A whole history remains to be written of spaces‹which would at the same time be the history of powers‹...from the great strategies of geo-politics to the little tactics of the habitat.
Michel Foucault, Power/Knowledge

The work reported here is the fourth in a sequence of studies by the author (Umiker-Sebeok 1979, 1981, 1985) on representations of gender and the life cycle in U.S. print ads. The first three studies draw inspiration primarily from the field of ethology and the method of visual analysis of gender representations the sociologist Erving Goffman derived from that discipline (Goffman 1979). Space plays a central role in this approach to the construction of gender, as it does in general in Goffman's interaction-based social theory. Echoing Goffman in a work which appeared in the same year that Goffman's Gender Advertisements was published, Anthony Giddens writes that
 

Most forms of social theory have failed to take seriously enough not only the temporality of social conduct but also its spatial attributes. ...Neither time nor space have been incorporated into the centre of social theory; rather, they are ordinarily treated more as "environments" in which social conduct is enacted...rather than as integral to its occurrence. (1979: 201-10).

In a later elaboration of the theory of structuration, Giddens follows Goffman in adopting the term Umwelt to refer to "a phenomenal world with which the individual is routinely 'in touch' in respect of potential dangers and alarms....a core of (accomplished) normalcy with which individuals and groups surround themselves." (Giddens 1991: 244). Goffman had borrowed the term from ethology (probably from Konrad Lorenz) without, unfortunately, tracing the term to its source, in the genial writings of the biologist Jakob von Uexküll (1982 [1940]; 1992 [1934]). Von Uexküll's Bedeutungslehre sets out a biologically-grounded theory of meaning based on Immanuel Kant's idea that it is the a priori categories of space and time that make it possible for a subject to perceive a world. The subjective "cocoon" or umwelt is an intricate mosaic which is built up out of three interrelated spatial mosaics - which he calls operational tactile, and visual space - through the mediation of local signs, directional signs, impulse-to-operation signs, and time signs, "self-expressions (the ego-qualities) of cells that are distributed either as tactile cells in the skin on the surface of our bodies or as specific perceptive cells in the retinas of our eyes" (T. von Uexküll 1982: 10-11). In the umwelt model, space is "the subjective framework consisting of organizing signs with a system of coordinates corresponding to the three directional levels: right-left, above-below, in front-behind. The origin of these levels lies in the inner ear." (T. von Uexküll 1982: 87). During the construction and recognition of objects and processes (or complex signs), "it is not static memory-images that are used" but the process of image-formation itself is repeated and the sequence of impulses for the movements of our muscles...is thereby compared with programs of sequences of impulses that are stored in our memories" (T. von Uexküll 1982: 10-11: 16). Following Kant, J. von Uexküll calls these programs "schemata".

Recently, a number of authors such as Mark Johnson (1987), George Lakoff (1987, 1988), Eleanor Rosch, and Francisco Varela (Varela, Thompson and Rosch 1991) have given new life to the Kantian tradition by proposing models of cognition as embodied action. A basic tenet of this view is that "Cognition is not the representation of a pregiven world by a pregiven mind but is rather the enactment of a world and a mind on the basis of a history of the variety of actions that a being in the world performs." (Varela, Thompson and Rosch 1991: 9).
 

Meaningful conceptual structures arise from two sources: (1) from the structured nature of bodily and social experience and (2) from our innate capacity to imaginatively project from certain well-structured aspects of bodily and interactional experience to abstract conceptual structures. Rational thought is the application of very general cognitive processes‹focusing, scanning, superimposition, figure-ground reversal, etc.‹ to such structures. (Lakoff 1988: 88)

Power and Kinesthetic Image Schemas

The kinesthetic image schemas of Compulsion, Blockage and Containment (Johnson 1987) lie at the root of our everyday conception of power, which may be defined as the ability to control movement through space/time in order to maximize access to those resources in the environment which enhance survivability.

Compulsion 

The compulsion schema consists in a force which travels along a trajectory at a certain speed and moves or carries an object or person along in its path (Johnson 1987: 45). Learned during early interactions with the environment, such as when a child is swept up and carried away by a parent, the compulsion schema, like all image schemas, is extremely productive in terms of social structuration. Discursively elaborated, it plays a role in our understanding of the power of discourse ("The crowd was carried away by the demagogue's powerful oratory"), social institutions ("The tax system drove them to ruin"), and our psychological and emotional life ("He was swept along by his love for her"). To be empowered is to serve as the force that controls the actions of others to ones own advantage. From the perspective of the person who is acted upon, compulsion may be empowering, if it leads to the acquisition of resources needed to survive, e.g., when an emotion "drives us" toward an alliance with someone with whom we can expand our access to resources, or when a rescuer carries an accident victim to medical help. Since it takes control of our actions from us and we cannot count on the good will of the controlling force, compulsion is generally seen as disempowering, however.

Blockage and Containment

If power is the ability to move through space/time in order to access resources, the blockage schema - "a force vector encountering a barrier and then taking any number of possible directions" (Johnson 1987:45) - is an important one. Whether in the case of a baby encountering a sofa in its path toward the other side of the room, the less tangible blockage of a social proscription against eating certain kinds of foods or marrying a certain kind of person, or discursive impediments to movement ("During the debate, the Republican candidate frequently backed the Democratic hopeful into a corner"), blockages are one of the primary schemas in controlling movement and access to resources. Generally speaking, the person who can control the blockage (and can remove the restraint) has more power than the person who cannot.

An elaboration of the blockage schema wherein the blockage is continuous so as to separate "inside" from "outside", the containment or container schema lies at the heart of our systems of categorization (Zerubavel 1991), logic (Johnson 1987) and territorialization of space and time. Containers are designed both to define and protect resources, whether it is a stall around a herd of horses or a word or image which "delineates" a certain semiotic "content" or resource. The body is the quintessential container and the model for many of our other containers. As a living and therefore open system, the body must take in enough information to sustain itself but not too much information, which would risk destroying the system. Each exchange of information poses a danger, as we try to avoid the risk of boredom, on the one hand, and anxiety, on the other, in our search for optimal experience (Csikszentmihalyi 1990). As Goffman notes, "a body is a piece of consequential equipment, and its owner is always putting it on the line" (1972: 167). Around this primary container, people construct multiple protective layers of spatio-temporal organization (settings and events) which constitute what Giddens, in his discussion of risk and security in the late modern age calls a "protective cocoon...the mantle of trust that makes possible the sustaining of a viable Umwelt" (1991: 129). In order to increase security, through the increase of control over resources, one must take risks. The greater the risk, the greater the potential gain. Power is the ability to control movement to and from containers. For example, membership "in" the socially-sanctioned container known as a "family" will entail power to allow outsiders inside the family compound or to keep them out. By virtue of their status "inside" or "outside" the category containers such as "child" or "female", family members will have differing amounts of power to move to and from the compound, introduce outsiders into it, access family resources, and move resources into or out of various family containers such as bank accounts, closets, refrigerators, and computers. Containment both empowers (by protecting resources inside from forces outside) and disempowers (by controlling our behavior as insiders, e.g., as husbands and wives "in a marriage", prisoners "in" prison, students "in" school, or professionals "in" a profession).

Gender, Power and Kinesthetic Image Schemas in Print Advertising: Background

The structuration of gender, or the construction of gendered spaces, is viewed here as a process of interaction ‹ the interaction of those containers we call our bodies. The code for the social construction of traditional gender normalcy in print advertising in the 1970's, as described by Goffman (1979) and Umiker-Sebeok (1979, 1981, 1985), may be restated as follows in terms of kinesthetic image schemas: Masculinity and masculine spaces are constructed through repeated instances of (or the exhibition of the potential for) exerting force over animate and inanimate objects and overcoming obstacles, resulting in an increase in the size of territory controlled. Femininity and feminine spaces are constructed through submission to force and avoidance of or submission to obstacles, with a resultant decrease in the size of territory controlled. Stereotypical nonverbal gender displays included:

• relative size (males take up more space than females in terms of body size and/or posture);
• psychological withdrawal from the social context (males stay attuned and ready for potential threats to their control of the situation while females' attention drifts away)
• ritualized subordination (females adopt postures which indicate submission to control by others; females show more appeasement behaviors such as smiling; females adopt softer [i.e., curvilinear] postures than males, who exhibit more dominant [i.e., angular] postures);
• function ranking (males control the actions of others, while females are controlled);
• locations (females are found primarily in domestic settings or containers, males in a wide variety of other locations).
• touch/manipulation (males manipulate objects while females caress or finger them);
• movement (males move faster and farther than females);

The monopoly of the older [patriarchal] code has been broken...."de-linking of masculinity [and femininity] from fixed familial, cosmological, and sexual positions has effectively transformed it [them], ...into a floating signifier, free within any given promotional context to swirl around and substitute for its paired opposite at will. (Wernick 1991: 62-3)

Methodology

As a means of approaching the question of the current state of gender portrayal in print ads, it was decided to analyze a random sample of contemporary U.S. print ads from the point of view of the umwelt theory Goffman never had a chance to fully exploit. Based on the results of this analysis, a second study would be conducted to see if some of the elements of gender portrayal in print ads are interpreted in similar way outside of print ads.

Analysis of Ads

A random selection of magazines was made over a period of three weeks from municipal recycle bins in Bloomington, Indiana. From this corpus, two issues of each of 38 different magazines were selected for analysis. The issues were published between July, 1993 and January, 1994. Twelve of the magazines are aimed at primarily male audiences (American Legion, Business Week, Details, Entrepreneur, Esquire, Forbes, GQ, Inc., Inside Edge, Men's Fitness, Men's Journal, Sports Illustrated). Thirteen of the magazines were aimed primarily at female audiences (Better Homes & Gardens, Cosmopolitan, Country Living, Family Circle, Glamour, Good Housekeeping, Harper's, Ladies' Home Journal, Lear's, Mademoiselle, Sassy, Woman's Day, YM). Sixteen magazines targeted mixed-gender audiences (Discover, Life, Midwest Living, Modern Maturity, Money, Muscle & Fitness, New Choices, New York Times Magazine, The New Yorker, Newsweek, People, Time, US News & World Report). The first twenty full-page ads featuring at least one human form were chosen from each of two issues of each magazine. The corpus of ads thus selected consisted of 1006 ads, with 387 aimed as male audiences, 406 at female audiences, and 213 at mixed audiences.
The author and two student assistants (one male, one female) independently coded the ads for models' approximate age, gender, race, social grouping, body size and build, relative size, availability to the environment, postures, facial expressions, actions, interactions, and settings. Agreement among coders was .82. Chi square tests of the frequency with which males and females were associated with different features were performed. In the discussion that follows, asterisks will be used to refer to the following levels of significance derived from the chi square tests:

* = Significant (p <.05)
** = Moderately significant (p < .01)
*** = Highly significant (p <.001)

Experimental Test

Based on the above analysis, a hypothesis about the relationship between space, power and gender was made and submitted to an experimental test. Fourteen features which seemed to be gender marked in the ads were taken for testing. Two unisex metal sculptural armatures were used to represent these features. For each feature, the armatures were shaped and positioned in such a way as to keep all features constant except the one in question (e.g., for relative height, the armatures differed only in height, not in other body dimensions, posture or activity level). For those features which involved one figure doing something to another (body containment or enclosure, lifting, and shoulder hold), one image was made which contained the two, contrasting figures (container/contained; lifter/lifted; shoulder holder/shoulder held), for a total of three images. For the eleven other features, each figure was contained in a separate image (e.g., wide stance in one, narrow stance in another), for a total of 22 images. 60 subjects (28 male, 32 female), ranging in age from 19 to 50 years, were tested. In one type of experimental session, 29 subjects were shown the pictures of the armatures and asked to rate them on a seven-point semantic differential scale, from Weak (1) to Strong (7). In another type of session, 31 different individuals were asked to rate the same images on a seven-point semantic differential scale, from Masculine (1) to Feminine (7). The average rating on both strength and gender was calculated for each figure.

Results

Within this umwelt framework, the gender system in U.S. print advertising today remains largely the same as it was in earlier studies. Maleness and femaleness define the endpoints of a continuum of control over spatio-temporal resources. Males still exhibit more control over spatio-temporal resources than do females. Risk-taking is associated with increased control, and males exhibit more risk-taking behaviors in ads than do females.

Relative Size

Americans have repeatedly demonstrated a preference for body types which are neither too thin (and thus lacking in muscle mass and strength) nor too fat (and whose bulk inhibits the ability of the body to move); (see the review in Jackson 1992). During this century, however, U.S. females have come to diverge from males by preferring a thin or ectomorphic body type for themselves, with thinness being interpreted as a sign of high social status (and power). The media have long promoted the ectomorphic body type for women (e.g., Gagnard 1986)2 and the research being presented here confirms the continuation of this tradition and indicates that the predicted shift to a new fit ideal of female body attractiveness has not occurred, at least in print advertising (Jackson 1992). Male bodies tended to take up more space than those of females. Females were judged to be thin more frequently than males were*** and males were judged to be of average build more than females***. There was a greater frequency of obese males in the ads than obese females**. A number of explanations of the current thin ideal have been proposed (see the review in Jackson 1992). Most appealing from the point of view of this paper and the model it proposes is the idea that thinness in a woman makes her more angular and hence more masculine (see below re. curvilinear forms), and therefore signifies her identification with more masculine goals such as power and status (cf. Beller 1977). Since being thin also reduces body mass and hence masculinity, women can both signal their pursuit of masculine goals and still appear feminine and hence attractive to the opposite sex.

Males also tended to be taller than females in the ads*** and they were more likely to be looking down**. Males are more likely to be looking down if there is a female present than in single-gender settings***, whether or not the female is of a different height. As Lakoff (1988) has noted, power (control) is commonly conceived of as coming from above, or "the high ground". Empirical evidence on the intrapersonal and professional implications of body appearance (reviewed in Jackson 1992) indicates that being tall confers many advantages on males in terms of controlling events and people. Similar relationships between tallness and control have not been found for women, perhaps because being tall makes a woman appear more threatening and hence more masculine (due to increased size), but, unlike thinness, does not at the same time reduce size (and masculinity) in another dimension, thereby softening the effect and reducing the likelihood that some negative interactional outcomes will occur.

Postures can, of course, make the body appear larger or smaller. In the ads, males exhibited postures which increase the space occupied by their body more than did females** , and females displayed postures which reduce body size more than did males*. Females were shown lying down more than males*, and males standing up more than females*.

Psychological Withdrawal

As Goffman noted, in the traditional gender code of advertising, "Women more than men...are pictured engaged in involvements which remove them psychologically from the social situation at large, leaving them unoriented to it, and presumably, therefore, dependent on the protectiveness and goodwill of others who are (or might come to be) present" (1979: 57). Models were judged whether they were in a posture which indicated a readiness to respond to the social situation in which they found themselves, and males were determined to adopt such postures more frequently than females***.
One common way to indicate psychological withdrawal is to cover the face or mouth, something which females did in the ads more than did males*. Self-enclosure, where the body partially withdraws from the surrounding space into its own self-made cocoon, was exhibited more by females than males**. Females were also associated more frequently than males with more extreme forms of self-enclosure, such as curling up in a fetal position, which signal fear of or vulnerability to the surrounding space. Males tended to use forms of self-enclosure such as folding the arms across the chest, a posture which carries the meaning of "defiance" rather than anxiety about it.

Ritualized Subordination

There were highly significant gender differences in the frequency with which they displayed certain emotions (judged by facial expression), with males being associated with more aggressive expressions, females with looks of appeasement. Males were more frequently "serious"*** or angry* than females. Females were seen as happy** or seductive*** more frequently than males were. Females also smiled more than males**. Curvilinear forms are usually seen as softer and weaker than rectilinear forms (Espe and Krampen 1994), and females in the ads were found to adopt soft, curvilinear, "bending" postures significantly more than males did***.

Function Ranking

Males tended to control the actions of females rather than vice versa. Male bodies tended to take control of the female body by lifting it significantly more often than females lifted males***. Females were also more likely to hold onto the arm of a male as the male led them than were males likely to do this with females*. In addition, when females put their arm around the shoulder of a male (see below), they tended to rest their weight on the male, thus allowing his body to support theirs.

One way in which an individual signals/maintains power over another is by containing the space of the other person, generally by encircling the body with the arms and/or legs. In these ads, the male body tended to encircle the female body more frequently than the female body contained the male body**. Further, in 65% of the cases where the female enclosed the male's body, the female's body was also being enclosed by the male, this reciprocal encasement being found in only 31% of the cases where the male body contains the female. In other cases where the female body enclosed a male body, the female was seated behind the male on a motorcycle or other means of transportation, holding onto the male for safety. Body containment is the visual metaphor employed to signify the merging of individual body containers into a broader territory or cocoon (a couple or family) and is thus a strategy employed almost exclusively in mixed-gender contexts. Except where there are babies or young children, there were no examples of single-gender body containment.

Another method of controlling someone is to block access to that person. Males tended to extend their arm to define/control their companion's space (and block access to her) more than females did this to males***, for example by putting a hand on the wall behind the head of the partner. Men did this more frequently with females than with males***. It was also more common for access to females to be blocked by objects more than access to males**.

Shoulder holds - where one person puts their arm around the shoulders of another person - is another nonverbal means of controlling another person's space. In the ads, males and females did this about as often, but shoulder holds by females tended to be accompanied by having the woman rest her weight on the man, while men rarely do this with females. The impression created by this posture is that the female's space is dependent upon the male for support*** (cf. above re. function ranking).

 Experimental Evidence

The results of the semantic differential experiment provide additional support for the relationship between body space, gender and power being proposed here. Table 1 below shows that the more space a figure took up (e.g., wide stance, height, arms out, big shoulders), the more it was perceived to be "masculine" and the more it was perceived as "strong". The more the figure controlled the space of the other figure (e.g., enclosing, lifting, or shoulder holding). Not all of the figures proved to be as reliable gender cues as others, however. Figures which seem to be among the reliable cues to gender include: Body containment, Body support, Height, and Wide stance. Outstretched arms, wide shoulders, and shoulder holding appear to be weaker gender signals. Since stance was a strong cue, further research is needed to test the hypothesis that it is lower torso signals such as wide stance that are more important in gender display than are those which involve the upper torso. (It is interesting to note in this regard that, in the ads examined here, males were more likely to be shown with the lower torso only than were females**.) There were no major differences between the semantic differential rankings by male and female subjects.
 

 Table 1 Relationship Between Gender, Body Space and Strength
 

Name of Figure	Average Gender Ranking*	Name of Figure	Average Strength Ranking**
Enclosing figure	2.45	Lifting figure	5.66
Lifting figure	2.71	Enclosing figure	5.52
Taller	2.97	Wide stance	5.48
Wide stance	2.97	Taller	5.48
Arms out	3.52	Arms out	4.28
Shoulder holder	3.61	Wide shoulders	4.21
Wide shoulders	3.90	Shoulder holder	4.10
Shoulder held	4.48	Narrow shoulders	3.97
Narrow shoulders	4.61	Shoulder held	3.72
Arms at sides	5.0	Arms at side	3.06
Narrow stance	5.25	Narrow stance	2.97
Shorter	5.38	Lifted figure	2.61
Lifted figure	5.45	Enclosed figure	2.45
Enclosed figure	5.52	Shorter	2.0

* Scale: Masculine 1 2 3 4 5 6 7 Feminine
** Scale: Weak 1 2 3 4 5 6 7 Strong

Locations

When a location could not be determined for an advertisement, it was coded as "Other", and it was found that females were much more likely to be de-contextualized than were males***. The presentation code whereby the female body stands as a sign for the universal, ahistorical essence of woman continues to operate within these advertisements. Females gazed more at the camera*** than did males, especially in all-female ads*** as opposed to mixed-gender ads. Since females were also judged to be more seductive in single-gender than in mixed-gender ads***, and more seductive than males in single-gender ads***, it may be concluded that the camera represents the male observer, not the female. (Males were less seductive in single- than mixed-gender ads**.) The cultural ritual of exposure of the female body for male evaluation (Henley 1977; Spitzack 1990) does not appear to have diminished in print advertising.

If the body is the individual's most basic territory, the built environment may be seen as many layers of more or less protected and protective containers or secondary territories that the individual builds around itself. The private home is perhaps the most intimate of these cocoons and is itself constructed of several layers, some more private than others. Beyond the home, we find ever expanding circles of containment: the family compound, the neighborhood, the town, etc. The only spaces in the built environment which were marked as female were in the home, while four of the five built spaces marked as male were outside the home. Males thus appear to be associated with wider, more public spaces.

There was not a marked difference in the frequency with which females and males were pictured in homes: 20% of the ads which contained a female featured a domestic setting, compared with 12% of ads which featured a male model. However, it was the more private and protected spaces within the home ‹ bedrooms, bathrooms and, to a lesser extent, kitchens ‹ which were marked as female. Gender-neutral interior and exterior domestic spaces included the more public spaces of the home: dining rooms, living rooms, and the border areas of backyards and driveways. (No examples of attics, basements or garages were found in the ads.)

The family bathroom is, along with the bedroom, one of the most private and "protected" of rooms in U.S. culture. In the ads, men appeared in bathrooms more than did women**. Females in bathrooms were shown, in the vast majority of cases, in self-absorbed grooming, with only a few cleaning the room or conversing with a female family member. Far fewer of the males are pictured grooming, and, if they were, for example, bathing, they were usually upright in a vigorous shower rather than luxuriating in a bathtub. They are as likely to be making love or playing as in the more feminine detached, non-controlling activities. Another indicator of the gender definition of the bedroom is the fact that bedrooms were featured in ads in female-audience magazines at four times the rate they appeared in ads in male-audience magazines.

The second space within the home that was defined as female is the bedroom. Females were more likely to be found in bedrooms than were males*. Again, the activities that women engaged in the bedroom appear to be different than male activities in the same space. Half of the bedroom scenes with females showed them modeling clothes or makeup (more self-absorbed detachment from the space), and the second most frequent activity was sleeping (almost complete detachment and lack of control). By contrast, the most common activities for males were making love and lying awake in bed, worrying about financial problems. Another indicator of the gender definition of the bedroom is the fact that bedrooms were featured in ads in female-audience magazines at twice the rate that they were in ads in male-audience magazines.

Feminist criticism of advertising in the eighties regularly focused on the fact that women were commonly shown in servile and often ridiculous roles in the family kitchen. It is clear that advertisers have taken this to heart, for, while females were still found more frequently in the kitchen than males, this difference was not statistically significant. However, if one looks at how events are controlled in kitchens, it appears that these rooms are still something of a female domain. 38% of the males pictured in kitchens were handling food, compared with 71% of females. Further, all but one of these men were cooking alongside females, while 67% of the females were cooking without help. The one male who was cooking "solo" was a young man who was clowning around, clearly displaying his distance from the role he was performing. While women do more of the cooking, men do more of the eating. Almost half of the males in kitchens were eating, while less than a quarter of females were. Another indicator of the gender definition of the kitchen is the fact that kitchens were featured in ads in female-audience magazines at 20 times the rate at which they were included in ads in male-audience magazines.

Only one space in the private home was marked as male: the den or study, where males appeared more frequently than females*. It is certainly no coincidence that this is the only room in the home that contradicts the fact that "the home is the spatial institution containing the least amount of socially valued knowledge" (Spain 1992: 235). As in the examples of kitchens, the differences between males and females in terms of what they are doing in these spaces is significant. All of the males shown in dens were working (writing, communicating, lecturing, reading, etc.), while only 60% of women were doing so. Furthermore, men tended to associate with men rather than women in these rooms (67%), while women tended to be with men (60%). Another indicator of the gender definition of dens is the fact that they were featured in ads in male-audience magazines at five times the rate of ads in female-audience magazines.

Males were over three times more likely to be pictured in non-domestic settings as in the home***, while females were about as likely to be shown in the home as outside it. Of all the non-domestic settings examined (Workplaces, Sports facilities, Urban, Rural and Wilderness), only Suburban and Urban settings‹ which tended to be associated heterosexual courtship, marriage, and the family ‹ were gender-neutral. The rest ‹ workplaces, sports facilities, rural sites and the wilderness ‹ were marked as male, and were associated with all-male groupings more than with mixed-gender or all-female sets of participants.

Males were far more likely than females to be pictured in the workplace***. They were also more likely to be found in the workplace with other males than females were likely to be found with other females*. Another indicator of the gender definition of the workplace is the fact that they were featured in ads in male-audience magazines at more than six times the rate they appeared in ads in female-audience magazines (and more than three times the rate for mixed-gender audiences).

Males were more likely to be depicted in rural surroundings than females**, and men were more likely to be pictured in rural surroundings with other males, rather than with females*. Not only were females less likely to be found in rural settings, they were more likely to be found in such contexts with males rather than other females*. Females, it would appear, need male protection when they wander far from home.

Males were five times more likely to be situated in sports facilities than were females***, and, again, males tended to be associated in these sites with other males*, while females were likely to be accompanied by males than by females*. Ads featuring sports facilities were 24 more times likely to be featured in male-audience magazines than magazines aimed at a primarily female audience, and sixteen times more frequent than in mixed-audience magazines.

Males more frequently linked to wilderness settings than were females* and, again, they were more likely to be there with other males than with females*. Females, on the other hand, were more likely to be with males than females in wilderness locations*. Wilderness scenes were twice as likely to be found in male-audience magazines as in female-audience or mixed-audience magazines.

 Touch/Manipulation

Control of events involves command of both what Giddens (1979) calls "allocative resources" - dominion over the material world - and control of "authoritative resources", or dominion over the social world. We have argued that the control of authoritative resources derives from command of allocative resources. We have already seen that, in the ads in question, males tended to dominate space physically in a number of ways, and that this management of spatio-temporal resources translates into management of social resources. Spaces in the built environment in the ads which were associated with males - dens, workplaces, sports facilities, rural settings, and wilderness areas - tended to be filled with cultural tools which could augment models' ability to control allocative resources. Even when females were in these "male" settings, they were less likely to be using such tools. The only female domain which emphasized control of allocative resources was the kitchen, with its various tools and appliances for the manipulation of food stuffs. The result was that males were generally far more likely to be demonstrating dominion over the material world than were females**.

The tools of manipulation varied with the type of setting, with dens/studies and workplaces emphasizing tools for obtaining, manipulating and communicating business and financial information (computer equipment with spreadsheets and stock quotations, and various sorts of communications technologies) and tools for manipulating physical objects (manufacturing machinery, medical appurtenances, construction equipment and blueprints, vehicle repair supplies). Table 2 shows occupations associated with males and females as well as gender-neutral professions. Male occupations were much more likely to involve manipulation of the material world than were the female occupations.

In rural and wilderness settings, males were 2.5 times more likely than females to be shown controlling the material world. As the distance from "civilized society" increased, tools and equipment associated with hunting, ranching, fishing, farming, jogging, boating, horseback riding, baseball and exploring become the prominent cultural implements for command of the material world.

Table 2 Occupations in the Workplace (in % of occupations found)
 

	Males	Females
Male Occupations:
Business owners/managers Mechanics/skilled craftsmen  Factory workers/laborers Engineers/architects  Physicians/pharmacists/dentists Researchers/Professors	49% 13 8 6 4 2	19% 5 0 0 2 0
	82%	26%
Female Occupations:
Office workers/clerical   School teachers	3 0	32 22
	3%	54%
Gender-neutral Occupations:
Musicians/artists/interior designers  Writers/ journalists	11 4	14 5
	15%	19%

 
 
Movement

Males demonstrated their ability to control larger amounts of both horizontal and vertical space by engaging more frequently than females in more active pastimes such as auto racing, diving, football, boating, hunting, ranching, fishing, farming, jogging, boating, horseback riding and racing, baseball, karate, marathon running, mountain climbing, rodeo, competitive roller skating, rugby, soccer and white water kayaking and exploring*.

Motor vehicles in the ads served a dual function as signs of both freedom to escape the boundaries of civilization (speed, power and sexual potency), on the one hand, and of the middle-class family home, on the other (Umiker-Sebeok 1994). Males were pictured in or with motor vehicles more than were females, but this difference was not statistically significant. When pictured with or in motor vehicles, both males and females were more likely to be with members of the opposite sex than with someone of their own gender, a relationship which was statistically significant for females** but not for males. Looking at who is controlling the motor vehicles in these ads, however, one finds that it is more likely to be the male than the female. Males were 2.3 times more likely to be driving the vehicle and 12 times more likely to be repairing it. Females were 2.7 times more likely than males to be a passenger in the vehicle. The types of motor vehicles with which males and females were pictured also differed: males with trucks, tractors and airplanes, females with passenger cars or minivans. The two types of motor vehicles for which this gender difference did not exist ‹ sports cars and motorcycles‹were those used for courtship scenes, where the power of the vehicle served as a metaphor for the sexual potency of the couple. In these ads, the male was always driving, so that one assumes that the vehicle (and the power) belongs to him more than to the female passenger.

 Risk

Only males were observed involved in high-risk physical activities such as diving, football, exploration, hunting, mountain climbing, horse and auto racing, rugby, rodeo, white water kayaking and karate and were much more likely to be taking financial risks than were women**. A key element in the representation of these risky activities was the breaking down of natural or cultural obstacles: forging rivers, scaling mountains, breaking speed records, taming the wild, invading a competitor's market, investing large sums of money. If power can only be achieved through risk-taking, then, judging from these ads, males are much more likely than females to achieve power.

 Conclusion

 Evidence has been presented to support the claim that, while some shift has taken place in the gender code in print advertising, these changes have been superficial, leaving the underlying image schemas largely unchanged. When looked at from the point of view of a random selection of ads, and the millions of tiny interactions through which gendered spaces are constructed in them, gender in print advertising is not the "floating signifier" that Wernick claimed (1991), nor does it resemble the free negotiation of gender outside advertising described in Ginsburg and Tsing 1990. This research has shown the relationship between power and the structuration of gendered spaces in print advertising. It remains for future research to apply the same frame of analysis to advertising in other media, such as television commercials and World Wide Web promotions, which offer greater levels of freedom to introduce ambiguity and thus to introduce more radical shifts in the underlying gender code.

References

Ardener, Shirley (ed.) 1993 Women and Space. Ground Rules and Social Maps. Oxford: Berg Publishers.
Belknap, P. and W.M. Leonard 1991 "A conceptual replication and extension of Erving Goffman's study of gender advertisements," Sex Roles 25 (3/4): 103-18.
Beller, A.S. 1977 Fat and Thin: A Natural History of Obesity. New York: Farrar, Straus, and Giroux.
Courtney, A. and S. Lockeretz 1971 "A woman's place: An analysis of the roles portrayed by women in magazine advertisements," Journal of Marketing 8(1): 92-95.
Courtney, A. and T. Whipple 1983 Sex Stereotyping in Advertising. Lexington, MA: D. C. Heath.
Csikszentmihalyi, Mihaly 1990 Flow. The Psychology of Optimal Experience, New York: Harper & Row.
Espe, Hartmut and Martin Krampen 1994 "Eindruckswirkungen visueller Elementarformen und deren Interaktion mit Farben," unpublished manuscript.
Ferguson, J.H., P.J. Kreshel and S.F. Tinkham 1990 "In the pages of Ms.: Sex role portrayals of women in advertising," Journal of Advertising 19 (1): 40-51.
Foucault, Michel 1980 Power/Knowledge. New York: Pantheon Books.
Gagnard, A. 1986 "From feast to famine: Depiction of ideal body type in magazine advertising: 1950-1984," Proceedings of the 1986 Conference of the American Academy of Advertising, R46-50.
Giddens, Anthony 1979 Central Problems in Social Theory: Action, Structure and Contradiction in Social Analysis. Berkeley: University of California Press.
-- 1991 Modernity and Self-Identity. Self and Society in the Late Modern Age. Stanford: Stanford University of Press.
Ginsburg, Faye and Anna L. Tsing, eds. 1990 Uncertain Terms. Negotiating Gender in American Culture. Boston: Beacon Press.
Goffman, Erving 1972 Interaction Ritual. London: Allen Lane.
--1979 Gender Advertisements. Cambridge, MA: Harvard University Press.
Henley, Nancy 1977 Body Politics: Power, Sex, and Nonverbal Communication. Englewood Cliffs, NJ: Prentice-Hall.
Jackson, Linda A. 1992 Physical Appearance and Gender. Sociobiological and Sociocultural Perspectives. Albany, NY: State University of New York.
Johnson, Mark 1987 The Body in the Mind. The Bodily Basis of Meaning, Imagination, and Reason. Chicago: University of Chicago Press.
Klassen, Michael L. and Cynthia R. Jasper and Anne M. Schwartz 1993 "Men and Women: Images of their Relationships in Magazine Advertisements," Journal of Advertising Research March/April: 30-39.
Lakoff, George 1987 Women, Fire and Dangerous Things: What Categories Reveal about the Mind. Chicago: University of Chicago Press.
-- 1988 Cognitive semantics. In Meaning and Mental Representation, ed. Umberto Eco et al. Bloomington: Indiana University Press.
Nakayama, Tom 1989 Stale Roles and Tight Buns: Video examines images of men in advertising. Media&Values No. 48: 17.
Ruggiero, J.A. and L.C. Weston 1985 "Work options for women in women's magazines: The medium and the message," Sex Roles 12 (5/6): 535-47.
Soley, L. and G. Kurzbard 1986 "Sex in advertising: A comparison of 1964 and 1984 magazine advertisements," Journal of Advertising 15(3): 46-54, 64.
Spain, Daphne 1992 Gendered Spaces. Chapel Hill, NC: University of North Carolina Press.
Spitzack, Carole 1990 Confessing Excess. Women and the Politics of Body Reduction. Albany, NY: State University of New York Press.
Stale Roles and Tight Buns: Images of Men in Advertising. 1989. Videotape produced by OASIS (Men Organized against Sexism and Institutionalized Stereotypes. Boston, MA.
Steinem, Gloria 1990 "Sex, lies and advertising," Ms. July/August.
Sullivan, G.L. and P.J. O'Connor 1988 "Women's role portrayals in magazine advertising: 1958-83," Sex Roles 18 (3/4): 181-89.
Uexküll, Jakob von 1973 [1928] Theoretische Biologie. (=Suhrkamp Wissenschaft 20.) Frankfurt: Suhrkamp.
-- 1982 [1940] "The theory of meaning," Semiotica 42(1): 25-82. [Originally published, in German, in 1940.]
-- 1992 [1934] "A stroll through the worlds of animals and men: A picture book of invisible worlds," Semiotica 89(4): 319-391. [Originally published, in German, in 1934.]
Uexküll, Thure von 1982 "Introduction: Meaning and science in Jakob von Uexküll's concept of biology," Semiotica 42 (1): 1-24.
Umiker-Sebeok, Jean 1979 "Nature's way? Visual representations of childhood in American advertisements," in Semiotics of Culture, eds. Jean Umiker-Sebeok and Irene Portis Winner, pp. 173-220. The Hague: Mouton.
-- 1981 "The seven ages of woman: A view from American magazine advertisements," in Gender and Nonverbal Behavior, eds. Clara Mayo and Nancy M. Henley, pp. 209-52. New York: Springer. [German translation with German ad illustrations: "Die 7 Lebensalter der Frau ‹ ein Blick auf die 70er Jahre," in Frauen der Werbung. Aufklärung über Fabeltiere, ed. Christiane Schmerl, pp. 93-130. Munich: Frauenoffensive.
-- 1985 "Growing signs: From firstness to thirdness in life and advertising," in Iconicity. Essays on the Nature of Culture, eds. Paul Bouissac, Michael Herzfeld, and Roland Posner, pp. 527-73. Tübingen: Stauffenburg Verlag.
-- 1994 "'New' images of the family in advertising: The view from print ads in the United States". Paper presented to the International Conference on Family, Cultures and New Identities of "Motherhood" and "Fatherhood", Università di Roma "La Sapienza", Rome.
Varela, Francisco J., Evan Thompson & Eleanor Rosch 1991 The Embodied Mind. Cognitive Science and Human Experience. Cambridge, MA: MIT Press.
Venkatesan, M. and J. Losco 1975 "Women in magazine ads: 1959-1971," Journal of Advertising Research 15 (5): 49-54.
Wernick, Andrew 1991 Promotional Culture. Advertising, Ideology and Symbolic Expression. Newbury Park: Sage.
Whipple, T. and A. Courtney 1980 "How to portray women in TV commercials," Journal of Advertising Research 20(2): 53-59.
Zerubavel, Eviatar. 1991 The Fine Line: Making Distinctions in Everyday Life. New York: The Free Press.

Footnotes

 1 The author would like to acknowledge the stimulating and helpful suggestions of the participants in the International Conference on Gender, Space, Perception and the Construction of Social Reality (University of Rome, Italy, November, 1994). Special thanks go to Martin Krampen for having suggested that the author take another look at gender and space in advertising.

 2 The promotion of the thin female has taken place despite the fact that research has shown that males actually prefer the mesomorphic type for both sexes and that the average weight for U.S. females has not only never approached the promoted ideal but has actually been increasing during the time the promotion has been going on (Jackson 1992).