It is important to remember that card sending, in general, is part of what Di Leonardo (1987, p. 443) calls “kin work”, a gender based phenomena in which women bear the responsibility for:

…the conception, maintenance, and ritual celebration of cross-household ties, including visits, letters, telephone calls, presents, and cards to kin; the organization of holiday gatherings; the creation and maintenance of quasi-kin relations ; [and] decisions to neglect or to intensify particular ties…it is kinship ties across households, as much as women’s work within them, that fulfill our cultural expectations of satisfying family life.

De Leonardo’s argument is strengthened by research results which indicate that women express positive sentiments more often than men (cf. Cancian 1985), are disproportionately represented in the card purchasing and sending population (Audits and Survey s, Inc. 1996; Brown 1995) and, therefore, are more often used in consumer testing of cards (Rhodes 1971; “The Hallmark Story” 1996). Further, considerable evidence exists that women buy cards for others to send — most notably for husband and childre n (Greene and Polivka 1985; Kansas City Star 1996), and are more likely to send birthday cards than men, 17 cards a year on the average (Brown 1995).

Hochschild’s analysis of “emotion work” is also relevant for she argues that women not only do more emotion work but do “extra emotion work — especially work that affirms, enhances, and celebrates the well-being and status of others” (Hoch schild 1983: 165). Consistent with De Leonardo’s (1987) and Hochschild’s (1983) theoretical insights, gift giving, in general, has been found to be “women’s work” from selection and purchase, to wrapping and presentation (Cheal 1988; Caplow 1982; 1984; Mi ller 1993; Komter 1996; Komter and Volleberge 1997).

The sentiment recorded in the present investigation as one measure of deference is likely to be a woman’s definition of the appropriateness of expressing certain sentiments to various role occupants or, minimally, her definition of what she beli eves others believe to be the ideal. The card quality or decoration she selects also varies directly with the demeanor she believes should be conveyed of herself or of the sender for which she is buying the card. The closer the relationship, the more positive the image conveyed — this act of ceremonial idiom conveying a presentation of self as a valued role occupant — a good son, a fine daughter, or a loving husband. At the center of ceremony, at least in terms of the communicati on of deference and demeanor through birthday cards, is a women — her choices limited by card availability and card availability determined by the choices she makes in consumer testing trials.

The greeting card industry, however, is changing. Hallmark and American Greeting, now in the business of communication, have opened “social expression stores” on the Internet, video and audio “do-it-yourself” greeting cards are available on manufacture rs’ homepages, “build-a-card” computer machines frequent discount stores, and Microsoft and Hallmark offer “personal expression” software. These changes may alter the expression of deference and demeanor between sender and receiver as sender, rather than manufacturer or marketing research, determines card sentiment. Additionally, the differential access to and availability of “cybercards” may impact the demographics of the card sending and receiving populations.

In addition to new card technologies, there has also been a significant increase in the sale of non-occasion or “anyday cards” with unit sales increasing at an estimated rate of 10% a year compared to a growth rate of 3% per year for all cards (Wandycz 1991; “The Hallmark Story” 1996). Traditionally, cards are sent to mark special occasions many of which are socially constructed as store displays and industry commercials dictate appropriate card-sending events — Father’s Day, Sweetest Day, Secret ary’s Day, Nurses’ Day, and the like.

In part, the increased popularity of “anyday cards” speaks to the industries success at socially constructing non-traditional card sending occasions — the need for an apology — the expression of anger — even the inability to express s entiments becomes worthy of a ceremonial token or “sign vehicle”. Congratulations cards aren’t just for wedding announcements, babies, and promotions. One Hallmark card in their “Just How I Feel” line states, “I’m really proud of you! Not only did you hav e the courage to admit you needed help, but you’ve worked hard to put your life back in order and keep it that way.”

Finally, the increase in non-occasion cards may also reflect that today “… people prefer cards that are honest and immediate in their sentiments. This is a major shift from the formal and oblique messages in cards of earlier years” (Heubusch 1997: 32 ). Anyday cards thus provide the researcher with an unobtrusive and welcomed measure of emotional culture as senders present themselves, as well as their regard for Other, in a variety of newly defined card sending events as evidenced by Hallmarks “Just H ow I Feel” line. It is through these cards that senders express a wider range of socially significant emotions — including the elements of deference and demeanor.

References

Andrus, David M., Edward Silver and E. Dallas Johnson. 1996.”Status Brand Management and Gift Purchase: A Discriminant Analysis.” Journal of Consumer Marketing 3 (1): 5-9.

American Greetings Corporation (AGC). 1996. “Corporate Profile.” http://www.amgreetings.com.

Audits and Surveys, Inc. 1996. “What a Card.” Psychology Today. January. p. 18.

Brown, Camala. 1995. “Cakes, Cards and Candles. American Demographics(March).

Cancion, F.M. 1985. “Gender Politics: Love and Power in the Public and Public and Private Sphere. Pp. 253-264 in Gender and the Life Course Alice Rossi (ed). New York: Aldine.

Clark, Candace. 1989. “Studying Sympathy: Methodological Confessions”. Pp. 137-152 in Sociology of Emotions: Original Essays and Research Papers. David Frans and E. Doyle McCarthy (eds.). Greenwich, Conn: JAI Press.

Company News, 1996. “Demand Grows for Ethnic Valentines.” Press Release. http://www.prnewswire.com.

Caplow, Theodore. 1982. Christmas Gifts and Kin Networks. American Sociological Review 47: 383-92.

Caplow, Theodore. 1984. Middletown. American Journal of Sociology 89: 1306-23.

Cheal, D. 1988. The Gift Economy. London: Routledge.

Denzin, Norman. (1970). Rules of Conduct and the Study of Deviant Behavior: Some Notes on Social Relationships. Pp. 120-159 in J. Douglas (ed.) Deviance and Respectability: The Social Construction of Moral Meaning. New York: Basic Books.

Di Leonardo, M. 1987.”The Female World of Cards and Holidays: Women, Families, and the Work of Kinship.” Signs 12: 440-453.

Emerson, Ralph Waldo. 1936. “Gifts” in Emerson’s Essays. Philadelphia: Spencer Press.

Finn, P. 1980. “Attitudes toward Drinking Conveyed in Studio Greeting Cards.” American Journal of Public Health 70: 826-829.

Furnham, Adrian. 1991. “Send as You Would be Sent To.” New Scientist December (21-28): 60-61.

Gerhards, Jurgen. 1986. “Georg Simmel’s Contribution to a Theory of Emotions. ” Social Science Information 25 (4): 901-924.

Goffman, Erving. 1967 Interactional Ritual. Chicago: Aldine.

Goffman, Erving. 1979. Gender Advertisements. New York: Harper and Row.

Goffman, Erving. 1959. The Presentation of Self in Everyday Life. Garden City, N.Y.: Doubleday.

Gordon, Steven L. 1989. “Institutional and Compulsive Orientations in Selectively Appropriating Emotions to Self.” Pp. 115-135 in Sociology of Emotions: Original Essays and Research Papers. David Frans and E. Doyle McCarthy (eds.). Greenwich, Conn: JAI Press.

Greene, R. and J. Polivka. 1985. “The Meaning of Grandparents’ Day Cards: An Analysis of the Intergenerational Network.” Family Relations 34: 221-225.

Hallmark, Story The. 1996. “Hallmark History is the Story of Birth of an Industry.” http://www.hallmark.com/ourcompany_bin/corporate/hisbirth.asp.

Heubusch, Kevin. 1997. “Cards for Life’s Bummers.” American Demographics 19 (10): 32.

Hochschild, Arlie R. 1983. The Managed Heart. Berkeley: University of California Press.

Jackson, Stevi. 1993. “Even Sociologists Fall in Love: An Exploration of the Sociology of Emotions.” Sociology 27(2): 201-217.

Kansas City Star, 1996. “Hallmark, Microsoft Team up to Create “Social Expression Products”. Nando.net.

Komter, Aafke Elisabeth. 1996. “Reciprocity as a Principle of Exclusion: Gift-Giving in the Netherlands.” Sociology 30: 299-319.

Komter, Aafke Elisabeth and Wilma Vollebergh. 1997. “Gift Giving and the Emotional Significance to Family and Friends.” Journal of Marriage and Family 59 (3): 747-311.

Levi-Strauss, Claude. 1965. “The Principle of Reciprocity” in Lewis Coser and Bernard Rosenberg (eds.) Sociological Theory. New York: Macmillian.

Malinowski, B. 1922. Argonauts of the Western Pacific. London: Routledge.

Mauss, Marcel. 1989 [1923]. The Gift: The Form and Reason for Exchange in Archaic Societies. London: Routledge.

Michener, H.A., J. DeLamamater and S. Schwartz. 1991. Social Psychology. San Diego: Harcourt.

Miller, Daniel. 1993. Unwrapping Christmas. Oxford: Clarendon Press.

Rhodes,R. 1971. “Packaged Sentiment.” Harpers Magazine(December): 61-66.

Sinha, Anjana Miatra. 1995. “Greetings with Cards.” The Hindu. http//: www.webpage.com/hindu/index.

Sahlins, M. 1972. Stone age Economics. London: Tavistock.

Schwartz, Barry. 1967. “The Social Psychology of the Gift.” American Journal of Sociology 73: 1-11.

Wandycz, K. 1991. “Love Means Never Having to Say Anything.” Forbes. April. Pp. 88-89.

Woodyard, Chris. 1997. “Father’s Day Cards Evolve with Times.” USA Today June 12.

Wright, Susan E. 1986. Social Science Statistics. Newton, MA: Allyn and Bacon, Inc.

Copyright 1998 Electronic Journal of Sociology