Images of Old Men’s Masculinity: Still a Man?

Sex Roles (2006) 55:633–648 DOI 10.1007/s11199-006-9119-7 ORIGINAL ARTICLE Images of Old Men’s Masculinity: Still a Man? Edward H. Thompson, Jr. Published online: 6 December 2006 # Springer Science + Business Media, Inc. 2006 Abstract Relatively little attention has been paid to later life masculinities. Separate studies were designed to identify the commonly held images of old men and then determine whether or not an old man is perceived as a man or more in terms of a gender-neutral elderly person. The first assessed the commonly held images of positive and negative later life masculinities. A second study examined if old men’s gendered character was perceived similarly in comparison to a young man or an old woman. Findings from the studies revealed age expectations and masculinity norms jointly affected people’s images of old men. Old men were favorably described more often than negatively stereotyped, and there was evidence of distinct late life masculinity expectations. Keywords Old men . Later life masculinities . Stereotypes Recent interest in the social worlds of old men and how age-specific masculinities differentially structure men’s experiences has been noticeable (Arber, Davidson, & Ginn, 2003; Calasanti, 2004; Emslie, Hunt, & O’Brien, 2004; Kosberg & Kaye, 1997; Thompson, 1994). Yet very little is known about what gender and age norms are most applicable to old men’s lives, or how others perceive old men. This information gap raises an intriguing question: To what extent is an elder man recognized as a man? Do people generally view him in terms of his gender as much as his age? Or, does the perceptual salience of gender recede from foreground to background to become less E. H. Thompson, Jr. (*) Department of Sociology & Anthropology, College of the Holy Cross, 1 College Street, Worcester, MA 01610-2395, USA e-mail: [email protected] important in the eyes of others as a man ages, and thus is the older man recognized more in terms of his age-status as “an old man?” I have argued elsewhere (Thompson, 1994) that old men have remained relatively invisible in contemporary social life. They are less visible in comparison to the attention given to younger aged men and boys by both academic researchers and the general public. Discussions of old men and their masculinities within the burgeoning literature on men and masculinities are infrequent. Instead, almost all of the available research is in the gerontology literature and emphasizes men’s aging and the difficulties that old men face in later life. Any discourse on old men as men is very peripheral to the predominant effort to understand men’s aging bodies and health problems, in particular their medical troubles. By principally framing old men’s lives through a lens that medicalizes late life, the old man’s masculinity is pushed into the background. Aging overshadows gender. This is the insidious effect of ageism (Butler, 1969; Green, 1993). An old man is imagined as “old” whether he is aged 67, fit, active, independent, and employed, or age 90, frail and living in a nursing home. Even these differently aged old men surely live gendered lives. Are their masculinities and age-specific gender performances recognized? Despite the large body of research that has examined the way that the public perceives older adults (e.g., Brewer, Dull, & Lui, 1981; Hummert, Garstka, Shaner, & Strahm, 1994; Kite, Stockdale, Whitley, & Johnson, 2005), and the equally established research tradition examining gender stereotyping, very little is known about the way age and gender constructions jointly influence conceptions of old men. The research that has examined the dual effects of a target’s age and gender on people’s descriptions of and attitudes toward old men offers equivocal information about how much or little masculinity expectations are evident. 634 A few studies (e.g., Braithwaite, 1986; Jackson & Sullivan, 1988; Kite, Deaux, & Miele, 1991; Minnigerode & Lee, 1978) report that the age-of-the-target influenced people’s perceptions and descriptions more often than the gender-of-the-target and thus old men’s age status as old men was more influential in determining evaluations of the old man. For example, Kite et al. (1991) found that when respondents described all the qualities they thought characteristic of old men or old women, the attributes provided for the two targets were highly similar (r=0.79) but quite different characteristics were used to describe older and younger men (r=0.12). Kite et al. also noted that the descriptions of old men were less likely to include the masculine traits, behaviors, and physical characteristics ascribed to the younger men. Similarly, Levin (1988) reported that when undergraduates evaluated photographs of the same man at three ages, they saw age whittling away at the attributes commonly associated with masculinity. By comparison, several other studies (e.g., Canetto, Kaminski, & Felicio, 1995; Deutsch, Zalenski, & Clark, 1986; Zepelin, Sills, & Heath, 1987) observed that gender norms remain potent in late life and the gender-of-the-target had an important affect on the way men were perceived and described. Canetto et al. (1995) found that old men were evaluated more favorably in terms of intellectual competence and autonomy than old women; they were perceived as more self-reliant and less passive and dependent. Deutsch et al. (1986) reported that old men were regarded as more attractive than old women, and Hummert, Garskta, and Shaner (1997) noted that old men were less likely to be described negatively than old women. Although researchers have begun to untangle the ways age and gender stereotyping frame how old men are discussed and can be imaged, no study has directly examined the content of the stereotypes about old men or assessed the extent to which an old man’s masculinity is or is not veiled by becoming old. This article details the findings of one study that examined the content that makes up the conceptions young adults have of old men and another that examined the extent to which perceptions of an old man’s trait-masculinity differ when he is compared to a younger man or an old woman. Conceptual Footing It is a basic proposition of the social constructionist perspective on masculinity that there are different kinds of masculinities within society; it is also understood that lived masculinities are negotiated performances that help maintain the gender scripts that are “out there” in the culture, in institutions, and in relationships and reveal relations of dominance and subordination (Connell, 1995; Martin, Sex Roles (2006) 55:633–648 2004; West & Zimmerman, 1987). The culturally idealized form of masculinity may not be the lived form of masculinity at all, but it remains a powerful, perhaps dominant, script against which self and others are evaluated across the life course (Morgan, 1992). Conversations about age-related masculinities are not uncommon, and the masculinity expectations for boys and adult men have been systematically addressed (e.g., Connell, 2000; Kimmel, 1996). But research addressing later life masculinities is in its infancy. Neither the general norms scripting later life masculinity nor the behavioral norms older men institute have been charted. The blueprints of older men’s masculinities remain hazy. But later life masculinities seem to be defined by norms of sociability and reveal more vividly men seeking connections with others, instead of the acts of individuation. Older men’s experiences become more centered on the emotion work of relational concerns (Bleiszner, 1989; Kidder, 1993) and reflection (Ryff, 1991) and less by the triumphs they used to define themselves as younger men. A “busy ethic” (Ekerdt, 1986) replaces the performance-oriented work ethic; a search for coherence replaces the self-reliant, “give ‘em hell” values of earlier masculinities (Erikson, Kivnick, & Erikson, 1994). Study 1 For several decades studies have documented the multidimensionality of stereotypes associated with older adults and identified some of the content within the stereotypes. Initially, Brewer et al. (1981; Brewer & Lui, 1984) established that when people are asked to describe an older adult, rarely do they envision a general superordinate category, rather they maintain images of specific kinds of older adults. The presence of multiple stereotypes explains why incompatible traits such as mellow and complaining are used at the same time to describe an older adult. The work of Brewer et al. also established that people’s cognitive representations involved at least three recognizable stereotypes—grandmothers, elder statesman, and senior citizens—and distinct characteristics were associated with each image. As is suggested by its name, the “elder statesman” is an old man stereotype that exemplifies a distinguished, conservative old man. Traits attributed to this stereotype included some stereotypic masculine attributes such as aggressive, strong-willed, and competitive. The second stereotype associated with old men, the “senior citizen,” emphasized the target’s age rather than gender and represented an inactive, isolated, and old-fashioned “senior citizen” who could be a woman as much as a man. In an influential study of the content of stereotypes of an old person, Schmidt and Boland (1986) asked informants to Sex Roles (2006) 55:633–648 generate traits by describing “the typical elderly adult;” 99 distinct traits were identified. The short phrases or single words listed by the informants represented different content domains, including physical characteristics, social characteristics, personality traits, and emotions. When the traits were sorted by other participants, 12 middle-range stereotypes described “old people.” Four of the 12 were favorable stereotypes—sage, John Wayne conservative, perfect grandparent, and liberal patriarch/matriarch; twice as many stereotypes were negative—nosy neighbor, shrew/curmudgeon, bag lady/vagrant, recluse, despondent, vulnerable, mildly impaired, and severely impaired. Interestingly, among the 99 traits, only ten physical traits (e.g., grayhaired, wrinkled skin, shaky hands) and a few social and psychological traits (e.g., retired) applied exclusively to older adults; the remainder were as descriptive of young and middle-age adults as older adults (e.g., enjoys life, curious, interesting, demanding). Hardly any trait was gendered to be more descriptive of a man than a woman. None of the 12 stereotypes was gendered in ways to describe an old man any more than an old woman. The content of the “perfect grandparent” stereotype entailed gender-neutral traits such as generous, understanding, family-oriented, wise, loving, and useful; even the content of the “John Wayne conservative” stereotype was defined in genderneutral terms such as patriotic, wealthy, and Republican. Hummert’s (1990; Hummert et al., 1994) studies of the content of the stereotypes of older adults extended the work of Schmidt and Boland. She demonstrated that there is, in fact, little similarity in the content of the stereotypes that describe older adults and younger adults, as well as minimal agreement among people of different ages about what stereotypes are associated with an older adult. Using the 86 Schmidt and Boland traits that were not age-specific, three positive and seven negative stereotypes were identified in the 1990 study for older adults. Only two of these ten were analogous to informants’ images of young adults—the “liberal matriarch/patriarch” was similar to “activist,” and “severely impaired” older adult paralleled “invalid” young adult. The 1994 study (Hummert et al., 1994) found fewer stereotypes were generated by the young adults—eight rather than the expected ten: three positive (Golden ager, perfect grandparent, John Wayne conservative) and five negative (shrew/ curmudgeon, despondent, vulnerable, severely impaired, and recluse). Her middle-age informants detailed 12 stereotypes, reintroducing “liberal matriarch/ patriarch” to the list of positive and adding “self-centered” to the list of negative stereotypes. Her older informants revealed even greater diversity in their images of older adults: They identified five positive and seven negative stereotypes from the same list of attributes. As much as the earlier studies uncovered many distinct images of older adults, the studies were not designed to 635 reveal public images of an old man. People were only asked to generate a list of attributes and traits for a generalized target—e.g., an older adult or the elderly, thus gender was masked (Braithwaite, Gibson, & Holman, 1985–86). This may account for why the content generated was overwhelmingly gender-neutral. As much as the informants in the Schmidt and Boland study and Hummert’s studies may have envisioned a “European American male” as the default category and thereby gendered the old person they were evaluating, which Kite et al. (2005) suggest is likely, in fact by not specifying the target’s gender, researchers have not knowingly determined if the gender of the target affects the way informants characterize the target. The present study was designed to reveal people’s images of an “old man.” Remaining as one with the earlier studies, no information about the target’s class or ethnic/cultural background was provided. Only the gender and age of the target was cued. I anticipated that the images people reported would be as contradictory and multidimensional as prior studies have detailed for the gender-neutral older adult. Method Participants Participants were 210 liberal arts undergraduates (age range=18–25, X=20.7, SD=1.78) recruited over a four year period from the author’s introductory sociology courses (n=7), which often had reverse registration priorities and enrolled fourth- and third-year students first. The sample was made up of 45% males, and most students (93%) were white. None were married, and nearly all lived on-campus. At the time data were collected no participant was majoring in sociology, and more participants reported a major in the humanities or natural sciences than one of the other social sciences. Data on the undergraduates’ socioeconomic status is not available; however, tuition at the college is expensive, and more students come from uppermiddle class families than middle-class. Procedure The content of the stereotypes for the target “an old man” was generated with three sentence completions. Sentence completions generally yield fuller descriptions of a target compared to a word association exercise (cf., Schmidt & Boland, 1986). Similar to comparing photographs, the sentence completion method also provides an opportunity to anchor the old man target to a comparison. For the present study the sentence completion exercise was designed to maximize consideration of the old man’s masculinity and age. Two sentence stems anchored the phrase an “old man” to a comparison—a young man or an old woman, effectively producing a demand for informants to pay attention to the age-of-the-target or gender-of-thetarget and hopefully priming rich descriptions of the old man. 636 In the third week of the author’s introductory sociology courses, students were asked to participate in an in-class exercise at the beginning of the class. This was the second in-class exercise in which students were asked to anonymously reveal a personal opinion on a 4×6 piece of paper, submit their comments to the instructor, watch the papers shuffled on a desk, then asked to take back one of their peer’s comments and read aloud the comments to the class as a trigger for class discussion. In this exercise students were again informed that their participation was voluntary. Only one of 211 students chose to not take part in the exercise. The participants were asked to complete the sentence “An old man...” on a 4×6 piece of paper. No verb was provided in the sentence stem. No other verbal instructions were provided, other than repeating the task expectation, to please complete the sentence “An old man...” Once everyone completed the initial sentence, another 4×6 piece of paper was distributed, and the participants were asked to complete the sentence stem “Compared to a young man, an old man...” When everyone completed the second sentence, a third 4×6 piece of paper was distributed and the students were asked to complete the sentence “Compared to an old woman, an old man...” The presentation order of the two comparative sentences varied so that three classes detailed their old man/young man images before they disclosed their old man/old woman images. The priming for informants to attend to the age- or gender-ofthe-target in the final two sentence completions was deliberate; a key objective was to assess if the content of the stereotypes might differ when the old man target was or was not anchored to a comparison other. The attributes, behaviors, personality traits, physical capabilities, physical appearance characteristics, and other qualities disclosed in the informants’ sentence completions provided the data for the study. The full range of attributes/ activities used to describe the “old man” target was preserved, thus each unique expression within the sentence completion was treated as a separate example of stereotype content (e.g., “...is grandfatherly, considerate, full of experience, and walks every morning” was coded as four descriptions). Analysis Strategy In the first step of the content analysis, the qualities attributed to the “old man” target were coded as (a) descriptors, attributes, or activities identifiable as likely characteristic of older men, and (b) non-meaningful sentence completions—e.g., An old man... “accosted my younger sister, therefore I beat the hell out of him.” Coding was completed by two independent judges; each descriptive statement and attribute/activity reported by the informants that could not be agreed upon by the judges was reviewed by a third judge. Interrater reliability was assessed by Sex Roles (2006) 55:633–648 examining the assessments of two judges for the first 50 informants; the percentage of agreement was 0.93. Overall, the third judge was needed in fewer than 10% of the descriptive statements or attributes/activities disclosed in the sentence completions; e.g., when the stimulus “An old man...” was finished with completions such as “...is in many works of art I studied last semester,” the information was coded as a non-meaningful descriptor of old men by both primary judges; but the sentence completion “...is on the dollar bill” was arbitrated and eventually coded as a non-meaningful descriptor rather than evidence of a wise, worldly sage. This process identified 95 descriptive statements about the first unanchored old man target, 101 descriptors for the old man target when compared to a young man, and 89 descriptors for the older man target when compared to an old woman. The next step in the analysis required a coding taxonomy to identify the underlying representations. The same three coders separately reviewed the entire list of descriptive statements and generated possible thematic categories to capture the meaning of the statements and/or qualities; e.g., wisdom, experienced, worldly, and ready to give advice upon being asked were regarded as analogous and evidence of an image that regarded the old man as a “sage.” After the coders generated their lists of possible stereotype categories, lists were shared; similarities among the generated categories were discussed, and after several cycles consensus was reached on the final stereotype categories and their meta-themes. Finally, with the taxonomy completed, the coders sorted the descriptive statements into the categories to identify the frequency for each stereotypes. Because an informant might mention the same stereotype two or more times in the sentence completion, each descriptor had been coded with the informant’s ID number. One frequency measure, absolute frequency, tallied the number of descriptors for stereotype; the second tallied just the number of respondents who mentioned the stereotype in their sentence completions. Interrater agreement among the three coders on the first round was 86%. Descriptive statements that were initially sorted into dissimilar categories were discussed and consensus agreement was reached. This process was repeated for the sets of descriptive statements that defined the old man in comparison to a young man and an old woman. Results Overall, the initial characterizations of the old man target provided different content and fewer images of an old man than the descriptions linked to a “young man” and an “old woman” (see Fig. 1). And generally speaking, as much as Sex Roles (2006) 55:633–648 637 Fig. 1 Stereotypes of an old man. age-related descriptions of the old man recur, underneath these images, evidence of later life masculinities was also prevalent. The range of stereotypes that emerged from the three sets of descriptive statements are summarized in Fig. 1 and reviewed in detailed in this section. There is evidence of four meta-themes involving 12 stereotypes within the initial descriptive statements, attributes, or activities linked to the sentence stem “An old man....” Table 1 summarizes the meta-themes, the stereotypes, and their content. Representations ranged from positive to negative images of later life masculinities and from neutral to negative images of how aging likely affects men. Images of old men’s masculinities involved the positive stereotypes of the old man as a sage (n=40), nurturing, sociable grandfather (n=49), and active retired man (n=70). The sage and grandfather stereotypes are analogous to Schmidt and Boland’s sage and perfect grandparent stereotypes, and the active, retired man is a bit similar to non-gendered “golden ager” of Hummert et al. (1994). Depicted significantly less often, t (209)=4.34, p<0.001, later life masculinities are also evident in negative stereotypes—the curmudgeon (n=10), solitary old man (n=15), failed old man (n=9), and dirty old man (n=4). Although the curmudgeon stereotype was first identified by Schmidt and Boland, the content of the old man curmudgeon in the present study is only analogous to the trait cluster produced by the elder informants in Hummert et al. (1994), not the younger or middle-age informants in their study, nor the student informants in Schmidt and Boland. Similarly, when informants imaged a “solitary old man,” their descriptions seem comparable to the “despondent” stereotype Hummert (1990; Hummert et al., 1994) described, but informants’ sentence completions characterized the solitary man as socially alone, not emotionally lonely. On its face, the “failed old man” stereotype also appears similar to the “vulnerable” and “vagrant” stereotypes of the old person that younger informants described in Schmidt and Boland and Hummert et al., but again the content of the stereotypes markedly differ. Theirs emphasized a worried, fearful, impoverished person, whereas the current failed old man stereotype emphasizes the man’s failure as a (successful) man and/or his victimization as an old man. An equal number (n=178) of informants used descriptive statements, attributes, and activities that largely disregarded the gender of the old man and emphasized his age. What emerged as an unexpected age-related description was the association of an old man with death; in fact, 5% of 638 Sex Roles (2006) 55:633–648 Table 1 Content of the “Old Man” stereotypes. Positive later life masculinities The sage—wise, experienced Has wisdom Has experience Ready to give advice when asked Is wise/knowledgeable/worldly Listens intently and speaks when he has something important to say Knows more than young give credit for Should be well-respected/valued for his experience Fun to talk to The grandfather—nurturing, sociable Gentle/kind Considerate Sweet & harmless A (friendly) grandfather/reminds me of my grandfather(s) Bought/gave me candy Watches children play Enjoyed the company of his grandchildren Many stories to tell Sitting with a little boy next to him, listening intently The active retired man—engaged, busy Enjoys life because of available time/more free time Retires/receives Social Security Walks/walked around neighborhood for exercise Walked down/across the street/lane Supervising the job (others are doing) Still prefers to drive, not be driven Drives an old car Drove/went to the store Plays bocce/tends to his garden Fishes (with grandson/little boy) Men growing old—neutral images of aging The tranquil old man—at rest, philosophical Has live a long life/ was once young Spends time around the house Sits on porch Sitting in rocking chair Sits smoking a pipe Sat on park bench reading/feeding birds/talking Relaxed on bench in park Sat next to me in church Sat home and watched television Reflects on life/years past The timeworn man—aging appearances Been though a lot & that is shown in his face Wrinkled Gray/white hair Balding Wearing hat Brown age spots Old men—negative images of his aging The diminished man—poorer well-being Walks/shuffles slowly Is bent over Drives slow Hearing/vision losses Uses/caries a cane Less physically active/athletic Physically weak/weaker Withered Tired/less energy Must listen carefully May or may not be in good health and state of mind The impaired man—restricted, curtailed Suffering from chronic illness Will not get much older Cannot work Feels a loss of power Helpless Feeble Fragile Senile/crazy Dependent Driveling Considered as not responsible The mortal old man Died/is dead Negative later life masculinities The curmudgeon—ill-tempered (Sometimes) grumpy Cranky Stubborn in his ways More settled and set in ways Sits around the house very restless Longed for the days of his youth Complaining The solitary old man—dispirited, alone Feels rejects Discontented Suffers from loneliness Lonely man without wife and family His time has past & with that his family (Sometimes) lonely Needs companionship Lives by himself Sat alone in corner of his room Lives in a retirement home The failed old man—singular Slept on a park bench Slumped on a park bench with a bottle Standing on the corner beneath the street light On the street came up and ask me for spare change Got mugged by four teenagers Came into shop asking for a cup of coffee Suffers from many problems given to him by society The dirty old man Touched my girlfriend Married a younger woman Longed for sex Sex Roles (2006) 55:633–648 the informants finished the “An old man...” sentence stem with “is dead” or “died” (n=13). The rest of the age characterizations ranged from negative ageist stereotypes to neutral descriptions of how old men may appear—as timeworn and graying (n=32), tranquil (n=65), diminished and coping with lesser capabilities (n=34), and restricted and impaired (n=34). These characterizations are equivalent to the stereotypes reported by Schmidt and Boland (1986) and Hummert (1990, Hummert et al., 1994) for an old person whose gender was not specified. For example, the content of the “timeworn man” is very similar to the list of age-related physical traits reported in Schmidt and Boland’s study; the content of the “diminished man” stereotype is comparable to the content of the “mildly impaired” old person that the middle-age and elder informants provided in the studies by Hummert et al.; and, the “impaired man” is analogous to the “severely impaired” older person in both Schmidt and Boland and Hummert et al. Reviewing Table 1, it is evident that a considerable proportion of the descriptions of the old man (41%) call attention to the age-of-the-target. Two meta-themes and five of the 12 stereotypes were strictly age-specific, as when an informant’s narrative emphasized the experiences of growing older (e.g., “...been through a lot and that is shown in his face”). Nearly all of these age-related descriptive statements, attributes, and activities that make up the agespecific stereotypes were not gendered and could have defined an old woman as much as an old man (e.g., drives slow, reflects on life, is bent over). Nonetheless, the old man target was more frequently (59%) described in terms that recognize his later life masculinity—mostly in the context of old men’s social roles as a grandfather and former breadwinner, and to a lesser extent the way his social isolation or his temperament as a curmudgeon is noted. As a grandfather and/or former breadwinner, the old man’s nurturing, sociable character and active lifestyle were defined by his story telling and side-by-side masculine companionship activities with boys and other older men. As a sage, a concept that dictionaries classify as masculine, the content of the stereotype underscores the long standing social construction of the old man as venerated and able to advise, not as someone who offers unsolicited advice. These positive characterizations of the old man outnumbered the negative. Even for the negative masculinity stereotypes, the descriptors disclosed traces of agency and interpersonal involvement more often than just an emotional state. For example, the old man “feels rejected” or “suffers from loneliness” rather than “feels alone,” or the old man “is discontented” rather than “sad” or “melancholy,” suggests something about the man’s social life as much as his emotional state or well-being. When informants were asked to describe the old man compared to a young man in the second sentence 639 completion, their characterizations again produced a pair of themes about later life masculinities and about aging. Not unexpectedly the same-sex/different-age comparison enhanced the salience of the target’s age; very often informants described the old man in terms of having “more” or “less” of something than a young man. A total of 13 stereotypes (Table 2) were disclosed. Eight of the stereotypes were age-related, and their content comprised 64% of informants’ descriptive statements, a marked increase over the age-related imagery disclosed for the unanchored old man target. It is noteworthy that one-third of these characterizations suggested a positive imaging of aging and centered on three new stereotypes: the seasoned, mature man (n=40), the successful, accomplished man (n= 25), and the principled gentleman (n=21). In each of these stereotypes the descriptive statements about the old man highlighted ways that growing older can yield integrity and classiness, compared to a young man. Consistent with the ageism in American culture, the number of informants using negative descriptors was greater than the number who mentioned positive images about men’s aging, n=126 and 86, respectively, χ2 =7.55, p<0.01. Many of these negative, age-specific characterizations were also explicitly or subtly gendered, such as when informants produced the distinct, gendered negative stereotype about men becoming old—the old man was characterized as a former beast in bed (n=17). Similarly informants emphasized the old man’s diminished status in society more than his pitiful nature when they described him as vulnerable. Informants reestablished the images of an old man as diminished and coping with lesser capabilities (n=76), timeworn and graying (n=18), and, less often, as restricted and impaired (n=7). When compared to a young man, these images of the old man no longer contained neutral characterizations. The timeworn man was more likely defined as “worse looking” and “less attractive” rather than described plainly in terms of his age-related wrinkles and gray hair; the prevalent description of a tranquil old man noted in the initial descriptions was replaced by an emphasis on how aging likely diminishes a man’s sexual, physical, and psychological capabilities and how men are believed to lose agency with age. In all, the negative characterizations accentuated the physiological aspects of growing older and the imagined effects of the “rolelessness” that some young adults think men might experience in later life. Anchoring the old man to a young man also partly reframed the way informants perceived later life masculinities. Altogether, the number of informants imaging later life masculinity positively was far greater than the number of informants defining it negatively (n = 74 and 47, respectively), yet the number of discrete negative depictions 640 Sex Roles (2006) 55:633–648 Table 2 Content of the “Old Man” stereotypes, compared to “Young Man.” Positive later life masculinities The sage—wise, experienced More knowledge/more wise about life (or ways of the world) Is wise/has great wisdom/acts wisely More experienced Possesses air of worldliness/more worldly A lot of character/by far more interesting Greater foresight More intellectually mature Shares knowledge with young man, probably to keep the men dominant The Companion—sociable, considerate More interested in you/more sincere Much nicer/easier to talk with More friendly/kindly More understanding More likely to tell you things about himself More jovial/nicer personality Old men—positive images of his aging The mature man—seasoned, tempered Becomes mellow/more at ease (to experience) More mature and self-assured Acts as he feels and speaks what he thinks Sexier/more experienced in bed More distinguished More content, relax, at ease More calm, stable, controlled & settled down Less macho, competitive Less threatening Less reckless, impulsive The gentleman—conventional, principled More traditional, conservative, conventional More settled Better manners/a sense of chivalry Worries more about (or more concerned with) his family Still married to an old woman The accomplished old man—successful, retired More authoritative power Doesn’t have to impress or prove himself to anyone More independent Less pretentious Has established himself in working work Less concerned with monetary success More financially secure/wealthier No longer earns much money Works less/has fewer family responsibilities Old men—negative images of his aging The former beast in bed—lost sexuality Much less a sex object Fewer sexual desires, interests, fantasies Less sexually active More trouble in the sack/can get it up as long Must fantasize about women more The timeworn man—aging appearances Less appealing/attractive Looks different, uglier Worse looking Fatter Grayer hair Balding Wrinkles The diminished man—his poorer well-being Not as quick, slower and more careful Slowed down and has less energy Not as capable or happy/more pessimistic Appears less outgoing Doesn’t have as much fun Less enthusiastic More physically restricted Less athletic, physically active/fit Tired, less vitality/stamina Less agile, less able Skinner, less muscle Physically and mentally weaker Not as sharp or alert Poorer sight More unhealthy/has more health problems The impaired man—restricted, curtailed Helpless Senile Dependent, not self-reliant Isn’t capable of taking care of himself Should be cared for Has a minimal existence Likely to die sooner than later The vulnerable old man—alone, mistreated Fewer life/job opportunities Has a minimal existence Sad figure in society More depressed about place in society Doesn’t have it/doesn’t stand a chance Should be cared for Negative later life masculinities The curmudgeon—ill-tempered Less calm More demanding, stubborn more grouchy, cranky Crabbier Grumpier, less friendly Tougher time accepting act of aging Isn’t as much fun to be with Frustrated that he is not young The uncompromising old man—inflexible, self-centered More set in ways (values/habits) Less apt to adopt new methods of living Not as open minded Not as adventurous Thinks he is always wiser/knows more about everything Little alternative than to reminisce in past glories The solitary old man—dispirited, alone Feels less useful and productive Pushed into retirement Retirement yields helplessness More withdrawn/less exciting Lacks incentive/ambition Sex Roles (2006) 55:633–648 Table 2 continued. Less determined Lonely and isolated Stripped of security Considered less productive Perceived as less important/influential of later life masculinities was greater than the positive behaviors and attributes mentioned. The old man again was described in terms of the sage stereotype and its emphasis on how aging yields wisdom (n = 62). He also was stereotyped as an affable companion and someone who is interpersonally engaging (n=12). This companion image seems to be a variant of the stereotype of an old man as grandfatherly. Informants’ descriptions also maintained the negative stereotypes of later life masculinity that characterized an old man as alone and dispirited (n=19) and a curmudgeon (n=14). Their descriptive imagery added a stereotype of an uncompromising, inflexible old man (n= 14), which might be a blending of the “inflexible senior citizen” and “self-centered” images Hummert (1990) identified for “older adults.” In the final sentence completion, informants’ characterizations of an old man compared to an old woman yielded slightly fewer descriptive statements, yet 13 images of an old man emerged from the data and revealed the same meta-themes of how positive and negative images of later life masculinities and aging underpin informants’ comments (see Table 3). Compared to an old woman, nearly 25% of the informants (n=46) made some reference to the old man’s shorter life span. Otherwise, the aging double standard evident in American culture cast a positive halo on the way an old man is described in comparison to an old woman. Thus unlike the less favorable imagery evident in the descriptions of the old man when compared to a young man, where only 36% of the content mentioned something positive about an old man, description of the old man compared to an old woman yielded a sizeable proportion of the positive descriptive statements (56%), and the number of informants reporting a positive characteristic exceeded the number that mentioned a negative, n=129 and 114, respectively. Further, if the one set of descriptive statements that mentioned men’s shorter life span was reclassified as a description of a true sex difference rather than a negative image of older men, then the number of informants reporting a positive description of the old man would far exceed the number that mentioned a negative, n=129 and 68. Embedded within the imagery are five positive stereotypes about later life masculinities and two positive stereotypes of men’s aging. Not surprisingly nearly all of the content of the seven stereotypes unmistakably acknowl- 641 edges the gender-of-the-target. Later life masculinities were described positively in terms of an old man being a sage (n=7), an amiable companion (n=19), and an active, retired man (n=14). This is at least the second time these three masculinity stereotypes were evident within informants’ sentence completions. Two new later life masculinity stereotypes were introduced—the broadminded, philosophical man (n=7) who is perceived as more tolerant and accepting than an old woman, and descriptions of an old man that emphasized his continuing manliness, or being “still a man” in character and lifestyle (n=22). The two images that characterized an old man in positive aging terms centered on the stereotype of a seasoned, mature man who is perceived as more respected and aging gracefully than an old woman (n=17) and a new stereotype of the old man as handsome (n=43). As formidable as the number of people positively stereotyping the old man when compared to an old woman, almost half of the unique content (46%) is again characterized in negative terms. Two underlying images continued to characterize the old man’s later life masculinities in terms of him being a curmudgeon (n=36) and a solitary, dispirited man (n=15), and three stigmatizing images of his aging again defined an old man in terms of his timeworn appearance (n=22), diminished agency (n=37), and impaired healthy and lifestyle (n=8). Discussion Old men can be talked about and imaged through conversations, as well as sentence completions. The range of distinct images detected in this study strongly suggests that people are, in fact, not of one mind when they think about an old man. Images of the old man comprised positive and negative depictions of aging as well as later life masculinities. Most of the time informants’ characterizations of the old man included multiple later life masculinities. Often enough the content within an individual’s sentence completion included contradictory descriptors (e.g., “...is valued for his experience yet he is at times rejected because he is old”). As important the findings are in beginning to describe the ways old men are perceived, the findings are restricted. On one hand, the “old man” target’s ethnic/cultural background, health and employment statuses, living arrangements, and so on were not identified, and surely further information about the target could have affected the way the old man was discussed. Whether or not the images and stereotypes are dependent on the respondent’s gender is also something that needs to be investigated. This study consciously did not assess if young men defined the old man differently than young women. The intent of the study was to provide an initial description of a gendered, older 642 Sex Roles (2006) 55:633–648 Table 3 Content of the “Old Man” stereotypes, compared to an “Old Woman.” Positive later life masculinities The sage—wise, experienced More experience in life More wise/ more aware/worldly More sexually competent The companion—sociable, considerate More fun to be around, less synthetic More lively Has more fun/enjoys life more/full of life More freedom Better sense of humor More compassionate Softer Easier to get along with/talk more Friendlier/likely to talk to you The broadminded old man—philosophical More tolerant Accepts old age better/more at ease with his age Gets older more gracefully Accepts younger generations more readily Less concerned about death/about dying Greater pride in regard to his age The active retired man—engaged, busy Spends less time in the home Hands out more with his buddies More active/ plays more golf Likes to walk Likes younger women more than an old woman likes younger men Thinks he stays young at heart longer Likes to read/ talk (Still) The man—conventional/unfeminine lifestyle Can be a man’s man Thinks about sex more Tries to prove himself sometimes Cooks less often/doesn’t shop or gossip Not perceive to be nearly as weak Less vulnerable/ less dependent More self-sufficient More adventurous Less involved with family worries, more leisure Still feels he has more financial responsibility for his family Old men—positive images of his aging The handsome old man—appearance, spunk Ages/looks better, looks younger Appears more distinguished Less concerned about his physical aging Ages more slowly Not fat and ugly More desirable/ more attractive Less wrinkles Doesn’t worry about facial wrinkles Younger acting Possesses more vigor/more vitality The Mature man—seasoned, tempered More respected More established/financially secure More independent Less crabby More easygoing/laid back Old men—negative images of his aging The timeworn man—aging appearances Often looks older/uglier Shows the effects of the tension/stress that was his life Balding/less hair More unkempt More wrinkles Isn’t expect to/can’t hide his age as easily The diminished man—his poorer well-being Less self-sufficient Wants more done for him/needs more care and attention Less interested/ curious More sickly/not as healthy Slower/ less active/ lazier More depressed with old age More unhappy/less happy with getting older The impaired man—restricted, curtailed Wears hearing aid Older, usually more senile Acts (as if) older The mortal old man—compressed lifespan Dies earlier/sooner Dies younger than the old woman Shorter life/does not live as long Enjoys a briefer life span Negative later life masculinities The curmudgeon—ill-tempered Not as pleasant Not as much fun to be around More stubborn, more (or very) demanding More grouchy, cranky Grumpier, fussier, set in ways Not adaptive to times, more conservative More restless/edgy Harder to take care of More negative complaining Bad disposition Much more particular The solitary old man—dispirited, alone Drinks more More lonely Often lonely with his wife Has a hard time with getting old Sat on a park bench Sits around/does nothing Sleeps more Watches more television Sex Roles (2006) 55:633–648 adult target; the number of participants was too small to reliably assess, in a post hoc study, whether or not a gender difference exists in the way the young adults characterize the old man target. Furthermore, studies need to go beyond comparing men’s and women’s perceptions, and investigate within-gender differences, determining whether men who endorse particular masculinity values (e.g., self-reliance) report more or less favorable images of old men. Some of the images of the old man seem to be widely held by the participants in this study, such as representations of an old man as an active, yet retired man, a sage, and a curmudgeon. Other representations were less frequently mentioned and are reported after the salience of the old man’s gender and age was primed with a contrasting target, such as the old man being depicted as an “accomplished man” and a “gentleman” when compared to a younger man or as “handsome” and “(still) a man” when compared to an old woman. Just six of the 22 representations were found in all three descriptions—the sage, the timeworn man, the diminished man, the impaired man, the curmudgeon, and the solitary man. It is noteworthy that all but one of the “common” images are unfavorable; two call to mind negative late life masculinities and three evoked negative views of aging. Ageist images of later life masculinities—from being a dirty old man to a dispirited, solitary old man to a curmudgeon—are representations presented frequently in the mass media and literature, from Jack Nicholson in As Good as it Gets to the Grumpy Old Men films to Hemmingway’s Old Man at Sea. Finding the old man was at times favorably visualized as grandfatherly, an affable companion, and handsome is quite consistent with the way older men are portrayed in film vis-à-vis older women (Markson & Taylor, 2000). Thus, uncovering ageist (and sexist) representations is not surprising, given the pervasiveness and persistence of ageist imagery of men and women in our culture (Chasteen, Schwarz, & Park, 2002) and the evaluatively-mixed nature of ageist stereotypes (Cuddy, Norton, & Fiske, 2005)—e.g., old men equally described as both a sage and as incompetent. Principally, the images summarized in Fig. 1 and the content of detailed in Tables 1, 2, 3 attest to the continued importance of the target’s gender. Unlike the gender-neutral findings reported by Schmidt and Boland (1986) and affirmed by Hummert at al. (1990, 1994), even the findings from the initial sentence completions make clear that without priming informants’ sensitivity to the old man’s gender with a comparison to the old woman, later life masculinities were disclosed. A greater number of the underlying images of an old man addressed a dimension of later life masculinity, sometimes positively and other times negatively, and less than half (41%) of the content describing the old man was strictly age-specific. 643 When compared to a young man and an old woman, other distinctive later life masculinities emerged. For example, the image of an old man as a “former beast in bed” unquestionably brings to the forefront the old man’s age and the effects of aging on his sexuality. By addressing his sexuality the informants were addressing a central dimension of masculinity (Connell, 1995), one that has gained greater public recognition with the recent pharmaceutical advertisements for erectile dysfunction (Calasanti & King, 2005). Similarly, the positive images of how aging could affect an old man—by being a “gentleman” and an “accomplished old man”—certainly contain evidence of later life masculinities and the continuities of expectations for men to be “successful” in whatever they do as men. Study 2 As much as the prior study reveals evidence of distinct later life masculinities within discourses on old men, the extent to which old men’s perceived masculinity differs from younger men was not investigated. Since old men are talked about and imaged in terms of masculine forms such of “the gentleman” and “the sage,” what is the likelihood that their perceived masculinity is diminished by their aging? Does one type of gender performance—being a grandfather, a sage, an active, retired man—replace earlier masculinity performances—being a breadwinner, athlete—and maintain an aura of respectability and masculinity? Is the old man still perceived as masculine, and any less so than a younger man? A second study was designed to extend the initial query on whether an old man is thought to be masculine. It is probable that an old man is perceived as less masculine because he is imaged as less active and his appearance timeworn; in the prior study he was often portrayed by the unflattering depiction of a “diminished man.” His so-called diminished masculinity may be especially salient when he is paired with a young man. Does the context within which an old man target is evaluated—when he is alone versus with a younger man versus with a same aged woman—have bearing on his evaluation? Smiler (2004) has noted that investigations of how stereotyping is contextually specific is an underdeveloped area of masculinity research. Method Participants Two hundred fifty six undergraduates participated in a questionnaire-based study of “men’s and women’s beliefs about other people.” Participants were recruited early in the semester from classes and asked to take a short questionnaire with them upon leaving class. Three hundred questionnaires were handed out. The completed questionnaires were returned through campus 644 mail within the week, yielding an 85% response rate. Participants ranged in age from 18 to 26 (mean=19.6, SD= 1.3). Women (66%) outnumbered men two to one, and very nearly all students were white (97%). Their father’s age ranged from 39 to 72 (mean=49.5, SD=5.6); 54% of the students’ grandfathers were alive, and all but one student had interacted with a grandfather in the past year. Questionnaires Three versions of the questionnaire were used, yet participants only filled out one version. One was designed to have participants report their views about “an old man,” another asked the participants to describe an old man compared to “an old woman,” and the other asked the participants to describe an old man in comparison to a “young man.” Each participant rated the old man target either without a comparison other or in comparison to an old woman or a young man. The number of completed questionnaires for each version was 86, 82, and 88, respectively. Measures Participants were initially instructed to describe the old man using a list of personality attributes (the Bem Sex Role Inventory, BSRI; Bem, 1982). They revealed their image of the old man for the 30 attributes in the short form of the BSRI by indicating on a 7-point scale (ranging from “never or almost never true” to “always or almost always true”) the extent to which they saw the attribute describing an old man. Among the listed cultural traits are ten socially desirable, instrumental attributes (e.g., assertive, independent) that have been used in many gender studies to describe “masculinity,” ten other attributes (e.g., sensitive to others, compassionate) assessing traits more commonly associated with women and “femininity,” and ten nongendered attributes (e.g., tactful, jealous) that are ignored. To construct summary measures of the instrumental/ masculinity traits and expressive/femininity traits, ratings on each set of ten attributes were algebraically summed. Reliability estimates for the “masculinity” and “femininity” trait measures reveal that they are internally consistent for each version of the questionnaire (αs range from 0.80 to 0.85 for “masculinity” and 0.89 to 0.93 for “femininity”). For further discussion of BSRI interpretation, see Bem (1981) and Spence, Losoff, and Robbins (1991). A modified from of the Osgood Semantic Differential developed by Levin (1988) was also used to assess participants’ descriptions of an old man. Numerous studies have established the validity of the semantic differential as a procedure for measuring meaning, and the method has proven useful for assessing differences in attitudes and conceptual understanding when perceptual targets are compared (cf., Levin, 1988). By assessing attitudes toward the target with opposing, or bi-polar, adjectives, the semantic differential taps the connotative meaning participants assigned to the three old man targets. Sex Roles (2006) 55:633–648 The semantic differential consisted of 24 bipolar pairs of adjectives, each rated on a seven-point scale. The items were designed to measure attitudes about older persons’ level of instrumentality, competence, acceptability, and integrity (cf., Intriri, von Eye, & Kelly, 1995). A factor analysis (varimax orthogonal rotation) of the respondents’ ratings of the old man targets produced a two factor solution. Numerous labels have been used by researchers to characterize these two dimensions (e.g., task orientation vs. social orientation), yet the closest to match to the actual content of the factors are the terms instrumentality and acceptability. The first scale evaluates the target’s instrumental/ autonomous character (α=0.80). It is an algebraic mean of ratings on the 11 paired attributes that loaded highly on the factor: active–passive, incompetent–competent, powerful–weak, healthy–sickly, confident–scared, creative–uncreative, slow–fast, energetic–lazy, wealthy–poor, educated– uneducated, and lonely–supported. The item pair flexible– rigid was discarded. The second scale examines an interpersonal side of the target’s perceived character (α=0.82) with 12 adjective pairs: calm–irritable, attractive–ugly, unpleasant–pleasant, unreliable–reliable, important–unimportant, depressed–excited, involved–socially isolated, poor memory–good memory, generous–selfish, meaningless–meaningful, awkward–graceful, and loving–angry. For purposes of scale construction, the item pairs were all coded for a higher score to reflect greater endorsement of the positive trait. Analysis Strategy The research design involved three distinct targets—two old man targets that explicitly involved comparison others, a young man and an old woman, and one unanchored target. The means for the three sets of evaluations of the BSRI traits and the semantic differential ratings were compared using multivariant analyses of covariance. To guard against the ways that routine involvement with older men could differentially affect some respondents’ evaluation of a target, the participant’s father’s age and a dummy variable for whether or not the respondent was interacting with a grandfather were entered first as covariates. Results The two covariates explained little variance among respondents’ assessments. Respondents with older fathers were less likely to evaluate older men in instrumental/ masculine terms, F (1,246)=4.01, p<0.05. One significant main effect for the sex of the participants, F (1,246)=4.82, p < 0.05, was also observed for the way old men’s femininity was assessed. Women generally saw greater expressiveness/ femininity in the old man target than men. Sex Roles (2006) 55:633–648 645 Table 4 Means for masculinity and femininity, instrumentality and acceptability ratings for three old man stimulus targets. Measure Old man Old man/ young man Old man/ old woman F (2,246) BSRI M scale BSRI F scale Instrumentality Acceptability 4.47a 4.72b 4.46c 4.94 4.16 4.83b 4.07 4.60d 4.76a 4.14 4.62c 4.64d 12.02*** 14.52*** 12.02*** 6.39*** Row means sharing a superscript do not significantly differ from one another at p<0.05 (one-tailed); all other paired comparisons differ. ***p<0.001. Neither respondents’ sex nor the covariates had any bearing on the semantic differential assessments of instrumentality and acceptability. The MANCOVA examining the three assessments of the old man’s perceived gender was significant, Wilks=0.834, F (4,476)=11.30, p<0.001. As expected, the old man target was differentially described in the three questionnaires in terms of the “masculinity” scale, F (2,239)=10.58, p< 0.001, and “femininity” scale, F (2,239)=14.49, p<0.001. Table 4 summarizes the group means and pairwise multiple comparisons (Bonferroni) for the three old man targets. Evaluations of the old man differ markedly when he was judged in situations that highlighted either his age or gender Fig. 2 Evaluations of old men’s perceived gender. (see Fig. 2). When gender and age were not explicitly cued with a comparison target, the old man was perceived to have greater feminine qualities than masculine. Similarly, when attention was called to his age, respondents again rated the feminine qualities as more characteristic of an old man than masculine qualities. Not surprisingly, when compared to an old woman he was perceived to be least feminine and more masculine. Ratings on the semantic differential scales were significantly swayed by the old man target evaluated, Wilks= 0.825, F (4,478)=12.04, p<0.001. His instrumentality was assessed in patterns similar to evaluation of his trait masculinity (see Table 4 and Fig. 2) and thus deemphasized when the target was compared to a young man and accented when he was compared to an old women, F=13.67 (2,240), p<0.001. The old man’s acceptability was, however, uniformly diminished when he was compared to another person, which strongly suggests that an old man’s acceptability is affected by both age and gender norms, F=6.36 (2,240), p<0.01. Age norms make the old man less acceptable than a young man, and gender norms make him less tolerable than an old woman. Discussion The differential assessment of three old man targets demonstrates that age and gender norms both underlie 646 young adults’ images of elderly men’s gendered lives. Not unexpectedly, the old man was perceived as significantly less instrumental/masculine when compared to a younger man. Whether assessed with the BSRI or the semantic differential scales, the instrumental attributes associated with manhood in our culture—men being forceful, selfreliance, self-assured, strong and robust—were rated as less characteristic of an old man. His perceived masculinity was diminished when he was compared to a young man. Viewing masculinity through the lens of age, the old man does not measure up to the hegemonic standards of masculinity typically associated with younger men’s lives (Connell, 1995). But should he? Ironically, the BSRI “masculinity” mean for the old man in the unanchored target is congruent with the normative data Bem (1982) reported for adult men’s self-evaluations. It was when the target’s age was intentionally highlighted through direct comparison to a young man that the respondents diminished his masculine character. If these evaluations could be generalized to truly reflect young adults’ beliefs about late life masculinities, as a man ages he is no longer expected to evidence the characteristics associated with younger men and manhood. Yet he is still perceived as a man, especially when compared to an older woman, when his instrumental qualities and masculine attributes are heavily accented and his feminine side becomes, comparatively speaking, much less visible. Perhaps the antifemininity dimension inherent in the standards of manhood (Thompson & Pleck, 1995) swayed respondents’ evaluation to deemphasize the old man’s perceived femininity when attention was drawn to his gender by asking informants to compare him to an old woman. Interestingly, without an explicit comparison target, an old man’s masculine character is not devalued. The results uphold an argument that gender norms do not fade away as men age as well as an argument that age prejudices and gender stereotypes cast old men as both old and men. Interestingly, the gender of participants had minimal effect on perceptions; with the exception that women attributed greater “femininity” to the old man than the men did, college-age men and women imaged the old man similarly. General Discussion and Conclusions This article examined whether or not men’s later life likely remains scripted by masculinity expectations and, more directly, what types of age-based gender stereotyping is ordinary for men in later life. According to the seminal work of Neugarten (Neugarten, Moore, & Lowe, 1965), all of us sustain an internal timetable about age-appropriate behavior by which we measure our own and others’ march across the life course. Age norms define the appropriateness or inappropriateness of self-presentation, expressed emo- Sex Roles (2006) 55:633–648 tions, and other specific behaviors. Gender is also salient in our judgments of elders’ age-appropriate behavior (cf., Plath & Ikeda, 1975). The two studies in this article reveal that gender and age are both perceptually salient when old men are thought of. People seem to hold clear stereotypic beliefs that differentiated older and younger men, including beliefs that highlight old men’s masculinities as well as personal attributes. As Hearn (1995, p. 99) proposed, men and masculinities “are talked about, pictured, imaged. To do so invokes age and agedness. Men and masculinities, whether in image, text, social practice, or social structure, are necessarily about aging, sometimes explicitly so, often implicitly. There is a recurring inseparability of ‘men’ and age/aging.” Further, this research underscores how important gender is an inherent part of later life. Older men are perceived as men, more so than simply as an elder. This finding is instructive theoretically, because it is inconsistent with the gender convergence assertion that in later life age status outweighs gender and, by implication, an androgynous standard defines late life (Gutmann, 1987; Sinnott, 1986). The convergence hypothesis was built on clinical-based opinion and research showing older men were less masculine/more feminine than young men when assessed with M-F scales. As evidenced by the descriptions of the old man compared to an old woman, aging does not seem to result in the homogenizing of gender expectations. This pattern of differential evaluation also lends empirical support to the argument that there are multiple masculinities embedded in American culture, as has been theorized by men’s studies scholars for the past decade (cf., Connell, 1995; Thompson & Whearty, 2004). Researchers have persuasively demonstrated that men with physical disabilities are viewed and treated from a different masculinity standard than the ones applicable to men without disabilities (Gerschick & Miller, 1994), and that masculinities vary by class and race (cf., Duneier, 1992; Turner & Turner, 1991). Logically, masculinities vary by age as well, and the descriptions of old men observed in this research strongly suggest the value of systematically outlining the blueprints of masculinities that different old men encounter in their everyday lives. As Smiler (2004) noted, the relevance of masculinity theory is to reexamine another construct, in this case “old,” through the lens of masculinity and ask how manhood and gender influence being old. Because the late life masculinity stereotypes of old men seem to rest on different gender standards than applied to men at earlier stages in the life course, much more empirical research is needed which outlines the common and unique dimensions of old men’s gender script (cf., Calasanti, 2004). Also affirmed is that our understanding of old men’s lives cannot be fully understood without an appreciation of Sex Roles (2006) 55:633–648 the multidimensionality of age-based gender expectations. Research has demonstrated that people maintain multiple stereotypes of the elderly rather than a single, global stereotype. This line of research principally focused on the superordinate category older adult (cf., Hummert et al., 1994; Schmidt & Boland, 1986. Adding gender distinctions, as the present study did with its assessments of the categories old man (and old woman), it is striking how many ways the old man target was perceived. One important limitation of the study is the likelihood that some of the stereotypes underneath the four metathemes (Tables 1, 2, 3) may be only characteristic of how young adults perceive old men. The research was restricted to undergraduates, and no effort was made to identify the ethnicity/racial background of the participants in this study. The study of Hummert et al. (1994) revealed that different age groups hold somewhat different images of a target. And a study of three generations describing themselves at ages 20, 45, and 70 found the self-descriptions by the men (and women) were structured by both age and gender norms (Puglisi, 1983). In the study, young men’s projections into later life anticipated that their personal characteristics would reflect rising masculinity to mid-life, followed by decreases in later life. Older men looking backward made finer life stage discriminations regarding gender traits and had richer, more complex cognitive schemas about gender. Similar studies by McCreary (1990), in which elders rated themselves at the present time and in two retrospective points (work-entry and parenthood) and younger respondents reported their gender attributes at the present time and in two prospective points (parenthood and retirement), effectively demonstrated that informants’ presentation of former, present, and future selves were sensitive to models of aging that acknowledge age-based masculinities (and femininities). The common denominator is that people visualize their own changing identities as well as others’ lives in terms of age-based masculinities and femininities and these need to be more thoroughly investigated with other ages than young adults. Being cautious in interpreting the strength of this research is necessary. The first study tacitly yields the warning that when participants are asked to describe an older man target without a comparison, the images of the old man almost involved as much age stereotyping as gender stereotyping. The phrase “old man” might have activated what Cole (1992) called bipolar ageism, which vacillates between discredited negative stereotypes of old age and positive constructions (cf., Cuddy et al., 2005; McHugh, 2003). Thus, informants’ images and the content they divulged about their images of an old man could be unduly guided by the label “old man” as well as the comparison labeling “young man” and “old woman.” Braithwaite et al. (1985–86) and more recently the meta- 647 analytical review of Kite et al. (2005) have shown that informants respond in more stereotypic manner when asked to assess a general versus specific target. Given that the present research was based on evaluations of a general target word “old man,” it is possible that students responded to the label in an exaggerated fashion. Whether such exaggeration mars and/or masks their underlying beliefs about and expectations of older men is also uncharted. A second caution is linked to the findings of the second study. I have interpreted the findings as affirmation of the joint effects of age and gender expectations. Nonetheless, the lack of a reliable difference in the one paired evaluation of the old man and old woman on the semantic differential acceptability scale could be evidence for the premise of Kite et al. (1991) that age scripts and stereotyping outweigh those of gender in late life. Nonetheless, to think of old men as less like younger men and more similar to old women seems to introduce a deficit model of how aging diminishes men’s vitality (cf., Vaillant, 2002). It also suggests that there is only one standard of masculinity in American culture. Surely in old age the social worlds of men and women might be more similar than they were earlier, but older men and women bring into later life a gendered biography. This study has shown that people also seem to anticipate distinct gendered lives in late life. Standards of masculinity may not remain stable across the life-span, but neither do they disappear. Old men are still men after all, and “being a man” is an expectation that extends into old age. References Arber, S., Davidson, K., & Ginn, J. (2003). Gender and ageing: Changing roles and relationships. Maidenhead, England: Open University Press. Bem, S. L. (1981). Gender schema theory: A cognitive account of sextyping. Psychological Review, 88, 354–364. Bem, S. L. (1982). Bem sex-role inventory: Professional manual. Palo Alto, CA: Consulting Psychologists Press. Bleiszner, R. (1989). Developmental processes of friendship. In R. G. Adams & R. Bleiszner (Eds.), Old adult friendship (pp. 108– 126). Newbury Park, CA: Sage. Braithwaite, V. A. (1986). Old age stereotypes: Reconciling contradictions. Journal of Gerontology, 41, 353–360. Braithwaite, V. A., Gibson, D., & Holman, J. (1985–86). Age stereotyping: Are we oversimplifying the phenomenon? International Journal of Aging and Human Development, 22, 315–325. Brewer, M. B., Dull, V., & Lui, L. (1981). Perceptions of the elderly: Stereotypes as prototypes. Journal of Personality and Social Psychology, 41, 656–670. Brewer, M. B., & Lui, L. (1984). Categorization of the elderly by the elderly: Effects of perceiver’s category membership. Personality and Social Psychology Bulletin, 10, 585–595. Butler, R. N. (1969). Age-ism: Another form of bigotry. The Gerontologist, 9, 243–246. 648 Calasanti, T. M. (2004). Feminist gerontology and old men. Journal of Gerontology: Social Sciences, 59B, S305–S314. Calasanti, T. M., & King, N. (2005). Firming the floppy penis: Age, class, and gender relations in the lives of old men. Men and Masculinities, 8, 3–23. Canetto, S. S., Kaminski, P. L., & Felicio, D. M. (1995). Typical and optimal aging in women and men: Is there a double standard? International Journal of Aging and Human Development, 40, 187–207. Chasteen, A. L., Schwarz, N., & Park, D. C. (2002). The activation of aging stereotypes in younger and older adults. Journal of Gerontology: Psychological Sciences, 57B, P540–P547. Cole, T. R. (1992). The journey of life: A cultural history of aging in America. Cambridge, UK: Cambridge University Press. Connell, R. W. (1995). Masculinities. Berkeley, CA: University of California Press. Connell, R. W. (2000). The men and the boys. Berkeley, CA: University of California Press. Cuddy, A. J. C., Norton, M., & Fiske, S. T. (2005). This old stereotype: The pervasiveness and persistence of the elderly stereotype. Journal of Social Issues, 61, 267–285. Deutsch, F. M., Zalenski, C. M., & Clark, M. E. (1986). Is there a double standard of aging? Journal of Applied Social Psychology, 16, 771–785. Duneier, M. (1992). Slim’s table: Race, respectability, and masculinity. Chicago, IL: University of Chicago Press. Ekerdt, D. J. (1986). The busy ethic: Moral continuity between work and retirement. The Gerontologist, 26, 239–244. Emslie, C., Hunt, K., & O’Brien, R. (2004). Masculinities in older men: A qualitative study in the west of Scotland. Journal of Men’s Studies, 12, 207–226. Erikson, E., Kivnick, H. Q., & Erikson, J. M. (1994). Vital involvement in old age: The experience of old age in our time. New York: Norton. Gerschick, T. J., & Miller, A. S. (1994). Gender identities at the crossroads of masculinity and physical disability. Masculinities, 2, 34–55. Green, B. S. (1993). Gerontology and the construction of old age: A study in discourse analysis. Hawthorne, NY: Aldine de Gruyter. Gutmann, D. (1987). Reclaimed powers: Toward a new psychology of men and women in late life. New York: Basic. Hearn, J. (1995). Imaging the aging of men. In M. Featherstone & A. Wernick (Eds.), Images of aging: Cultural representations of later life (pp. 97–115). London, UK: Routledge. Hummert, M. L. (1990). Multiple stereotypes of elderly and young adults: A comparison of structure and evaluation. Psychology and Aging, 5, 182–193. Hummert, M. L., Garstka, T. A., & Shaner, J. L. (1997). Stereotyping of older adults: The role of target facial cues and perceived characteristics. Psychology and Aging, 12, 107–114. Hummert, M. L., Garstka, T. A., Shaner, J. L., & Strahm, S. (1994). Stereotypes of the elderly held by young, middle-age, and elderly adults. Journal of Gerontology: Psychological Sciences, 49, P240–P249. Intriri, R. C., von Eye, A., & Kelly, J. A. (1995). The aging semantic differential: A confirmatory factor analysis. The Gerontologist, 35, 616–621. Jackson, L. A., & Sullivan, L. A. (1988). Age stereotype disconfirming information and evaluations of old people. Journal of Social Psychology, 128, 721–729. Kidder, T. (1993). Old friends. Boston, MA: Houghton Mifflin. Kimmel, M. S. (1996). Manhood in America: A cultural history. New York: Free. Sex Roles (2006) 55:633–648 Kite, M. E., Deaux, K., & Miele, M. (1991). Stereotypes of young and old: Does age outweigh gender. Psychology and Aging, 6, 19–27. Kite, M. E., Stockdale, G. D., Whitley, B. E., & Johnson, B. T. (2005). Attitudes toward younger and older adults: An updated metaanalytic review. Journal of Social Issues, 61, 241–266. Kosberg, J., & Kaye, L. (1997). Elderly men: Special problems and professional challenges. Berlin, Heidelberg, New York: Springer. Levin, W. C. (1988). Age stereotyping: College student evaluations. Research on Aging, 10, 134–148. Markson, E. W., & Taylor, C. A. (2000). The mirror has two faces. Ageing and Society, 20, 137–160. Martin, P. Y. (2004). Gender as social institution. Social Forces, 84, 1249–1273. McCreary, D. R. (1990). Self-perceptions of life-span gender-role development. International Journal of Aging and Human Development, 31, 135–146. McHugh, K. E. (2003). Three faces of ageism: Society, image, and place. Ageing and Society, 23, 165–185. Minnigerode, F. A., & Lee, J. A. (1978). Young adults’ perceptions of social sex roles across the life span. Sex Roles, 4, 563–569. Morgan, D. H. (1992). Discovering men: Sociology and masculinities. New York: Routledge. Neugarten, B. L., Moore, J. W., & Lowe, J. C. (1965). Age norms, age constraints, and adult socialization. American Journal of Sociology, 70, 710–717. Plath, D., & Ikeda, K. (1975). After coming of age: Adult awareness of age norms. In T. Williams (Ed.), Socialization and communication in primary groups (pp. 107–123). Mouton, The Netherlands: The Hague. Puglisi, J. T. (1983). Self-perceived age changes in sex role selfconcept. International Journal of Aging and Human Development, 16, 183–191. Ryff, C. D. (1991). Possible selves in adulthood and old age: A tale of shifting horizons. Psychology and Aging, 6, 286–295. Schmidt, D. F., & Boland, S. M. (1986). Structure of perceptions of older adults: Evidence for multiple stereotypes. Psychology and Aging, 1, 255–260. Sinnott, J. D. (1986). Sex roles and aging: Theory and research from a systems perspective. Basel, Switzerland: Karger. Smiler, A. P. (2004). Thirty years after the discovery of gender: Psychological concepts and measures of masculinity. Sex Roles, 50, 15–26. Spence, J. T., Losoff, M., & Robbins, A. S. (1991). Sexually aggressive tactics in dating relationships: Personality and attitudinal correlates. Journal of Social and Clinical Psychology, 10, 289–304. Thompson, E. H. (1994). Older men’s lives. Thousand Oaks, CA: Sage. Thompson, E. H., & Pleck, J. H. (1995). Masculinity ideologies: A review of research instrumentation on men and masculinities. In R. Levant & W. Pollack (Eds.), A new psychology of men (pp. 129–163). New York: Basic. Thompson, E. H., & Whearty, P. M. (2004). Old men’s social participation: The importance of masculinity ideology. Journal of Men’s Studies, 13, 5–24. Turner, B. F., & Turner, C. B. (1991). Bem sex-role inventory stereotypes for men and women varying in age and race among National Register psychologists. Psychological Reports, 69, 931–944. Vaillant, G. E. (2002). Aging well. New York: Little, Brown. West, C., & Zimmerman, D. H. (1987). Doing gender. Gender and Society,1, 125–151. Zepelin, H., Sills, R. A., & Heath, M. W. (1987). Is age becoming irrelevant? An exploratory study of perceived age norms. International Journal of Aging and Human Development, 24, 241–256.